Pro * THE - COMPLETE : MSX: rammers Guide “ SH Toshiyuki Sato Paul Mapstone Isabella Muriel etdcuonny M OTHER MSX COMPUTER TITLES MSX Games Book (Lacey) MSX Exposed (Pritchard) MSX Machine Language for the Absolute Beginner (Pritchard) Ultra High-Performance MSX Programs (Sato, Muriel, Mapstone) Z80 Reference Guide (Tully) Toshiyuki Sato Paul Mapstone Isabella Muriel _—_—_ iar ae © / VE Melbourne House Publishers © 1984 Toshiyuki Sato, Paul Mapstone, Isabella Muriel. All rights reserved. This book is copyright and no part may be copied or stored by electromagnetic, electronic, photographic, mechanical or any other means whatsoever except as provided by national law. All enquiries should be addressed to the publishers: IN THE UNITED KINGDOM -—— Melbourne House (Publishers) Ltd Castle Yard House Castle Yard Richmond, TW10 6TF | CONTENTS IN THE UNITED STATES OF AMERICA -—- Melbourne House Software Inc. 347 Reed D i Ror ar2r Section 1: Introduction to MSX Basic IN AUSTRALIA — | Melbourne House (Australia) Pty Ltd 70 Park Street 1 South Melbourne, Victoria 3205 SETTING ur 3 A guide to the MSX computer keyboard and special keys <SHIFT> <CAPS LOCK><TAB> <GRAPH>> <CODE> <RETURN= <CLS> <HOME> <BS> Cursor keys <DEL> <INS> Cataloguing in Publication guing 2 COMMAND MODE 7 Sato, Toshiyuki. String and Numeric Variables Complete MSX Programmer's Guide. PRINT LET INPUT Includes index | pee a ISBN 0 86161 173 X. [1]. MSX Basic (Computer program language). 3 WRITING A PROGRAM 12 2. Microcomputers —- Programming. |. Muriel, Isabella. . RUN NEW LIST REM CLS I. Mapstone, Paul. III. Title. GOTO <STOP> CONT 001.64'24 4 MORE ON ARITHMETIC 16 A () INT f dein 76443241 : 5 THE USE OF LOOPS 19 Cy ee kD a a a ae FOR/NEXT loops and nested loops Yanar 90 89 88 B7? Bo BD 84 STEP 10 11 12 THE USE OF CONDITIONS IF/THEN/ELSE condition testing IF THEN ELSE Cites ——— _ SWAP AND OR USEFUL COMMANDS AND HINTS FOR WRITING PROGRAMS AUTO LIST DELETE RENUM CLEAR FRE TRON TROFF END STOP CONT RUN THE FUNCTION KEYS KEY KEY LIST KEY ON KEY OFF MORE ON PRINT AND THE SCREEN PRINT TAB LOCATE SCREEN WIDTH SPC SPACE$ INTERACTIVE PROGRAMMING INPUT INKEY$ LINE INPUT INPUT$ SAVING YOUR PROGRAM ON TAPE CSAVE CLOAD MOTORON = MOTOR OFF READING DATA INTO ARRAYS DIM READ DATA RESTORE 22 25 30 32 37 40 43 13 14 15 16 17 18 19 20 DATA HANDLING AND SORTING Bubble sorting and Multi-dimensional arrays SWAP ERASE DIM MANIPULATING STRINGS INSTR RIGHT$ —_LEFT$ MID FUNCTIONS INT FIX ABS SGN VAL STR$ LEN RND DEFINING YOUR OWN FUNCTIONS DEF FN STRUCTURING YOUR PROGRAMS How to use subroutines GOSUB RETURN PROGRAM BRANCHING ON GOSUB ON GOTO MATHEMATICAL FUNCTIONS Trigonometry and Exponentials SIN COS TAN ATN SQR EXP LOG TaN THE ASCII CODES CHR$ ASC STRING$ TIME 49 55 59 65 67 69 71 77 21 22 23 24 25 26 27 28 THE SCREEN MODES SCREEN COLOUR COLOR PLOTTING POINTS Explanation of Coordinates system PSET PRESET = POINT DRAWING LINES AND BOXES LINE DRAWING CIRCLES AND ELLIPSES CIRCLE THE GRAPHICS MACRO LANGUAGE How to draw on the graphics screen DRAW ; PAINTING Painting a part of screen in modes 2 and 3 and how to get rid of the smudge effect PAINT THE MUSIC MACRO LANGUAGE How to play a tune on your computer PLAY BEEP 80 84 87 91 95 99 105 109 Section 2: Advanced Programming Guide ADVANCED PROGRAM EDITING How to edit in sections List of CTRL keys and special function keys LIST AUTO DELETE RENUM CONSTANTS AND VARIABLES Integer, Single precision numbers, Double precision numbers and variables Type declaration Memory requirement of variables DEFSTR = DEFINT DEFSNG DEFDBL CLEAR DIM TYPE CONVERSION CINT CSNG CDBL VAL STR$ EXPRESSIONS AND OPERATORS Arithmetical and relational operators and expressions MOD ¥ (DIV) 119 124 129 133 SURVEY OF NUMBER SYSTEMS USED IN MSX 136 Binary, Octal, Hexadecimal, and Binary Coded Decimal systems &B BINS &O OCT$ &H HEX$ 6 BOOLEAN ALGEBRA (LOGICAL OPERATORS) 145 10 Gives explanation and truth tables for ail MSX iogical operators Some useful logical relations and De Morgan's Law AND OR NOT XOR EQV IMP BOOLEAN II: THE IF/THEN/ELSE Detailed explanation of coridition testing Minimisation using De Morgan's Law Nested IF/THEN/ELSE AND OR NOT PRINT USING PRINT USING LPRINT USING PRINT# USING EVENT HANDLING AND INTERRUPT BY BASIC ON INTERVAL GOSUB INTERVAL ON/OFF/STOP ON KEY GOSUB KEY () ON/OFF/STOP ON STOP GOSUB STOP ON/OFF/STOP ON STRIG GOSUB STRIG ON/OFF/STOP ERROR HANDLING Error Messages Table Error handling routines How to create your own customised errors ERROR ERL ERR RESUME ON ERROR GOTO 159 163 171 11 12 13 14 15 CASSETTE SAVING AND LOADING The cassette BAUD rate Saving and loading in ASCII format How to merge two programs Saving and loading a section of the computer's memory SAVE LOAD MERGE BSAVE BLOAD ADVANCED GRAPHICS | CHARACTERISTICS OF EACH SCREEN MODE Detailed description of each screen mode ADVANCED GRAPHICS II COLOUR IN HIGH RESOLUTION GRAPHICS MODE 2 Detailed explanation of how colour is managed in MODE 2 ADVANCED GRAPHICS Ill HOW TO PRINT TO GRAPHICS SCREEN USING FILES OPEN PRINT# PRINT#USING CLOSE ADVANCED GRAPHICS IV SPRITE GRAPHICS Sprite size How to define Sprites How to put a sprite onto the screen Sprite screen How to move Sprites Sprite colour How to hide Sprites The ‘Fifth’ Sprite rule How to animate your Sprites How to detect Sprite collisions SCREEN SPRITE$ STRINGS CHR$ ON SPRITE GOSUB PUT SPRITE SPRITE ON/OFF/STOP 181 186 192 196 200 16 17 18 19 20 21 ADVANCED GRAPHICS V HOW TO ACCESS THE VIDEO DISPLAY PROCESSOR How to access the TMS 9929A VDP Description of VDP register VDP ADVANCED GRAPHICS VI THE VIDEO RAM BASE VPEEK = VPOKE ADVANCED SOUND EFFECTS USING PSG How to write to the AY-3-8912 PSG to create sound effects. SOUND : HOW TO USE FILES MAXFILES OPEN CLOSE EOF PRINT# PRINT #USING INPUT# INPUT$(#) LINE INPUT# MEMORY MAP CLEAR FRE VARPTR PEEK USR FUNCTION AND MACHINE CODE How to define the USR function How to execute a machine code routine How to pass a pararmeter to a machine code routine from BASIC DEF USR 214 219 221 227 232 235 22 23 MSX MEMORY MANAGEMENT AND CARTRIDGE SLOT MECHANISM 238 Basic memory configuration Cartridge Basic slot arrangement and expanded slot configuration Slot selector mechanism How to select and enable a slot Expansion slots description Expansion slot buffer RAM search procedure Cartridge ROM software: ROM search procedure Description of system variables concerning the slot mechanism CALL PERIPHERAL DEVICES 249 Cassette Printer Joystick Games paddle Touch pad LLIST LPRINT LPRINT USING PAD PDL SCREEN Section 3: Reference Section 1 BASIC KEYWORDS section 4: The Operating System 1 RSTINSTRUCTIONS 2 ENTRY POINTS ASSOCIATED WITH SLOT HANDLING 3 BIOS ENTRY POINTS USED TO ACCESS THE CONSOLE 4 BIOS ENTRY POINTS WHICH CONTROL THE JOYSTICK PORTS 5 BIOS CALLS ASSOCIATED WITH THE CASSETTE INTERFACE 6 BIOS ENTRY POINTS DEALING WITH SOUND 7 BIOS ENTRY POINTS ASSOCIATED WITH THE VDP 8 THE USE OF HOOKS 9 THE SYSTEM RAM INDEX 255 495 498 507 519 525 559 SECTION 1 INTRODUCTION TO MSX BASIC CHAPTER 1 SETTING UP When you unpack your MSX computer you should find a video connect- ing lead packed with it. This lead has aPHONO connector on either end and is used to connect the computer to a television. For the moment, you only need to connect the computer to a televi- sion. Any television will do. A colour one is nice as you can then use the MSX colour command to full advantage, but a B/W set is perfectly adequate. To connect the TV to the computer, insert one of the PHONO plugs into the socket marked “TV” at the back of the computer; the other end of the lead plugs into the aerial connection of the TV. If your TV has two aerial sockets, marked UHF and VHF, use the UHF one (UK). Switch on both TV and computer at the mains. First turn on your TV at the set and allow it to warm up, then turn on your computer. If the TV is correctly tuned you will first see this screen .. . MSX system version 1.0 Copyright 1983 by Microsoft ...and, after a pause, this screen: MSX BASIC version 1.0 Copyright 1983 by Microsoft 28815 Bytes free OK B color auto goto list run bea. | If you do not see this second screen, adjust your UHF tuner until you do. (The number of bytes free will depend on how much memory is installed in your machine.) This happens every time the MSX is turned n. lf your TV has a continuously variable tuning control, adjust this until you get a clear picture. If you have a push button for each station, it is a good idea to tune in on an unused one. Now you are ready to use the keyboard. Experiment as much as you like, you will not damage the computer! Everything you type will apear on the TV screen. You will probably get several Error messages from the computer as well, but do not worry, you will learn how to avoid these later. If you look at the screen, on first switching on, you will see a small white square directly underneath the “OK” message. This is called the Text Cursor, and shows you where the next character to be printed on the screen will appear. Try typing some letters in. It's best to get into the habit of using both hands when typing. Pressing any of the Character Keys will result in the character appearing on the screen. Type: The quick brown fox jumps over the lazy dog. To get spaces, hit the long bar at the bottom of the Keyboard. To get upper case characters you press the <SHIFT> Key and the Character Key you want simultaneously. On the LHS of the keyboard there is a key marked <CAPS LOCK>. If you press this key, everything you type afterwards will automatically be printed in upper case, and the green light next to it will light up. To get out of this, press the key again. Now look at the numbers. See what happens if you press <SHIFT > and each number in turn from 1 to 9. At the very top of the keyboard there are five elongated Keys with F1 to F10 written on them. These are the Function Keys. The display at the bottom of the screen is due to these keys. Forget about them for the moment, as we will come to their use in CHAPTER 8. Now find the <STOP> Key. You will use this quite a lot with the Control Key, marked <CTRL>. The <ESC> key does nothing at the moment, but may be used for controlling a printer. The <TAB> key moves the cursor across the screen. It deletes anything in its way. If the cursor was originally on the LHS, it is moved to the 9th character position. On pressing <TAB> again it is moved to the 17th position, then the 25th. The <GRAPH> Key enables you to print the Graphics Characters, which are accessed via the Character Keys. On pressing <GRAPH> with any Character Key, you will get one of the Graphics Characters. You will get a different Graphics Character if you now press <SHIFT> <GRAPH*> and one of the Character Keys. Find the Graphics Char- acters %Y, @, banda. On pressing <CODE> and a Character Key, you can get the Euro- pean and Greek characters, e.g. a, B, 4, / etc. Experiment with this key. Try pressing <SHIFT> as well. Most of the keys will now give you a different CODE Character. A very important key is the <RETURN> Key. Whenever you want the computer to read something you have written on the screen, you must press <RETURN>. If you press it now, you will probably get an Error message. Do not worry, it just means that the computer has not under- stood what you have typed in. By now the screen must be rather full. Notice that when you get to the end of the screen, all the text automatically moves up one line and you lose your top line. This is called “scrolling”. To clear the screen, press <SHIFT> and <HOME> simultaneously. <HOME> is one of the four editing Keys on the top RHS of the keyboard. Notice that after doing this, the cursor automatically jumps back to the top LHS of the screen. If ever the screen gets very messy, you Can always clear it in this way. Pressing this key without <SHIFT>, causes the Cursor to Home back to the top LHS of the screen without deleting the text. Now type in a whole line of characters. To delete these you can press the Back Space Key, <BS>. This will delete anything on the LHS of the cursor. Like any of the other keys, it repeats itself if you keep your finger on it, so by now you have probably deleted the complete line! Type in alot more characters and experiment with the Cursor Keys. which are marked a ea and move the cursor about in the marked direction without deleting what is underneath. Try to delete a particular character, using <BS> and the Cursor Keys. When you think you have got the idea, see what happens when you press <DEL =, which is another of the four Editing Keys on the RHS. This key DELetes the character underneath the cursor and moves everything to the right of the Cursor back one space. If you keep your finger ori ~DEL.» you will delete everything which was originally on the RHS of the Cursor Supposing you had the following typed in: abctg You might wish to insert “d" between ‘c” and “f’. To do this, move the Cursor until it is on top of the letter “f” and then press the INSert Key, <INS>. The Cursor will be replaced by a white line. If you now type “d", it will appear to the left of the line and be inserted between the ‘c" and “f" To get the Cursor back. press <iINS*- again. Practice using the Cursor and Editing Keys. The fourth Editing Key, <SELECT> does nothing at the moment. EXERCISES 1 Type out both Graphics Characters associated with each key, and the character itself. Now delete the alphanumeric characters. leav- ing just the Graphics Characters on the screen. 2 Print the number 0 then the letter O:; notice the difference between the two. You must use 0 when typing a number, otherwise the computer will report an Error. Also compare 1 with the lower case letter |. Once again, do not confuse the two. CHAPTER 2 COMMAND MODE Once you have become familiar with the keyboard and the use of <BS*> and the EDIT keys, you are ready to give the computer commands. First of all clear the screen, by pressing the <SHIFT> and <CLS> keys simultaneously. Do this whenever the screen gets rather full or messy. Now carefully type tn: PRINT “Welcome to the MSX” then press the <RETURN> key. You should see “Welcome to the MSX’ displayed directly underneath the line you have just typed in, with OK under this. Perhaps you have an error message instead. Do not worry, have another go, as you have probably made a typing mistake. On pressing <RETURN>, the computer carries out or executes a commana. It only recognises certain words, called Keywords, of which PRINT is one. On seeing PRINT, followed by some words in quotation marks, it Knew that it had to PRINT out everything included within the quotation marks. The computer does not mind if you use upper or lower case letters and ignores spaces. (Do not, however, put spaces between the letters ina KEYWORD, as this will not be accepted by the computer.) print “Welcome to the MSX”’ <RETURN> would produce the same result. Everything included within quotation marks will be printed exactly as you type it. The line: “Welcome to the MSX” is called a ‘string’. When the computer sees a PRINT statement fol- lowed by a string, it strips the quotation marks off the string and PRINTs the rest of the string exactly as it stands. More on strings later! 7 To PRINT anumber you do not need the quotation marks. Type in: PRINT 12345 <RETURN> You will see 12345 displayed directly below the last previous line on the screen. Here again it does not matter how many spaces you have between the PRINT statement and the number. Spaces within the num- ber to be PRINTed will be ignored. PRINT12 345 <RETURN> gives the same result. Now if you type in: PRINT 3+4 <RETURN> the computer will PRINT out the value 7. It has evaluated the expression 3+4 and given you the answer. If you wish to PRINT out 3+4 as it stands, you must put it in a string by using the quotation marks: PRINT “3+4” <RETURN> The computer now sees 3+ 4 as a string, and so PRINTs it exactly as you wrote it. To subtract anumber, use the “—"' sign. To multiply two numbers, the ‘*” sign is used, instead of the more usual multiplication sign, ‘x’. This is done to prevent confusion with the letter x". To divide, use the “/” sign, instead of “+”. Type in the following in turn. N.B. Don’t forget to press <RETURN> after each line. From now on this will be assumed. PRINT 473 PRINT “3+4=";3+4 In the last example you will see: 3+4= 7 When the computer sees a semicolon it realises that there is another item to be PRINTed, in this case the expression 3+4. It therefore evaluates this and PRINTs the answer straight after the string. You will notice that there is always a blank space PRINTed before and after a positive number. PRINT “aa” ; 3:4;-5) “bb’"'cc” gives: aa 3 4 -—5 bbcc As you can see no spaces are left when PRINTing strings. When PRINTing a negative number, the negative sign is put in to the preced- ing blank space. When no punctuation is used between strings, the computer assumes a semicolon was intended. You must put punctua- 8 tion between numbers, otherwise the computer will see this as one long number. The computer also recognises the comma in a PRINT statement, in which case items are PRINTed on the same line, the second item being PRINTed 15 character positions along from the first ttem PRINTed. Type in the next two lines: LET A=3 PRINTA The number 3 is displayed on the screen. The LET command assigns a value to the variable, which in this case is called A. When the computer receives the PRINT command it is expecting a number, as there are no quotation marks. It therefore checks its memory and finds that a number has been assigned to the variable A and so PRINTs this number. You can have more complicated expressions in a LET command: LET B=31*72*4 PRINT B The number 8928 is displayed. Now type: LET B=B+1000 You get: 9928 The above LET statement looks incorrect mathematically, but is per- fectly acceptable in computing. The expression on the RHS of an “=” sign is evaluated and put into the variable on the LHS. In this case, the variable B is now equal to its old value, plus 1000. Variable names must begin with a letter. After that, numbers may be used. Blanks are ignored in variable names so: A Ais equivalent to AA Upper or lower case letters can be used, but the computer will not distinguish between the two. ABBACUS is equivalent to abbacus The computer only reads the first two characters of a variable name, so: ab is equivalent to abbacus It is important not to include a KEYWORD in the variable name, so: MONTH is invalid as ON is aKEYWORD. Some words have been reserved, and may be used as KEYWORDs in a later version of the MSX. These words must not be used within variable names, e.g. NAME This is invalid, as “NAME” is a reserved word. Here are some permitted variable names: number 2 n2 NUM These names would not be allowed: an The first character must be a letter. num? “2” ig an invalid character in a variable name. Strings can also be assigned to variables: LET A$="Hello” PRINT AS Hello is displayed, as PRINT A$ is equivalent to typing PRINT “Hello” in this example. String Variable names must end with the °$" sign to distinguish them from Numeric Variables. Otherwise they must conform to exactly the same specifications as Numeric Variable names. Here are some valid String Variable names: YEAR$ INCOME$ LAS These are invalid: A No $ sign A.B$ “is an invalid character 1FIRSTS The first character must be a letter. ON$ ON is aKEYWORD TONES ON is aKEYWORD NAMES NAME is a reserved word. Another useful command is INPUT. Type: INPUT A When the computer sees this it displays a question mark and waits for you to type In anumber. This number is then assigned to the variable A Type in a number. You will see it appear after the question mark. If you now type: 10 PRINT A the number you have just typed in will be PRINTed. If you type ina string, you will get an error EXERCISE 1. Find the value of the A, where. A=49°38.07/13 CHAPTER 3 WRITING A PROGRAM So far you have been working in Command Mode. You have typed a command on screen, pressed <RETURN> and the computer has executed the command straight away. If you wanted to repeat the command, you have had to type it in again, which is rather time consuming. . lf you want the computer to remember a command, you must give ita line number, for example: 10 PRINT “THIS IS A TEST” <RETURN> On pressing <RETURN> the computer sees the line number, in this case 10, soit realises that this line is part of a program. It therefore stores the complete line in memory. When it has done this, it displays the “Ok” prompt on the screen. To get the computer to execute the line, the command RUN must be used. Type: RUN. <RETURN> - THIS IS A TEST is now displayed. To write a program, each command must be preceded by anumber. This can be any value up to 65529, but it is a good idea to start with 10 and increase this in steps of 10, i.e. have line numbers 10,20,30,40, etc. This way, when you have finished typing your program, there is room for you to insert new lines between these lines, as you can use line numbers 15, 25, 36 etc. as necessary. The computer executes a program, line by line, starting from the line with the lowest line number. Before you type in the next program, enter the following: NEW <RETURN> This command erases any old programs and variables in the com- puter’s memory. It is advisable to use this command before typing in a new program. Now type in the following program. Do not forget, the <RETURN> key must be pressed after typing each fine: 10 PRINT “WHAT {S YOUR NAME?" <RETURN> 20 INPUT NS <RETURN> 30 PRINT “HELLO ”;N$" WELCOME TO MSX” <RETURN> RUN this program. If you have not made any typing errors you will see a display as shown below: WHAT IS YOUR NAME? 5 The INPUT statement, in line 20, is waiting for a string variable to be entered. Type in yourname and press <RETURN>. The computer now PRINTs: HELLO name WELCOME TO MSX After you have typed in your name, the computer proceeds to line 30 and gives you the Welcome message. When the computer runs out of program lines to execute, it realises that it has reached the end of the program and finishes off by giving you the “Ok” prompt. You are now back in Command Mode. If you got an error message instead of the Welcome message, you have made a typing mistake in one of your lines. Before looking for errors, it’s a good idea to get a new listing of your program. To do this type: LIST This command will LIST your program out neatly, starting from the lowest line number. All Keywords and Variable Names will be LISTed in upper case, even if you typed them in using lower case. You should see: 10 PRINT “WHAT !S YOUR NAME?” 20 INPUT NS 30 PRINT “HELLO ”;N$;" WELCOME TO MSX” Now check the line referred to in the error message for typing mistakes. Once you have found the error, edit this line using the Editing and Cursor Keys as described in CHAPTER 1. Then, with the cursor still at the corrected line, press <RETURN>. This tells the computer to replace the old line with the corrected version. It's a good idea to get the cursor completely out of the program listing before typing RUN again. To do this, keep the < | > key depressed until the cursor is at the bottom of the screen. Now type RUN again, followed 13 by <RETURN->. Repeat the editing process until your program works! ‘ the screen gets very messy. press <SHIFT><CLS> before typing LIST. If you have made lots of mistakes in typing a line, instead of editing with the cursor key. it is sometimes quicker to retype the whole line, e.g. to edit: 10 PRONT WHst os yourname”™ it might be quicker to retype the whole line: 10 PRINT “Hello, what is yourname" The old line 10 is then deleted and replaced by the new one. if you wish to remove a line from your program, all you have to do is type in the line number and press <RETURN >. After you have done this you cannot get the line back. so be careful! 10 < RETURN> deletes line 10. Type in the next program. 10 REM This program repeats itself 20 CLS 30 INPUT “WHAT IS YOUR NAME": N$ 40 PRINT “HELLO "; NS", WELCOME TO MSX” 50 GOTO 30 The REM statement tells the computer to ignore everything that follows in that line. The rest of the line is used as a REMinder, for you, as to what the program does. It is a good idea to include REM statements in your program, otherwise it is very easy to forget how a program works when you come back to It tater. The second tine Clears the Screen. You have already found that one of the keys does this, <SHIFT> <HOME>, but if you want to incorporate this command into a program you must type in the command CLS. The command in line 30 should be familiar by now. This example “up do is put the message in quotation marks, follow the message by the semicolon and then the Variable Name. The order is important here. Line 50 just tells the computer to go back to line 30. and continue execution from there. This is called an ‘infinite loop’, as the computer loops back to line 30, every time it reaches line 50. This will continue until you break out of the program. Now try running this program. When you get bored with INPUTting your name, press <CTRL*- <STOP >. You will get the message: BREAK at LINE 30 14 To continue execution, type in CONT. You will find that execution re- starts from where tt left off, i.e. at the INPUT statement in line 30. If you had broken into the program at any other line, |.e. one which did not contain an INPUT statement, you would find that execution starts from the next line. EXERCISE 1. Write a program which asks for your name and age, and then PRINTs these out. 15 CHAPTER 4 MORE ON ARITHMETIC So far you have met the mathematical symbols: +-—/* Another useful symbol is ‘ *”. This raises a number to the power of another number, i.e. 23 is written 2 A 3 Type in: PRINT 4*5 This gives the answer 1024. Brackets can be used in arithmetic expression. Any expression within brackets is executed first. All the symbols listed above can be combined in one expression. The expression will then be evaluated in the following order: () Expressions within ( ) are evaluated first : Exponentials . “yh Multiplication and division have the same priority +,- These are evaluated last For example, the computer evaluates the following expression in three stages: (3+4)* 7A2 First stage 7* 7A2 Second stage 7% 49 Third stage 343 16 You must include the “*” sign between brackets, i.e. PRINT (8+4)(4—2) iS WRONG! This line should be: . PRINT (3+4)*(4—2) In the following example program a temperature is asked for in fahrenheit and is then converted to centigrade: 10 REM temperature conversion 20 CLS 30 INPUT “FAHRENHEIT TEMPERATURE”; F 40 C=(F-32)*5/9 50 PRINT “EQUIVALENT CENTIGRADE TEMPERATURE IS"; C 60 GOTO 30 Notice in line 40 that the LET statement has been omitted. This state- ment is optional. As soon as the computer sees a variable next to an equals sign, it assigns the value of the expression on the RHS of the "=" to the Numeric Variable on the LHS. When you run this program, you will find that the centigrade tempera- ture is given to 14 significant figures! If you wish to round the tem- perature off to the nearest integer, the function INT may be used. INT converts real numbers, i.e. numbers which contain decimal places, to integers, i.e. whole numbers. The argument must be en- closed in brackets. The INT function rounds a real number down to the nearest integer: 3.4 becomes 3 -—27 becomes —3 To round the real number to the nearest integer and not to the next lowest, 0.5 must be added: INT(3.9) evaluates to 3 INT(3.9 + 0.5) evaluates to 4 You must let the computer know when you want it to store integers and when real numbers. To distinguish bettween the two, integer names end with the “%” sign. Replace lines 40 and 50 of the last program by: 40 C%=INT((F—32)*5/9 +0.5) 50 PRINT“EQUIVALENT CENTIGRADE TEMPERATURE |S”; C% The program will work if you use C instead of C%, but the number will still be stored to 14 decimal places, even if all the numbers after the decimal point are zeros. The zeros won't be PRINTed, but they are stored in 17 memory and so taking up valuable room. It is therefore better to use integer Variable Names whenever possibile. EXERCISE 1. Convert the temperature program so that it INPUTs the temperature In Centigrade and PRINTs it out in fahrenheit. You will need to use the equation: F=C"9/5+32 18 CHAPTER 5 THE USE OF LOOPS This program bother typing | PRINTS out any multiplication table you wish. Do not tin! 10 REM Multiplication Table 20 INPUT ‘Multiplier’; M 30 PRINT “1 times”: M: "=": 1*M 40 PRINT °2 times”: M: "=": 2°M 50 PRINT ’3times”M; "=". 3*M and soon. Th is is a rather long and repetitive program. A much more concise way of doing the same calculation is by using a FOR. . NEXT loop 10 REM Multiplication Table with loop 20 INP UT "Multiplier": M% 30 FORC%=1TO10 40 PRINTC%:; “times”: M%: “=") C%*M% 50 NEXT C% 60 GOTO 20 Notice that in and M%. The first tim passes on to| teger Numerical Variabies are used in this program, C% e the computer reaches line 30, it sets C%=1 and then ne 40. When it reaches line 50, the NEXT statement tells it lo return to line 30. Back at line 30, it increments the counter by one, i.e. C%-—-2 now. T his loop is repeated until C% =. 10, the limiting value; after this the computer ignores the NEXT statement in line 50 and carries on with line 60. YOu can increment the counter by STEPs greater than 1 by using the STEP stateme nt in the FOR. . .NEXT Line. Try the following: 10 FORN=0TO 12 STEP 2 20 PRINTN 30 NEXT You should see: 0) 2 4 6 8 10 12 The constraints on the counter name are the same as those for Numeric Variable Names. See CHAPTER 2. if you replace the limit in the above program by 13 it will have no effect on the results. The computer will still exit from the loop when the counter reaches 12. With a STEP size of two and starting from zero, the counter will never reach 13. You can use real numbers as well as integers to set the initial and limiting values of the counter and the step size. Negative STEPs can also be used: 10 FORN=107T0 2.5 STEP —2.5 20 PRINT N 30 NEXTN gives: 10 LS 5 20 Do be careful to start the counter at a higher value than the limiting value when using negative STEPs, otherwise an Error Statement will occur lt is possible to put aFOR. . .NEXT loop inside another FOR: . NEXT loop. This is then called anested loop. We will use this technique in the next example to PRINT out a pattern of *"'s: 10 REM a nested loop 20 PRINT"*" 30 FOR I=1 TO 9 40 FOR J=1 TO I > THE SO PRINT “#¥"5 >) NESTED 60 NEXT J >» LOOP 70 PRINT" BO NEXT I 20 You should see: k KKK KRaK KK tk RR KR Re KKK RK KIA DOK Kk Note the *:” at the end of the PRINT statement in line 50. This causes the next PRINT statement to PRINT on the same line, starting at the first empty character position, i.e. there is no carriage return. Line 30 puts |=1. This sets the limit for the J counter in the nested loop, so the first time round the | loop the “*” is only PRINTed once. The next time round the main loop, | is incremented to two. J can now take the values 1 and 2 so this loop is executed twice, resulting in two “*"'s being PRINTed on the same line in the loop. Different patterns can be obtained by varying the STEP size. EXERCISES 1. See what happens if you omit lines 20 and 70. 2. Try the previous program with STEP 4 in the main loop. Change the upper limit of this loop. With a different step size, lines 20 and 70 will need changing to keep a regular pattern. 3. Vary the STEP size in the nested loop. 21 CHAPTER 6 THE USE OF CONDITIONS Sametimes you will find that you want your program io take a certain path depending on whether a certain condition holds. The IF. . .THEN format is ised for this. Here is an example of its use: 20) IF A=B THEN GOTO 100 30 PRINTA.B Line 20 tells the computer that IF the contents of variable A are equal to the contents of variable B, THEN GOTO line 100, otherwise continue with the next line of the program, which in this case PRINTs the variables Aand B. Various expressions can follow the IF statement using the signs: : GREATER THAN LESS THAN = EQUAL TO veo GREATER THAN OR EQUAL TO LESS THAN OR EQUAL TO NOT EQUAL TO Any command can follow the THEN. For example: 20) IF A=BTHEN PRINTA The foilowing program INPUTs two numbers. then PRIN1s them, the larger first. 10 INPUT Numbers’: A.B 20 IF A’B THEN GOTO 40 30 SWAPA.B 40 PRINT “The larger number is’:A 50 PRINT “The smaller number is ';B 22 In this program both variables are INPUT using the one INPUT state- ment. The computer displays “Numbers?” and then waits until two values have been entered. You can either type in one of the numbers, a comma, the second number and then press <RETURN>, or you can type in the first number, press <RETURN>, and then the second number and <RETURN~>. If you chose the second option, the computer will dispiay two question marks after you press the first <RETURN>. The program then continues. In general, any number of variables can be INPUT with the one INPUT statement. as long as they are separated from the message in quotation marks by a semicolon, and from each other by commas. The SWAP Command, in tine 30, SWAPs the contents of variable A with the contents of variable B. So if the condition A>B in line 20 is not true, then, instead of jumping to line 40, the program continues with line 30. The SWAP Command ensures that the larger number is put into variable A. The next two lines PRINT the numbers out. SWAP can also be used to exchange the contents of two String Variables. The variables must be separated by cor: nas. The IF... THEN forsat can be extended to include the word ELSE. Everything after the word ELSE is executed IF the condition before THEN is not satisfied The previous program could have been written. 10 INPUT “Numbers”; A.B 20 IF AB THEN PRINT “The larger number is’;A ELSE PRIN| “The larger number is":B 30 IF A>BTHEN PRINT “The smaller number is”; B ELSE PRINT “The smailer number is’:A Multiple conditions can be used after the IF statement using AND and OR. For example: 10 INPUT A,B,C 20 IFA=BANDA>C THEN PRINT A The last short program will only PRINT A if it is equal to B as well as being greater than C. Compare the above to the following: 10 INPUT A,B,C 20 IFA=BORA®>C THEN PRINTA This version of the program will PRINT A if it is either equal to B, or greater than C. The inequaiities can also be used with strings. A string is stored as a number in the computer's memory, each character being represented by a different number. To compare two strings, the computer really compares the two numbers representing the first character of each string. If the first two characters are the same, then it compares the next two, and so on. Lower and upper case characters are represented by different numbers: 23 A-Z by the numbers 65 to 90 a-Z by the numbers 97 to 122 Other characters such as “*” can also be compared. More on this later! (See CHAPTER 20) The following strings are sorted into alphabetical order, with the “largest” first: To summarise what you have learnt in this chapter, here 's a short program which makes full use of the IF, THEN, and SWAP statements. This program sorts out three numbers given and displays them in order, the largest first: 10 REM SORT PROGRAM 20 INPUT “Three numbers”: A,B,C 30 IF B>A THEN SWAP A,B 40 iF C>B THEN SWAP B,C 50 JF A>BTHEN SWAP B.A 60 PRINTA,B,C Have a go at this program by INPUTting various numbers. Can you work out the iogic of this program? EXERCISES 1. Using the above program, write a program which sorts three strings into alphabetical order. 2. The example program can only cope with three numbers. Can you write a better one which can handie more than three numbers? One is given in CHAPTER 13. 24 CHAPTER 7 USEFUL COMMANDS AND HINTS FOR WRITING PROGRAMS A useful time saving command is AUTO. On typing this in, you will find that line number 10 AUTOmatically appears on the screen. Type REM after this and press <RETURN>>. Now line number 20 will appear. This process will go on indefinitely. To STOP the AUTOmatic Line Number- ing, press <CTRL><STOP>. Before typing in a program it is advisable to use the Command NEW, which deletes any old program lines and resets the variables. If you do not do this, you may well find that some of your old program lines are mixed in with the new program, especially if the last program was longer than the present one. This can result in chaos! As you are already aware, the LIST statement is used to LIST out your program on to the screen. If you just wish to LIST one line of your program (you might have an Error Message referring to this line). type in LIST followed by the Line Number: LIST 30 This just results in line 30 being displayed on the screen. On LISTing your program, you will probably have noticed that all the lower case characters in the keywords and variable names are replaced by upper case characters. It is always a good idea to title your program using a REM statement, and to sprinkle these liberally throughout the program to explain each part. This makes life a lot easier when you come to edit your program, especially if you are doing this some time after you originally wrote it. The REM statements can always be removed at a later date if you find 25 you are running out of memory. As you might well want to remove a REM statement later on, NEVER have a GOTO command referring to a REM statement. The simplest way to delete a single line from a program is simply to type the line number in and then <RETURN>. If you wish to delete a large chunk of a program, however, it is quicker to use the DELETE command: DELETE 20-80 This will delete all the fines with line numbers between 20 and 80 inclusive. Be careful not to delete the wrong lines by accident, as once you have used this commana, the only way to get the lines back is by typing them in again! More than one command can be included in one line but the com- mands must be separated by the colon. For example. 20 PRINT’a’:GOTO 10 This is a valid line. The letter “a” will be printed and then the cornputer will jump to line 10. Any number of commands can be linked together in this way. This will save on computer memory as a certain amount of memory Is necessary to save a line number. A program is generally more readable, however. if you avoid the use of multipie statements. There 1s also a limit on the number of characters which can be included in one line — up to 255 characters may be typed in, including spaces. The computer ignores anything over this. To demonstrate this, type in the following: 10 PR NTR PAA eee : tilling the screen with “*”’s. Now press <SHIFT> <CLS> and RUN the program. only 247 °*"s will be PRINTed. The computer has only ex- ecuted the line up to the 255th character. Notice that if you miss out the final quotation marks of a string, the computer automatically assumes their presence. However, do not get into the habit of doing this as it can produce confusing results. One more point about PRINT —- it has an abbreviated form, the question mark. Type in: 10? Now type LIST. The question mark will have been replaced by PRINT. On inspecting your listed program, you might find that you have missed a line out, or even a number of tines. by accident when typing your program in. It is possibile that there is not enough room to insert the missing lines due to inadequate spacing of the line numbers. If this happens. you can RENUMber your program using the RENUM Com- mand. For example: 26 10 REM 11. REM 12 REM To insert a line between lines 10 and 11, type RENUM. Now LIST your program, you should see: 10 REM 20 REM 30 REM You can now easily insert new lines between lines 10 and 20, using line numbers 11 to 19. It is possible to store more than one program in memory. To do this you could store the first program between line numbers 10 and 100. for example, and the second between 500 and 1000. When you type in RUN, you must prevent the compu and then going straight on to the statement. This must be the very ter from executing the first program second. To do this, use the END ast statement in the first program. When the computer encounters END it stops execution and puts you back into Command Mode. To continue execution, which in this case means RUNning the second program stored in memory, you can type CONT followed by «<. RETURN>-. Alternatively, to RUN the second program before the first, you Can go Straight to it using: RUN 506 This stants execution at line 500. The same result is obtained using: GOTO 500 The STOP Command can also be used in a program. This command, like the END Command, puts you back into Command Mode. However, it produces the message “BREAK in LiNE ..”, giving the line number of the STOP statement. After a STOP statement, as you are in Command Mode, you can continue execution by typing in CONT You can STOP arunning program at any point by pressing < STOP> . To continue execution press this key again. You will not be able to use the command CONT in this case, as this can only be used with STOP statements within programs. If you press <CTRL>> <STOP =, however, this STOPs the program and puts you back into Command Mode. Once in Command Mode you can restart execution from the next line by using CONT. Type in the following short program: 27 10 REM 20 PRINT “Goodbye” 30 GOTO 20 RUN this program. Now Press < STOP> . This will stop the program. The only way to restart execution is to press the< STOP> key for a second time. RUN the program again, this time STOP it using <CTRL= <STOP>. Restart the program using CONT, 10 REM 20) PRINT “HELLO” 30 STOP 40 PRINT “GOODBYE” 50 END 60 GOTO 20 Restart the program, using CONT after both the END and STOP statements. A rather different command, which you will not need to use at the moment but which you should be aware of, is CLEAR. This tells the computer to forget all the variables currently stored in memory. It can also be used to make more room for a variable. On turning the computer on, strings can only be 200 characters long. You canincrease this to 255 characters by typing: CLEAR 255 Another command which deals with the computer's memory is FRE. This tells you how much memory you have FREe in either the BASIC program area, or in the string storage area: PRINT FRE(O) Gives the memory left in tne BASIC programming area. PRINT FRE(*’) Gives the memory left in the string area. Finally, two. commands which are designed to help you TRace errors in your program. TRON, which stands for TRace ON, makes the com- puter display the line number of each line as it executes it. Type in: 10 REM USE OF TRON 20 PRINT “Hello, are you still awake?” 30 INPUT “yes or no”: A$ 40 IF A$="no” THEN BEEP : GOTO 20 50 PRINT “GOOD Now type TRON and RUN the program. You should get the following display: [10}[20] Hello, are you still awake? [30] yes or no? 28 Here the program waits for an INPUT {f you INPUT yes, you get: [40][50]GOOD If you INPUT no, you get: [40][20] Hello are you still awake? [30] yes or no? The BEEP Commana, in line 40, is the easiest command to use to get a noise out of the computer. To stop the tracing process, use TROFF. CHAPTER 8 THE FUNCTION KEYS These are the five keys at the top of the keyboard. Each key has two functions, one of which is displayed on pressing the key directly, the other on pressing <SHIFT> and the key SIMULTANEOUSLY, i.e. the functions F1, F2, F3, F4 and F5 are used by pressing the relevani key, while F6, F7, F8, F9 and F10 by pressing <SHIFT= and the correspond- ing key at fhe same time. On turning on the computer, the keys are already programmed for certain functions. A tist of these functions may be displayed by typing the command: KEY LIST The contents of the first five keys are displayed on the last line of the screen. The other five are displayed on pressing <SHIFT>>. To get rid of this display type KEY OFF, then, to bring it back, KEY ON. The last three commands can be inciuded in a program if so desired. For the moment, forget KEYs F1, F6 and F7. There will be more said about these KEYs later The functions of the other seven keys shouid be familiar to you by now, and you will probably find them useful KEY DISPLAY _ DESCRIPTION Fq color See CHAPTER 22 F2 auto This ts programmed to display AUTO on the screen, followed by a space, I.€. pressing this key 's equivalent to typing “AUTO’ F3 goto Pressing this key is equivalent to typing in GOTO followed by a space F4 list This is equivalent to typing LIST followed by a space. 30 KEY DISPLAY DESCRIPTION F5 run The function of this key Is slightly different from the ones you have already met. It is equivalent to typing RUN followed by <RETURN =. so this key will RUN a program, if one exists in memory, as soon as it is pressed. FO = color 15.4.7 See CHAPTER 22 F7 cload” See CHAPTER 11 F8 cont This is equivalent to CONT followed by <RETURN->. if you have stopped a program running by pressing CTRL «STOP =, or by an END or STOP statement in a program, you can restart execution. from the next line, by pressing <SHIFT = and F8 FQ list. FQ wili straight away LIST the program from the last line number typed in, and move the cursor to the beginning cf that line. FIQ run This function key Is useful as it CLeara the screen and then RUNs any program in memory lt is possible to redefine the action of any function key yourself iy using the command, KEY. followed by the key number, a comma, [nen & stang containing the new function. Type in: KEY 1, (PRINT ~ RETURN » This will result in the statement PRINT being displayed on the screen whenever KEY 1 is depressed. Notice that the list at the bottom of the screen has chanmed COL Cia nas been replaced by PRINT KEY 6, "NEW" + CHR$(73) programs KEY 6 to carry out the command NEW. CHB lt) Is equivalent to pressing ~ RETURN © (see Chapter 20 for areater detail) When defining a KEY. you may only use amaximumm of 15 characters in the string. Note that CHR$(13) only takes up one character posite EXERCISE 1. Program KEY 7 to RENUMber your prograrn immediately the KY vs pressed 31 CHAPTER 9 MORE ON PRINT AND THE SCREEN So far you have used the PRINT statement to PRINT strings and variables, with either a comma or semicolon separating the list of items to be PRINTed. The semicolon has resuited in the items being PRINTed next to each other; the comma has resulted in two ttems being PRINTed on the same line, the first at the beginning of the line, the second at the 15th character space. A semicolon at the end of a PRINT statement suppresses the carriage return, so the next PRINT statement will start PRINTing on the same line: 10 PRINT “To be, or not to be—”; 20 PRINT “that is the question;”’ This ts all PRINTed on one line: To be. or not to be —— that is the question; You will probably find that at some stage you want to print a table of results, or perhaps a list of aptions neatly tabulated. Two commands to help you to do this are TAB and LOCATE. To understand how to use these commands you should be aware of the different Screen Modes. There are two text screens, SCREEN 0 and SCREEN 1. On switching on the computer, you are automatically in one of these SCREENs. For this Chapter it will be assumed that the default screen is SCREEN 1. To make sure you are in this screen, type: SCREEN 1 This screen is 29 character positions wide by 24 long. Each character position occupies a square which is made up of 8x8 points. 32 The other text screen, SCREEN 0, Is 39 character positions wide by 24 long. The character positions in this screen are slightly smaller, each occupying a rectangle of 6x8 points. In this screen the border is the same colour as the background, so the whole screen is blue on initially switching on. Tu access this screen type: SCREEN 0 Now return to screen 1, by typing: SCREEN 1 The origin of both the TEXT SCREENS is taken to be at the top left hand corner of the screen. Notice that the first row and column on each screen is called Row 0, and Column 0, respectively. So, the last column on Screen 0 is Column 39 and on Screen 1 is Column 28. The last row on both screens is Row 23. You cannot PRINT on the last row unless you first use KEY OFF to get rid of the Function Key list. SCREEN 0 SCREEN 1 39 0 28 ) 0 a 0 | 23 23 You can change the maximum number of characters on each line from the default values by using the WIDTH statement, followed by a number. This number sets the maximum WIDTH of the screen. In Screen 0, the maximum number of characters you can set across the screen is 40, in Screen 1, it is 32. You can decrease the width of both screens down to just 1 character wide if you wish! SCREEN 0 WIDTH 20 These commands set the maximum WIDTH of SCREEN 0, to 20 characters. You will find now that you can only write in the 20 most central columns. To get back to the default width and screen, type: 33 SCREEN 0 WIDTH 29 Now back to tabulation! The TAB command is always used in conjunction with PRINT. This statement causes the computer to start PRINTing at the specified column (or X co-ordinate) given the TAB statement. For example: PRINT TAB(10) ‘This starts Printing at the 11th character position.” Don't forget the first character position on the screen is at Column 0, Row 0. Alternatively you can think of it as being at the X, Y character co-ordinate position (0,0). You can use more than one TAB statement in a PRINT statement: PRINT TAB(3)4 TAB(7)8 This will PRINT the number 4 at the fourth column. The blank space which always precedes a positive number is PRINTed at the third column. The number 8 is PRINTed on the same line, the preceding blank being at the seventh column. No punctuation is needed between TAB statements, but do not forget the brackets. You can only specify the column in a TAB statement, so if you wish to PRINT at a particular row as well as column, you must use the LOCATE Command just before the PRINT Command. This command sets the Cursor position: the first number in the LOCATE Commandis the column number (or X co-ordinate), the second the row number (or Y co- ordinate). For example: 10 LOCATE 6.2 20 PRINT “NAME” 30 LOCATE 20.2 40 PRINT “SCORE” which PRINTs: NAME SCORE on tne third line of the screen You need a LOCATE Command with each PRINT statement. Alternatively, you can just use LOCATE to set the row. and then use PRINT TAB. The following program gives exactly the same result as the last one: 40- CES 20 LOCATE .2 30 PRINT TAB (6) “NAME” TAB(20) “SCORE When the X co-ordinate ts left out of aLOCATE statement. the computer assumes you wish to stay ai the present X position. The CLS Command 34 in line 10 has set X=0, so this value of X is used when locating the Cursor in line 20. Note that although you may leave out the X co-ordinate, you must not leave out the comma. LOCATE cannot be used in Command Mode, because as soon as a command is executed in this mode, the cursor is set to the LHS of the next line, ready for anew command to be typed In. You can also use LOCATE to switch the cursor off within a program, i.e. to get rid of the little white square. To do this, put a comma after the Y co-ordinate and follow this with a zero. To switch the cursor back on, replace the 0 with a 1. For example: 10 LOCATE, 0 disables the Cursor. While: 10 LOCATE 10,10,1 enables the cursor and moves it to co-ordinate position (10,10). This effect cannot be observed in Command Mode. Now for another command, SPC. This command is used within a PRINT statement to specify the number of SPaCes to be PRINTed. The short program you typed in earlier, to PRINT out the headings “NAME” and “SCORE”, could be replaced by: 10 CLS 20 LOCATE ,2 30 PRINTTAB(6)"NAME” SPC(10) “SCORE” Any integer number between 1 and 255 may follow the SPC statement; the number must be enclosed within brackets. Whenever you wish to PRINT a String containing a lot of spaces, it is always advisable to use the SPC statement, as this saves on memory. There is an alternative method to PRINT spaces, using SPACES. This is a rather more versatile command than SPC as it can also be used outside a PRINT statement, whereas SPC cannot. SPACE$ is used to form strings made up entirely of spaces. For example: A$=SPACE$(10) is equivalent to: A$=" Using this command, we can assign our heading to a variable and PRINT this out as often as we wish: 10 H$="NAME” + SPACES(10) + “SCORE” 20 CLS 30 LOCATE 2 40 PRINTTAB(S) H$ 35 Don't forget you can't do this with SPC: A=SPC(10) EXERCISE 1. Write a program which INPUTs today’s date and your date of birth, then PRINTs out your age. Tabulate the messages printed on the screen using LOCATE. Turn the cursor off before the INPUT statement. IS WRONG! 36 CHAPTER 10 INTERACTIVE PROGRAMMING A program is called “interactive” if at some point during its execution it accepts an input from the keyboard. This chapter is therefore going to deal with the INPUT statement, INKEY$ and INPUTS. Although TAB can only be used with PRINT statements, LOCATE can be used with INPUT statements as well: 10 LOCATE(10,22) 20 INPUT “Article, Price’: A$, P You will see: Article, Price? appear on Row 22 of the screen. The computer will wait until you have INPUT a string and then anumber. To do this, as you already know, you can press <RETURN> after INPUTting the String Variable and then INPUT the number, or INPUT them together separating the string and number by a comma. If you use the statement LINE INPUT, however, you could INPUT the whole line of text into one string variable: 10 LOCATE(10,22) 20 LINE INPUT “Articles and price?” L$ You will see: Articles and price? on the 22nd Row of the screen. Now if you typed in: Bananas, peanuts and tea bags £25.00 37 the whole line, including the blanks, would be put into the variable L$. If you replace the LINE INPUT with an INPUT statement, everything after the comma is ignored. LINE INPUT can only be used with String Variables, so: 10 LINE INPUTA IS WRONG! The line you type in can be any length of up to 200 characters. The length of the line can be extended up to 255 characters, by using the CLEAR statement (see CHAPTER 7). Unlike INPUT, LINE INPUT does not automatically produce the question mark prompt, so if you want this you must include it in the message after the KEYWORD. More than one line can be INPUT in one LINE INPUT statement, but then the String Variables must be separated from each other by commas and from the message, if any, by a semi- coton. At some point in your program, you might want the computer to stop and ask you which of several options you want to take. For example, ina games program you might want the computer to ask the player if he wants the instructions displayed, or at the end of the game whether he wants to play again. You can use the INKEY$ Statement to do this. This function tests the keyboard to see if a key is being pressed. If one is, the key character is entered into the string Variable INKEYS. If not, an empty string is returned into the Variable INKEY§$ and the computer carries on with the next program line. The next part of a program could be put at the end of a game: 100 CLS 110 LOCATE ,10 120 PRINT "Do you wish to play again?" 130 PRINT "Press Y for Yes" 140 PRINT “Press N for No" 150 AS=INKEYS 160 IF AS="N" OR AS="n" THEN END 170 IF AS="Y" OR AS="y" GOTO 10 180 GOTO 150 The first time line 150 is reached, INKEY$ either contains an empty string or the character of any of the character keys being pressed. If any of the keys N.n,Y or y were pressed while the computer was at line 150, then the program either ENDs or starts again from line 10. The GOTO statement in line 180 makes the computer loop back to line 150 until one of the four character keys recognised by the computer is pressed. Do not forget that INKEY$ is a string and can only be equated to a String Variable, so: A=INKEY$ IS WRONG! and will result in a Type Mismatch Error. 38 Another statement, similar to INKEY$, is INPUTS. Like INKEY§, this reads the keyboard directly, but instead of just reading one character, it reads the number of characters specified in the INPUT$ statement, 1.e.: 10 A$=INPUT$(5) 20 PRINTAS$ The computer will wait at line 10 until five keys have been pressed. Line 20 then PRINTs these characters out. The number following INPUTS should be an integer in the range of 1 to 200. If a real number is used in this range tt will be truncated to the nearest lower integer. EXERCISE 1. Use an INPUTS statement at the beginning of a program to input a password. The first few lines of the program should ask for the password, and execution of the rest of the program should only start once the correct password has been entered. 39 CHAPTER 11 SAVING YOUR PROGRAM ON TAPE You have reached the stage where you can write fairly long programs. Instead of typing them in each time you wish to RUN one, you'll find it much more convenient to store your program on tape. Firstly, connect your cassette recorder to the computer. Most cassette recorders will do; a cheap monoplayer Is okay. It is very useful to have a tape counter to note where you have stored your program. The recorder must have an input socket for use with a microphone, and an output socket for use with earphones. These sockets must be 3.5mm jack sockets to take the jack plugs. The connector lead must have an 8 pin DIN plug on one end. If the lead is not included with your MSX computer, please buy one. You must specify that it is for an MSX computer. Plug this into the computer —— in the socket marked “CASSETTE”. The other end of the lead divides into three ieads, with red, black and white jack plugs on their ends. Plug the white lead into the socket marked “EAR” on your recorder and the red lead into the socket marked “MIC”. Forget the black lead for the moment. Put an empty cassette into your cassette recorder. Rewind it if necessary and then play it until you reach the end of the transparent leader at the beginning of the tape. Reset the tape counter. Now type a short program into your computer. Then type. CSAVE''name” <RETURN> You can give your program any name you wish as long as it does not exceed six characters. The characters may be upper or lower case 40 letters of the alphabet, or numbers. The first character of the name must be a letter. Now depress both the RECORD and PLAY buttons on your cassette recorder, then press <RETURN>: on the computer. When the computer has finished recording the program, it will come up with the “Ok’ message. Stop the tape and note the tape counter reading. To check if you have saved your program successfully, switch off the computer to get rd of the program in memory. Rewind the tape to the beginning or the zero setting of the counter. Now switch the computer back on and type: CLOAD''name” <RETURN> or just: CLOAD <RETURN> If the program name is not specified, the computer will load the first program it finds on the tape. Now depress the PLAY button on the cassette recorder. You should get the following message on the screen once the computer has found the program: FOUND: name When the program has been loaded you get the “Ok” message. Once that appears, stop the tape. Notice that the function key, F7, is programmed to display CLOAD" on the screen. Using this, you just need to type in the program name and the final quotation marks. You can verify that the program in the computer is the same as the program on the tape, i.e. that no saving or loading errors have occurred, by using the command CLOAD?. To use this, rewind the tape to the beginning of the program and type: CLOAD?'name”’ <RETURN > Start the tape. If the program in the memory is the same as the program on the tape you will get the “Ok” message. If you do not get this, you have not loaded the program successfully — see the problems section at the end of this chapter. When storing several programs on one tape, It is a good idea to space them out well and note down the tape counter readings at the beginning of each program. If you record a program on top of another program, obviously you will lose the first program. Supposing you have two programs stored on tape, the first called ‘plot’, the second called “point”. If you enter: CLOAD" point” A and start the tape from the beginning you wil! get the following display: CLOAD" point” skip :plot Found:point Ok The computer tells you it has passed over the program called “plot”, and has then loaded “point”. Now for the function of the black jack plug. You can only use this if you have a remote control socket on your cassette player, marked “REMOTE”. If you have, plug in the black plug and type: MOTOR ON <RETURN> If you now do a CLOAD/CLOAD? or a CSAVE, the motor in the cassette recorder will be automatically switched off by the computer when it has finished reading from or writing to the cassette recorder. This is very useful, as it is easy to forget to turn the cassette recorder off after using CLOAD or CSAVE. To stop remote control of the cassette recorder, type: MOTOR OFF <RETURN> Just typing: MOTOR <RETURN> will have the same effect as this switches control off if it is on, and on, if it is off. Saving and Loading Errors If you've been having problems saving and loading your programs, check: 1. You have plugged the cassette recorder tn! 2. You have got the leads correctly plugged in and that they have not come loose. 3. Try varying the volume and tone control on the cassette recorder. Usually it's best to leave the tone on its maximum setting and just vary the volume control. Start at the midway setting and try increasing or decreasing the volume from this. 4. Check you are spelling the name correctly if you cannot load a program you have already saved. If you still cannot save or load your programs, you have probably got an unreliable cassette recorder. 42 CHAPTER 12 READING DATA INTO ARRAYS So far you have only INPUT a limited number of items at any one time. There is bound to come a point when you will want to record a lot of data, perhaps along list of names or numbers. Supposing you wanted to keep a telephone directory of all the people you know. You could try the following: 10 REM DIRECTORY VERSION 1 20 INPUT‘ Name and Number’:N1$,NUM1 30 INPUT“ Name and Number";N2$,NUM2 40 0... and so on, but with only ten names this is going to be a long and repetitive program. It would be much better to use a loop, executing the loop once for each entry in the directory. To do this, you will need to use an Array. An Array is a whole group of variables which all have the same name but differ from each other by a subscripted number. A(n) is a Numeric Array. If n=10, then this Array can hold eleven variables called A(O), A(1), A(2) ... A(10). You can think of A as being a street name, in which case the following number gives you the house number in the street. If you are going to use arrays with more than 11 elements, you must let the computer know you are going to do this by telling it the size of the array and its name. This is done using a DIM statement: DIM NUM(12) sets up what is called a One Dimensional Numeric Array called NUM, which has thirteen elements, NUM(0), NUM(1), ... NUM(12). The array is 43 called One Dimensional as it can be represented by a One Dimensional diagram. ) ) ) The first character of an array name must be a letter of the alphabet the rest can be letters or numbers. As with simple variables. KEYWORDs or punctuation marks must not be included in the name. Only the first two characters are recognised by the computer, and lower case letters are not distinguishable from upper case. Therefore, the following all look the same to the computer: ab(10) AB(10) Abbacus(10) If you wish to read strings into an array, you must use a String Array. The only difference in appearance between this and a Numeric Array is that the name must end with the '$" sign (as indeed a String Variable Name must). Permitted names include: —~ += OONANHWNM +O Fe Fe Ls PL LEN LE C == 2CGNSGSEGSSS ¢ — CECE EEE ES — UW Game$(20) Address($10) Ai$(S) These would not be allowed: TOWNS(12) Inciudes Command TO Question?$(6) Punctuation marks are not allowed. name$(200) NAME is a reserved word. You can dimensionalise all the arrays you are going to use, in one statement at the beginning of the program: DIM A$(12),B(13),C(20) . Do not DIMensionalise more array space than you need, as on receiv- ing a DIM statement the computer sets aside sections of memory for 44 each array, so if you then don't use all of the array you are wasting memory. You can save on memory by using integer arrays, rather than real number arrays, i.e. NUM%(12). will hold 13 integer variables, in elements NUM%(0). NUM%(1)... NUM%(12). Multi-dimensional Arrays are dealt with in the next Chapter. You can now read your names and numbers into the telephone directory using a loop: 10 REM Telephone Directory Version 2 20 REM Input entries into Arrays NS and NUM$ 30 CLS 40 INPUT “Number of entries", E SO N=E-1 60 DIM N$C(N), NUMCN) 70 FOR I=0 TO N 80 CLS 90 LOCATE ,10 100 INPUT “Name "3; NSCI)D 110 INPUT “Number"; NUMCI) 120 NEXT I The variable N is used in the DIM statement in line 40, to ensure that there are the same number of elements in the arrays NS and NUM as entries in the directory. N is one less than E, as the first element of each array is element(0) and not element(1). All the names and telephone numbers have been entered into two arrays. The next part of the program displays the directory. The program now looks like this: 10 REM Telephone Directory Version 2 20 REM Input entries into Arrays NS and NUM$ 30 CLS 40 INPUT “Number of entries", E 50 N=E-1 60 DIM N$C(N), NUMCN) 70 FOR I=O TON so CLS 90 LOCATE ,190 100 INPUT "Name "s NSCI) 110 INPUT “Number"; NUMCI) 120 NEXT I 130 REM display directory 140 CLS 150 PRINT TAB(2) "NAME" TAB(18) "NUMBER" 160 PRINTTAB(2) Ceeeer rr rrr ere re Ss ts te ee 170 FOR I=0TO N 45 180 LOCATE 2,4+1 #2 190 PRINT NS$CI) 200 LOCATE 18, d+1#2 210 PRINT NUMCI) 220 NEXT I When the program finishes, you have lost all your names and numbers! It is possible to save data at the end of aprogram. You can then store the data on tape, with the program. To do this use the READ and DATA Commands. A DATA statement can be followed by a long list of numeric or string data, the data items being separated by commas: DATA Tom, 440 2073, Henry, 224 6607, Graham, 442 1708, .... The only constraint on the number of data items stored is that the total number of characters in the line must not exceed 255. You can have as many DATA statements as you like in a program. They can be stored anywhere you wish, as the computer will skip over DATA statements until it comes to a READ statement. It is often best to keep all the DATA statements at the end of the program, however, as this way it is easier to change DATA statements without hunting through the whole program. Let us now go back to the directory program. This time, instead of INPUTting the data while the program is running, the latest version of the program already contains the data in DATA statements. We know the number of entries in the directory as we have already put them into DATA statements, so we no longer need to INPUT the number of entries. The number of entries is put into the variable E, as you might want to change the number of entries in the directory at a later date, and so need to change this number. In that case, only line 30 will have to be altered. The program now looks like this: 10 REM Telephone Directory Version 3 20 REM Read Data into an Array 30 E=4 a0 N=E-1 SO DIM N$(N),NUMCN) 60 CLS 70 REM display DATA 80 PRINTTABC2) "NAME" TABC(18) "NUMBER" 90 FOR I =O TO N 100 READ N$CI), NUMCI) 110 LOCATE 2,d+I#2 120 PRINT N$CI1) 130 LOCATE 18, d+I*2 140 PRINT NUMCI) 46 170 NEXT I 175 END 180 DATA BATMAN ,774 0303, ROBIN, 420634 190 DATA “Dennis",dd4 206d, Baloo, 40 235 When the first READ statement is encountered, the computer READs the first item of DATA from the first DATA statement into the variable in the READ statement. In this case, BATMAN is READ into the element N$(0) of the array N$. The second time a READ statement is encountered, the second item of DATA in the DATA statement is READ. Once ail the items in a DATA statement have been READ, the computer goes to the next DATA statement and READs this sequentially. You might think that it is unnecesary to READ the DATA into an array in this program. You would be right! However, once the DATA is in an array, we can then process it, if we wish, before displaying it. The next Chapter gives the final version of this program, in which the entries are sorted into alphabetical order before they are displayed. You may use more than one READ statement to READ a DATA statement. The next READ statement starts from the first uNREAD DATA item. For example: 10 READA,B 20 READ A$,B$ 30 PRINT A;B:;A$;B$ 40 DATA 10, 20, “This is a sentence. “hello results in: 10 20 This is a sentence. hello Do be careful when using DATA statements. You must make sure that the order of the DATA in the statement corresponds to the order of the variable types in the READ statement. If you are READing into a Numerical Variable, you must have a number at the corresponding position in the DATA statement. You cannot put variables in DATA statements, only strings and numbers. Strings, in DATA statements, don’t have to be enclosed in quotation marks unless they contain commas, colons or spaces. Sometimes you might wish to use the same DATA statement twice. The first time round you might just wish to PRINT the DATA, the second time to do some calculations with it. To make the computer go back to a DATA statement it has already READ, you use the RESTORE command. RESTORE on its own makes the computer re-READ the DATA from the very first DATA statement. RESTORE, followed by a line number, makes the computer re-READ from the first DATA statement at or after that line number. Make sure that you follow RESTORE by a valid line number, i.e. one that exists in the program. Run the following program: 47 10 20 PRINT “A=""A.“B=” 30 RESTORE 40 READC,D 50 PRINT “C=":C, "‘D= 60 DATA 10,20 You will see: A=10 B=20 C=10 D=20 READ A,B ‘B “D 48 CHAPTER 13 DATA HANDLING AND SORTING In the last Chapter you were given a telephone directory program; it would be nice to PRINT the directory out in alphabetical order. We've already considered a rather limited way of sorting lists into numerical order in CHAPTER 6. The following method, however, is more flexible. (Although by no means the best way, it is one of the easiest to follow.) This method is called BUBBLE SORTING. Let us consider a list of numbers to be sorted: 7,4, 5,10, 2,8 We want to sort these numbers into decreasing order. Let us compare the first two numbers. |f the number on the LHS is smaller than the number on the RHS, then we SWAP these two numbers and then compare the second and third numbers. If the first two numbers had been in the correct order already, then we would have left them as they were and compared the second and third numbers straight away. So, in our example, we first compare 7 with 4. These two are in the right order so we move on to compare 4 with 5. 4 is less than 5 so we SWAP these two round. We now have: 7,5,4, 10, 2,8 Now we compare the third number with the fourth, i.e. 4 with 10. Again we SWAP these two: 7,5, 10,4, 2,8 Comparing the fourth with the fifth, we leave things as they are. Finally, comparing the fifth number, 2, with the last, 8, we SWAP, ending up with: 49 7,5, 10, 4, 8, 2 Well, we have not finished yet! But notice that the smallest number is already at the end of the list. Let’s see what happens the second time through: COMPARE SWAP? RESULT 1st/2nd no 7,5, 10, 4, 8, 2 2nd/3rd yes 7,10, 5, 4, 8,2 3rd/4th no 7,10, 5,4, 8,2 4th/Sth yes 7,10, 5, 8, 4,2 sth/6th no 7,10, 5, 8, 4,2 order at the end of the list. The maximum number of passes it will take to get all the numbers in order in any list is therefore one less than the total number of items in the list to be sorted, i.e. in the example given, the maximum number of passes required would be five. In fact, lists are often sorted before this as some of the items are likely to be in the correct order before you start sorting. It is therefore a good idea to check whether the list is already sorted at the end of each pass through. Let us now sort the telephone directory into alphabetical order. There are three parts to this program. The first part READs the DATA into an array, the second part sorts the DATA into alphabetical order, and the third part displays the sorted DATA. 10 REM Sorted Telephone Directory 20 REM READ DATA into two Arrays 30 E=d 40 N=E-1 SO DIM N$(N), NUMCN) 60 FOR I=0 TO N 70 READ N$C(I), NUMCI) 80 NEXT I 90 REM sort into order 100 M=N-1 110 P=0 120 FOR C=0 TO M 130 IF N$CC) > NS$(C+1) THEN SWAP NSC(C) ,NSC(C+1)5 SWAP NUMC(C), NUMCC+1) 8 P=P+1 140 NEXT C 150 IF P<>0O GOTO 110 160 REM display sorted DATA 170 CLS 180 PRINTTABC2)> "NAME" TABC(18) "NUMBER" 190 PRINTTABCZ2) "463636363636 46-4636- 3 363 HE HE 50 200 FOR I=0 TO N 210 LOCATE 2,4+I#2 220 PRINT NSCI) 230 LOCATE 18,4+I1#2 240 PRINT NUMCI) 250 NEXT I 300 DATA CHRIS, 3333333, DENNIS, ddddddd 310 DATA ANDREW,1111111, BELLA, 2222222 The first and last parts of this program should be familiar to you (see CHAPTER 12). Lines 90 to 150 contain the sorting routine. At the beginning of the loop the counter, P, is set to zero. Each time round the loop, an element of the array is compared to the next one along, te. the first time round, N$(O) is compared to N$(1). If the names in these elements are not in alphabetical order, the contents of the elements are SWAPped. The corresponding telephone number is also SWAPped. P is now set to 1, to indicate that a SWAP has taken place. The last time round the loop, N$(2) is compared to N$(3). The contents of these elements are SWAPped if necessary, and the counter, P, incremented. Execution then continues with line 150. Here the computer checks the value of P. If any elements have been SWAPped, P will be greater than 0. The computer assumes that the names are not yet sorted, and goes back to line 110. P is reset to 0, and the sorting loop, is executed again. If, at line 150, P has been found to be 0, the computer knows that no elements had been SWAPped, thus indicating that the names are now in alphabetical order. The computer, therefore, continues with the next line of the program, and displays the directory. If you RUN this program, you should see: NAME NUMBER ANDREW 1444117 BELLA 2222222 CHRIS 3333333 DENNIS 4444444 lf you wish to use the same array name twice in a program but with different data, you should ERASE the old array: ERASE N$ wipes out all the names stored in the array N$. You can now use the array N$ again, and re-DiMensionalise it using a DIM statement. If by chance you have a variable called Nf, it will be totally unaffected by the ERASE statement. Try not to have variables and arrays with the same name, however, as this can be rather confusing. 51 You can have Multi-dimensional Arrays, anything up to 255 dimen- sions! For the moment just consider a two dimensional array. These are very useful when storing tables: DATE CREDIT DEBIT BALANCE 30.6.84 34.05 34.05 5.7.84 10.00 24.05 10.7.84 120.00 144.05 12.7.84 50.00 94.05 20.7.84 200.00 294.05 These four columns of numbers can be stored in the two dimensiona! array, BANK(4,3). This array is large enough to store a table made up of five rows and four columns and may be represented as follows: BANK(O,0) BANK(0, 1) BANK(0,2) BANK(0,3) BANK(1,0) BANK(1,1) BANK(1,2) BANK(1,3) BANK(2,0) BANK(2,1) BANK(2,2) BANK(2,3) BANK(3,0) BANK(3,1) BANK(3,2) BANK(3,3) BANK(4,0) BANK(4,1) BANK(4,2) BANK(4,3) As this is aNumeric Array, the date willl have to be stored as one number, e.g. 300684. Each entry in the table is stored ina separate element of the array. The date 300684 can be stored in BANK(0,0), the final balance at BANK(4,3). The following program shows you how to put data into a two dimensional array. The four columns of the array are accessed by taking values of C=0 to 3. The number of rows depends on the number of transactions entered, if there are T transactions, then R takes the values O to T-—1. You enter the date of each transaction and the amount credited or debited, and the balance after each transaction is calculated and entered into column four of the array. The total Credits, Debits and Balance are then found and entered into the last row of the array. The table headings are kept in aseparate string array. This array does not need to be DiMensionalised before use, as it is a one dimensional array and has less than eleven elements. All multi-dimensional arrays must be DiMensionalised before use. 10 REM Calculate bank balance 15 REM Read table headings into an Array 20 CLS 30 FOR C=0 TO 3 aO READ HEADINGS(C) SO NEXT C 60 REM Input transactions into a 2-D Array 70 INPUT “Number of transactions" 3T 80 DIM BANK(T,3) 52 90 N=T-1 100 CLS 110 FOR C=0 TO 2 120 PRINTTAB (10*#C) HEADINGS(C)3 130 NEXT C 140 FOR R=0 TON 150 FOR C=O TO 2 160 LOCATE 10#C, R+2 170 INPUT BANKCR,C) 180 NEXT C 190 NEXT R 200 REM Calculate balance after each transaction 210 BANK(O,3)=BANK(0O, 1)—-BANK(0O, 2) 220 FOR R=1 TON 230 X=BANK(R, 1)~-BANK(R,2) 240 BANKCR,3)=BANK(R-1,3)+X 250 NEXT R 260 REM To calculate total balance 270 FOR R=O TON 280 BANK(T,1>=BANKCR,1>+BANK(CT,1) 290 BANK(T,2)=BANK(R,2)+BANK(T,2) 300 NEXT R 310 BANK(T,3)=BANK(T,1)—-BANK(T,2) 320 BANK(T,0)=BANK(N,O) 330 REM To display bank balance 340 SCREEN O 350 WIDTH @O 360 FOR C=0 TO 3 370 PRINTTABC10#C) HEADINGS(C)3 380 NEXT C 390 FOR R=O TO N a00 FOR C=0 TO 3 410 LOCATE 10#C, R+2 a20 PRINT BANK(R,C) a30 NEXT C 4aO NEXT R a50 PRINT" es a60 FOR C=0 TO 3 4790 PRINTTAB (10*C) BANKC(T,C)3 a80 NEXT C 490 END SOO DATA DATE, CREDIT, DEBIT, BALANCE If you enter debits or credits greater than 99999.99, then you will mess up the table as only ten character spaces have been allocated for each entry. 53 EXERCISE 1) Change this program slightly so that you can INPUT the date in the form 23.6.84 etc. You will need to use another string array; DATES(T) would be suitable. 54 CHAPTER 14 MANIPULATING STRINGS There are quite a lot of ways you can manipulate strings. So far, you have only added one string to another: PRINT “Good"+" Day” is PRINTed as: Good Day By using INSTR you can look for a specified string within a string. Ina program for an adventure game, you might want to ask the player where he or she is going next. The game might take different branches depending on whether the player types in North, South, East or West. This part of a program shows you how this could be done: 100 INPUT “Where are you going now’:B$ 110 IF BS="NORTH” GOTO 200 120 IF B§="SOUTH" GOTO 300 130 IF BS="EAST” GOTO 400 140 IF B6="WEST” GOTO 500 150 PRINT “TRY AGAIN!” : GOTO 100 The player, however, might well have typed in a whole sentence in response to the prompt in line 100, e.g. lM GOING EAST. As the program stands, this would mean that none of the branches would be executed and the player would be asked where he/she was going again. This would be repeated until the single word “EAST” was entered and would considerably slow the game down. |f would be much better if the computer could scan the whole INPUT string and see if it recognised a word within the string. This in fact is exactly what the INSTR command does. The following example will show you how: 55 100 INPUT “Where are you going”; BS 110 IF INSTR(BS$,“NORTH") <> 0 GOTO 200 120 IF INSTR(B$,“SOUTH”) <> 0 GOTO 300 130 JFINSTR(B$, EAST”) <> 0 GOTO 400 140 IF INSTR(BS$,"“WEST”) <> 0 GOTO 500 150 PRINT “TRY AGAIN!” :GOTO 100 In line 110, the Command INSTR searches the main string, BS, for the smaller string “NORTH”. If it finds it, the number corresponding to the position of the first letter of “NORTH” in the main string, is returned. If EAST is not faund, INSTR returns zero. So, if B4 was: “I'M GOING NORTH NOW” then, INSTR(B$, “NORTH") returns the number 11, as the N of NORTH occurs at the 11th character position. To check this, type in this short program: 10 A=INSTR(“I’M GOING NORTH", “NORTH”) 20 PRINTA The number 11 should be PRINTed. INSTR always returns anumber, so it must always be equated to a Numeric Variable. Be careful to enclose strings, in INSTR statements, within quotation marks. String Variables already include quotation marks so these can be put as they stand within the INSTR statement. When searching for a string you can start from any specified point: INSTR(10,A$,B$) This starts searching for B$ in A$, at or after the 10th character position. This command would be useful in a Hangman program. The next program searches the word: CONSTANTINOPLE For the letter N. 10 A$="CONSTANTINOPLE” 20 BS="N" 30 C=0 40 C=INSTR(C+1,A$,B$) 50 IF C=0 THEN END 60 PRINT “there is a’; BS’ at character position’; C 70 GOTO 40 In this program, the expression C+ 1 has been used to denote where the search is to begin. The first time the INSTR command finds an “N”, C=3, the next search therefore starts from character position 4. After the third “N” has been found, C=10, and the fourth and last search starts from character position 11. There are no more 'N’’s to be found so C is set to zero causing the program to END at line 50. 56 You will always get a zero returned by INSTR if you try to look for a larger string within a smaller string: INSTR( “day”, “good day”) will return a zero. Now for three very similar functions: RIGHT$, LEFT$ and MIDS First of all RIGHT$. Here’s an example of its use: 10 A$=RIGHT$(‘‘The weather is terrible today’’,5) 210 PRINT A$ gives: today The RIGHT most 5 characters have been returned into the String Variable A$. | expect you can guess what LEFT$ does! Here’s an example anyway: 10 A$=LEFTS$(‘‘The weather is terrible today”,11) 20 PRINT A$ gives: The weather Finally MID$: 10 A$=MIDS(‘The weather is terrible today’, 13,17) 20 PRINT A$ gives: is terrible today The first number in the MID$ statement gives the character position of the first character ta be returned. The second number tells the computer how many characters are to be returned. So, in the above example, the 17 characters from the 13th position onwards have been returned into the Variable A$. The number used in these functions must be between 1 and 255. You cannot have a line longer than 255 characters anyway, so this makes sense. If the first numeric argument in MID$ is greater than the number of characters in the string, then you will get an empty string returned. Once again, do not forget quotation marks round strings and don’t try to equate any of these three commands to a Numeric Variable; you must use String Variables as strings are returned by these functions. 57 EXERCISES 1) Using INSTR, find the result of: a. searching for an empty or null string, i.e. “*” b. searching for a string within an empty string c. searching for an empty string within an empty string. PRINT out the words in ‘Is there anyone there?” separately. You could use INSTR to find the position of each word-and then either only repeat the use of MIDS, or use RIGHT$, LEFT$ and MIDS. Change the last program so that you can separate the words in any string which you INPUT. 58 CHAPTER 15 FUNCTIONS You have already met quite a few functions, for example: INT CHAPTER 4 INSTR CHAPTER 14 LEFTS CHAPTER 14 MIDS CHAPTER 14 Generally speaking, each of these acts in the same way. You have an object, called the argument, on which the function acts to produce a result. Once a function is defined, it always produces its results in the same way if given a Suitable argument: INT(X) X is the argument here. The argument of a function is always enclosed in brackets. The argument must be a number, a numeric variable, or a numeric expression. As long as X is in this range, INT will always return the integer value of X by rounding it down to the nearest whole number. INT(7.9) Returns 7 INT(—2.3) Returns —3 If you get confused with negative numbers, refer to the number line below: 6—-5-4-3-2-1 0 1 2 3 4 5... The numbers are arranged in order of increasing size, going from left to right. From this you can see that —3 is in fact the nearest lower whole number to —2.2. There is another very similar function to INT, called FIX. FIX also acts on areal number argument to return an integer. In the case of positive numbers, it returns the nearest lower integer like INT, 59 but for negative numbers it returns the nearest higher integer. unlike INT: FIX(5.5) Returns 5 FIX(—2.2) Returns —2 Now for two more functions which also have numeric arguments, ABS and SGN. SGN returns the sign of the argument: If the argument is a negative number, the function returns —1. If the argument is O, the function returns 0. If the argument is a positive number, the function returns +1. 10 A=SGN(—402.2) 20 PRINT “SGN(—402.2)=""A gives: SGN(—402.2)=—1 ABS, on the other hand, is used to convert negative numbers into positive numbers. Positive numbers are left unchanged by this function: 10 A=ABS(~ 402.2) 20 PRINT “ABS(—402.2)=""A gives: ABS(—402.2)= 402.2 The arguments of both these functions can be either expressions or variables: SGN(4 +3*—2) Returns — 1 ABS(percent) Converts the number in the numeric Variable, percent, to a positive number. Here is a short program to show you a possible use for ABS: 10 REM This program calculates your age 20 INPUT"Today’s Date: DAY, MONTH, YEAR" DT,MT,YT 30 INPUT"Your Date of Birth: DAY, MONTH, YEAR” DB, MB, YB 40 DA=ABS(DT-DB) 5O MA=ABS(MT-MB) 60 YA=ABS(YT-YB) 70 PRINT "You are"“;YAs"“years,"3MA;3"“months,"3DA; “days.” Now for two tunctions which deal with both strings and numbers, VAL and STR$. if a string contains a number, VAL will return the number into a Numeric Variable, i.e. it strips off the quotation marks and removes any blanks within the string. For example:. 60 A=VAL(" 273”) PRINT “A=";A gives: A= 273 If there is more than one number in a string, VAL only returns the first number it comes across. For this reason VAL does not evaluate expressions within strings: PRINT VAL('3+4=7") gives: 3 Do not forget that although the argument of VAL is a string, the result returned by VAL is anumber and must be put into a Numeric Variable: A$=VAL(7) IS WRONG VAL cannot cope with numbers embedded between letters in a string: 10 A=VAL(" Today is my 20th birthday.”’) 20 PRINTA gives: 0 The opposite function of VAL is performed by STRS: It converts a numeric argument into a string: A$="lam “+STR$(60) +" today” PRINT AS gives: | am 6O today Do not forget that although the argument is a number, the result is a string, so you must return it into a String Variable. In CHAPTER 13 we used a Numeric Array. BANK, to store dates, credits, debits and balances. Unfortunately the date had to be entered as anumber. i.e. 240684 rather than 24/6/84. To enter it in the second form, you had to store the dates separately in another string array, so the program ended up using one Numeric Array for the transactions, and two string arrays for the headings and dates. Using STR$, all the data could have been stored in one large String Array. To do this the following kinds of alterations would have to be made: 170 INPUT BANK 175 BANKS(R,C)=STRS(BANK) To calculate the new balance after each transaction, you would now first have had to evaluate the strings in the credit and debit columns, 61 using VAL, and then convert the answer back into a string: 210 BANK$(0,3)=STR$(VAL( BANK&(0, 1))- VAL(BANK$(0,2))) Perhaps you would like to go back and alter the BANK BALANCE program! One last function which acts on a string is LEN. LEN returns the length of the string given in its argument: 10 A$="“CONSTANTINOPLE” 20 A=LEN(AS) 30 PRINT “The word ":A§$;" is’;A;‘characters long.” Again, note that although this function acts on a string, it always returns a number, So you must equate this function to a Numeric Variable. Finally, a rather different function, RND. RND generates a 14 digit random number between 0 and 1. In fact, the numbers generated are not truly RaNDom as they follow a fixed sequence. However, this sequence is so long, that for all intents and purposes, you can consider the numbers generated to be truly random. You can have three types of arguments with RND: If the argument is positive, the number generated by RND within a program is the next one in the sequence. The point at which the sequence is entered is always the same, and is reset by the computer at the end of the program. Type in: 10 PRINT RND(1) 20) PRINT RND(1) 30 PRINT RND(1) You will get the following sequence each time the program is RUN: 8952 1943994623 .10658628050158 .16877651772823 Changing the argument to a different positive number will have no effect. lf the argument is zero, the random number generated is the same as the last one. For example: 10 X=RND(1)+RND(1) 20 PRINT'The second random number generated was"; RND(0) 30 PRINT“ The first random number generated was”; X—RND(O) If the argument is negative, the part of the sequence from which the random numbers are taken is set by the negative argument: 62 10 A=RND(-3) 20 B=RND(1) 30 PRINTAB gives the seqence: 8438982042082 1 .29624868 166920 each time the program is RUN. Now replace —3 by —4. You will get a different sequence. Changing the positive argument will again have no effect. A random number between 0 and 1 is rarely of much use. You are much more likely to want a number between 1 and 6, for example. This program simulates the throwing of a dice: 10 REMdice program 20) A=INT(6*RND(1) +1) 30 PRINT You have thrown a"sA However, you get the same number each time the dice is thrown. If you want to get a different random number each time, put the line: RDRST=RND(-—TIME) at the beginning of your program. The TIME function returns the value of the computer's internal clock. On switching on the machine, the clock is set to O but is incremented every 1/50th of a second, i.e. TIME will equal 50 when 1 second has elapsed. So using TIME as the argument of a RND function, will result in the sequence of random numbers being read at a different point each time the RND function is encountered in the program. Do you see how you might use this and the last function in the Hangman game? The game might take this sort of a structure: 1) READ aword at RaNDom from a DATA statement (these statements could be put together after the END of your program). To READ a line at RaNDom, use RESTORE followed by a randomised line number. Think how you might access more than one word in a DATA statement. 2) Once you have READ your word, let the player know how long it is, using LEN. 3) Ask the player to guess a letter. 4) Using INSTR, check to see if the letter is in the word. (See CHAPTER 14 if you have forgotten how to do this.) 5) Count how many guesses the player has made. 6) (f the player has not got the word after 10 guesses, END the program. The computer has won. 7) Ask the player if he/she wishes to play again by using INKEY$. 63 EXERCISE 1) Write the Hangman Program! 64 CHAPTER 16 DEFINING YOUR OWN FUNCTIONS So much for the computer's defined functions, let us now define some of our own. This can be done using DEF FN. Once you have defined a function and given it aname, you can use it anywhere in the program. Let us take a mathematical example first. Supposing you want to define a function which will calculate the square of a number. |.e. raise the number to the power of two, which is the same as multiplying the number by itself. Let us call this function SQUARE and define it by: DEF FNSQUARE (x) = X *2 Note that the name of the function comes directly after the command DEF FN. X is what is called a ‘dummy parameter”. You only use it to show what the function does. When using the function, you must pass a real parameter into it to replace the dummy parameter. This real parameter has to be of the same type as the dummy parameter. In the example given therefore, the real parameter can be either a Numeric Variable or number. When defining a function, the dummy parameters used in the definition must be listed within brackets after the function name. Let us now use this function to calculate the SQUARE of 13, and put the result into the variable R. To do this, we use the command FN, followed by the name of the function and then the real parameter in brackets: R=FNSQUARE (13) On seeing this, the computer will look for the definition of the function SQUARE. On finding it, all the “X’’s in the definition are replaced by 13, 65 the result calculated, and entered into the variable R. X is also called a Local Variable. This means that it is only recognised locally, i.e. within the function: 10 X=4 20 DEF FNSQUARE(X)= X * 2 30 R=FNSQUARE(12) 40 PRINT'R=";R;" X=":X gives: R=144 X=4 You Can use more than one parameter in a function. All the param- eters used must be listed in the definition, the parameters being separated by commas. When calling the function, you must pass the same number of real parameters into the function as dummy param- eters used in the definition. You must also pass the correct type of parameter tn, !.e. anumber or Numeric Variable for a numeric dummy parameter. The next program will raise any number to the power of another number, both numbers being passed into the function definition: 10 DEF FNPOWER(X,Y)= X * Y 20 INPUT “Number”: A 30 INPUT “Exponent”; B 40 R=FNPOWER(A,B) 50 PRINTA;" *"-B:“=':R The definition of a function must fit into one program line. Now for an example using strings. This function chops off the first character of a string: 10 DEF FNSTS$(A$)=RIGHTS(A$, LEN(A$)—1) 20 INPUT “string” S$ 30 PRINT S$,FNSTS(S$) As this function returns a string, the function name must end with the ‘$" sign. Any valid Variable Name is acceptable as a function name. EXERCISE 1) Define a function which cuts off the last four characters of a string. 66 CHAPTER 17 STRUCTURING YOUR PROGRAM Often in a program you may find yourself wanting to do the same task many times, but at different points in the program. If you do find that it is necessary to repeat a program line many times, it is advisable to put that line into what is called a ‘subroutine’. When you need the line in your program you jump to the subroutine, using the GOSUB statement. When all the lines in the subroutine have been executed, you RETURN to the main program. A subroutine may contain as many program lines as necessary. In the following example, the subroutine lies between program lines 100-130: 10 CLS 20 PRINT “DO YOU LIKE ME?" 30 GOSUB 110 40 PRINT “DO YOU THINK I°M BRAINY?” SO GOSUB 110 60 PRINT “DO YOU THINK I°M GOOD LOOKING?" 70 GOSUB 110 80 PRINT "YOU"RE JUST SAYING THAT TO PLEASE ME!" 90 END 100 REM SUBROUTINE 110 INPUT “YES OR NO "3s YNS 120 IF YNS="NO" THEN PRINT “ OK. I*LL GO AWAY THEN." %END 130 RETURN Notice the RETURN statement at the end of the subroutine. This is very important, as it tells the computer to RETURN to the main program. Execution continues from the first line after the one containing the 67 GOSUB statement. It is a good idea to put a REM statement just before a subroutine. to distinguish it from the main program. Do not use the line number of the REM statement when calling the subroutine, however, as you might later want to get rid of all REM statements, if you start to run out of computer memory. Make sure the computer does not execute the subroutine by accident, by putting an END statement at the END of the main program, just before the subroutines. Another use of the GOSUB statement, is in the IF. . THEN. . .ELSE statement. This statement tends to get rather long and complicated: the easiest way to simplify it is by using a subroutine. The following short program asks for two different. positive numbers. The IF... THEN ...ELSE statement is used to test these numbers. if they are different and positive, their product and sum are PRINTed out. Otherwise you are asked to try again: 10 INPUT“ TWO POSITIVE, DIFFERENT NUMBERS” A,B 20 IF A=B OR A<0 OR B<0 THEN PRINT “TRY AGAIN!" ELSE PRINT “A=".A, °B=".B:PRINT “A+B=": A+B: PRINT “AXB=")A*B 30 GOTO 10 Now, using a subroutine: 10 INPUT “TWO POSITIVE, DIFFERENT NUMBERS"; A,B 20 IF A=B OR AXO OR B<O THEN PRINT “TRY AGAIN!" ELSE GOSUB 100 30 GOTO 10 90 REM SUBROUTINE 100 PRINT "A="sA,"B="3B 110 PRINT “A+B=";A+B 120 PRINT “AxB="3A*B 130 RETURN the program is slightly longer. bui much easier to read. 68 CHAPTER 18 PROGRAM BRANCHING At some point you might find that you would like to have amain program, or menu program, which offers you a number of options. For example, the first few lines of the menu program might cause the foliowing to be displayed on the screen: 1. Display Telephone Directory 2. Add New Entry to Directory 3. Remove Entry from Directory 4. End program ICCC e eee See eee ee eas The next part of the program may look something ike this 100 AS=INKEYS$ 120) iF A$="1° THEN GOSUB 1000 130 IF A$="2" THEN GOSUB 2000 140 !F A$="3" THEN GQSUB 3000 150 IF A$=="'4" THEN GOSUB 4000 160 GOTO 100 Line 160 is included in case a key other than 1.2.3, or 4 is pressed. There is a shorter way of writing this, using ON. ..GOSUB. Using this statement. you can jump to any number of subroutines you wish. In the above example, we want to be able to jump to one of four subroutines: we therefore list the line numbers of the four subroutines after the werd GOSUB: ON... .. GOSUB 1000,2000,3000,4000 Now we have to let the computer know which subroutine to jump to. This is done by dutting anumber or expression after the ON, and before the GOSUB Ifthe expression evaluates to 1, thenthe computer jumps io 69 the first subroutine listed after the GOSUB,; if it evaluates to 2, the second subroutine, and soon. We can now replace lines 100-160 by: 100 A$=INKEY$ 110 A=VAL(A$) 120 ONAGOSUB 1000,2000,3000,4000 130 GOTO 100 Notice that the string, A$, has been changed into the Numeric Variable A, before using it in the ON. ..GOSUB statement; otherwise this would have caused an errror, as only Numeric Variables may follow the word ON. On finishing the subroutine, the computer RETURNS to line 130. if the Numeric Variable, A, had contained a real number, e.g. 2.9, the computer would have taken the nearest lower INTeger value of this, and jumped to the second subroutine. The expression must, however, evaluate to a number between 1 and the total number of line numbers listed. ON. . .GOTO works in exactly the same way as ON. . .GOSUB. In this case, the computer jumps to the line number in the list which corre- sponds to the number following the ON statement. Execution of the program continues from this line number: 10 INPUT “1,2, 0r73":A 20 ONAGOTO 370,420,10 So, if: A=1_ the next line executed is line 370 A=2_ the next line executed is line 420 A=3_ the next line executed is line 10 Notice that the line numbers do not have to be innumeric order in the list. 70 CHAPTER 19 MATHEMATICAL FUNCTIONS If you have not done much mathematics and find this chapter rather heavy going, do not hesitate to skip to the next one. . This chapter is going to deal with the mathematical functions: * _SQR TAN, SIN, COS and ATN EXP and LOG . You already know how to raise a number to a power by using the sign: A= 16 (or as it is usually written 4° = 16) In general, raising anumber n, to a power p, is equivalent to multiplying n, by itself, p times. In the above example, n=4, p=2. So 4 * 2is equivalent to 4°4. The SQuare Root function, SQR, finds the square root of a number. This is the inverse of raising anumber to the power of two, or squaring it. if you square 5, i.e.5 * 2, you get 25. To get back to 5 from 25, you take the SQuare Root: 10 A=SQR(25) 20 PRINT “The square root of 25 is "A giving: The square root of 25 is 5 The argument of the SQR function must be a positive number or an expression which evaluates to a positive number. (The computer can't handle imaginary numbers!) Here are a few more points to think about. . What happens if you raise a number to a negative power? It is easy to 71 find out. By typing in: PRINT 4 * --4 you get : 25 whichis 1/4, so in general: n* —~p=1/n*p Let us see what happens if a number is raised to the power of a decimal number, or fraction. Typing in: PRINT 4 * 0.5 gives you the answer 2. 4 * 0.5 is equivalent to 4”, so in general: n~ 1/p gives you the pth root of n In the example, n=4, p=2. so you got the second, or as it is better known, the square root of 4, whichis 2. So, raising anumber to the power of a half, is equivalent to taking the square root of the number. SIN, COST, TAN and ATN are trigonometic functions. The arguments of these three functions must be angles. You are quite possibly used to thinking of angles in terms of degrees: Y-axis X-axis A circle subtends 360 degrees at its centre. The angie of the circle is measured from the X—axis, which bisects the circle with a line drawn trom the nine o'clock position to the three o'clock position, in the anti- clockwise direction. The Y-axis bisects the circle from the twelve o-clock position to the six o'clock position. The angle between these two 72 axes is therefore 90 degrees. However, the computer measures angles rather differently. To under- stand how, let us consider a circle with a radius of one unit. The type of the unit does not matter here: It canbe 1mm, 1m, 1km— anything in fact. The circumference of the circle is given by the formula: C=2*Pi*r C=Circumference P| is a special number used in maths, which starts 3.1415926.. Plis the Greek “p", and pronounced PI. ris the radius of the circle Therefore, for our circle, as r= 1: C=2"Pl Looking at this another way, we can say that the circumference of our circle subtends an angle of 2*Pi radians ai the centre of the circle. So: 2*PI| radians is equivalent to 360 degrees. In radians, therefore, the angle between the X and Y axes is Pl/2. lf we now have a point moving round on our circle, we can use the functions COS and SIN to find out where the point is: Y-axis a units X-axis At a certain time, the point might be at position P, where it has moved through an angle of a radians from the X axis, or if you prefer, it has moved a distance of a units along the circumference. \f we draw vertical and horizontal dotted lines from P, to cut the X and Y axes, we can find the position of the cuts on the X and Y axes by using 73 COS and SIN respectively: x=COS(a) COS stands for the Cosine of angle a Y=SIN(a) SIN stands for the Sine of angle a When we measure along the X-axis, we consider x to be positive if we are on the RHS of the Y-axis, and x is negative when we are on the LHS of the Y-axis. So we can predict: The COS of any angle between 0 and PI/2 is positive. The COS of any angle between PI/2 and 3*PI/2 is negative. The COS of any angle between 3*P1/2 and 2’P1 is positive. i.e. For COS: Pl/2 3/2P| When we measure along the Y-axis, y is positive when above the X-axis, and negative when below. So we can predict: The SIN of any angle between 0 and PI is positive. The SIN of any angle between PI! and 2*PI is negative, i.e. for SIN PI/2 SIN POS. PI O SIN NEG. 3/2P| 74 If we go round the circle more than once, COS and SIN take the same values as the first time round: SIN(a+27Pl) = SIN(a) COS(a+2*Pl) = COS(a) The tangent of an angle is given by: TAN(a)=SIN(a)/COS(a) Given the tangent of an angle, we can find the angle using ATN (which stands for arctan). ATN is the inverse function of TAN: a=ATN(0.6) returns the angle whose TAN is 0.6 into the variable a. The angle is returned in radians. The computer does not know the inverse functions of COS and SIN, but you can easily define your own functions for these using: ARCSIN(a)=ATN(a/SQR(—a’*a+1)) ARCCOS(a)=ATN(a/SQR(~—a*a+1))+P1/2 Another number often met in mathematics is represented by the letter e To have a look at the first 14 digits of this number, type: PRINT EXP(1) You will see: 2.7182818284588 The function EXP is defined by: EXP(x)=e * x So: EXP(1)=e The inverse of this function is given by the LOG function. This function gives you the log to base e of the argument, i.e. it gives you the natural logarithm of a number which is often written In. It is related to the EXP function as shown below: EXP(x)=e * x=n LOG(n)=x You might be more used to taking logs to the base 10. To get log tothe base 10, use the following identity: log10(x)=LOG(x)/LOG(10) 15 EXERCISES 1. Define a function called PI, which returns the value of Pl. Use the identity: PI=ATN(1)*4 Define a function which converts degrees to radians. Use the formula: a radians = (a*Pl)/180 degrees You will have to use 3.14... for Pl. Alternatively, use the function PI. 3. Define a function which gives the ARCCOS of its argument. Do not forget, -1 <= COS(a) <= 1, for any angle of a radians. So your function will only work for arguments between + and —1 inclusive. 4. Define a function which returns the logarithm to base 10 of its argument. in) 76 CHAPTER 20 THE ASCII CODES In CHAPTER 6, you discovered that the letters A-Z and a-z are stored inside the computer as the numbers 65-90 and 97-122 respectively. These numbers are called ASCII Codes. ASCII stands for the American Standard Code for Information Inter- change, and is commonly used by computers to represent characters. If you want to find the ASCII Code of a character, you may use the Command ASC. For example: 10 Al=ASC("A"):A2=ASC(‘a’’) 20 PRINT “The ASCII Code of Ais ":A1 30 PRINT “The ASCH Code of ais ":A2 The argument of the ASC function must be a string; if more than one character is contained in the string, the computer returns the ASCII code of the first character. The opposite function to ASC is CHR§$. When given a suitable numeric argument, CHR$ returns the corresponding ASCII CHaRacter, into a string. For example: 10 A1$=CHR$(65):A2$=CHRS§(97) 20 PRINT’The Character corresponding to ASCIl Code 65 is "\A1$ 30 PRINT'The Character corresponding to ASCII Code 97 is “A2$ The other characters on the keyboard also have ASCII Codes. Find the ASCII Codes of “%", °?” and “7”. In fact there are 256 ASCII codes altogether, lying between 0 arid 255. You can use CHR$ to find the characters associated with the codes between 32 and 255 inclusive. RUN the following program to find these characters: 77 10 REMASCII characters: 32-255 20 CLS 30 FORI=32 to 255 40 PRINT CHR§(i)+" ": 50 NEXT| You will see all the alphanumeric and code characters displayed on the screen, with most of the Graphics Characters. Notice that the last character, which is represented by the code 255, is in fact the cursor; and so changes as you move the cursor over text on the screen! The codes between 0 and 31 are rather special. These codes are called Control Codes. Instead of representing a character, they represent an operation. which is to be carried out by the computer when the code is used as an argument in a PRINT CHR$ statement. You have already met Control Code 13 in CHAPTER 8. Taking PRINT CHR$ of this was equivalent to pressing <RETURN>. This is the only 10 FORB=1T010 20 A$=STRINGS(B,42) 30 PRINT A$ 40 NEXTB gives the pattern: * ak koe we AOR Ke SOI ROR IO Oe ORO RO Ok OKO kk KKK KK RK If the second argument is a string, then the first character of this string is used, |e. if line 20 were to be replaced by: 20 PRINT STRING$(B,‘**”) the effect would be the same. way <RETURN> can be embedded within a program. Other Control Codes move the cursor around, clear the screen etc. Try RUNning the next program: 10 REMsome ASCII Control Codes 20 PRINT CHR$(12)+"Control Code 12 Clears the Screen.” 30 PRINT CHR$(7)+"'Control Code 7 BEEPs!”’ The codes 0 to 31 are also used to represent the rest of the Graphics Characters. To access these, however, you must PRINT CHR§(1) first, and then CHR$(A+64) where A is the code number. The next program PRINTs out the first 32 Graphics Characters: 10 REM Codes 0—31: more Graphics Characters 20 FORA=0T031 30 PRINT CHR$(1)+CHR$(A+64)+" ": 40 NEXTA The Keyword, STRINGS, is used to PRINT a string of a specified length and made up of a specified character. The string cannot exceed 200 characters, unless you first use a CLEAR statement. STRINGS must be followed by two arguments, separated by a comma, and enclosed in brackets. The first argument must be anumber, numeric expression or Numeric Variable. This specifies the length of the string. The second argument may be either anumber or a string. If anumber is used, this must be the ASCII Code of the character to be PRINTed in the string. For example: 78 79 CHAPTER 21 THE SCREEN MODES So far you have only used the Text Screens, Screens 0 and 1. There are in fact two more screens, screens 2 and 3. We will be using these in the next few Chapters. First of all, a summary of a few of the characteristics of each screen. The Text Screens You can alter the width of both these screens by using the KEYWORD WIDTH (see CHAPTER 9): SCREEN MODE DEFAULT WIDTH MAXIMUM WIDTH CHR SIZE SCREEN 0 37 40 6x8 SCREEN 14 29 32 8x8 Both screens hold 24 tines of text Although you can fit more characters across Screen 6 than Screen 1, the space allocated for each character is smaller. Oniy 6x8 poirits, or pixels as they are usually called, are used to form each character in Screen 0, whereas 8x8 are used In Screen 1. This means that the ailpnanumenc characters appear more compressed in Screen 0 than on Screen 1, as the last two pixel columns in the 8x8 pixel squares, used In Screen 1. are left blank, anc itis these two columns which are missing from the character pasitions tri en 0. The compression of the character positions is more obvious when comparing the graphics characters On turning on the computer. one of these text screens is ihe default screer ; The MSX ts capaole of producing 16 colours (see CHAPTEN 22 for details). In Screen : you oan set the foreground (text). background and 80 border colours to any combination of these 16 colours. You cannot have more than one foreground colour at any one time. The default colours are: FOREGROUND WHITE BACKGROUND DARK BLUE BORDER CYAN In Screen 0 you can only set the foreground and background colours, the default colours being the same as above. If at any point you change the foreground colour of a text screen, all the text on the screen changes colour. The High Resolution Graphics Screen Screen 2 is the high resolution Graphics Screen. It is 256 pixels wide, by 192 long. Each pixel can be resolved on this screen. The top left hand pixel is at co-ordinate position (0,0), the bottom right hand pixel at (255,191): SCREEN 2 Q 255 191 You cannot alter the width of this screen. If you have a WIDTH statement while in Screen 2, the computer will remember it, and on returning to a TEXT Screen will set the WIDTH as specified while in Screen 2. You may have typed: SCREEN 2 while in Command Mode and been surprised that nothing has changed. This is because you cannot access the graphics screens from Cemmand Mode. You must include the SCREEN Cornmand in a program in order to view the Graphics Screens, i.e. 10 SCREEN 2 20 GOTO 20 Line 20 freezes the Graphics Screen; this is necessary as the com- puter immediately returns to the default Text Mode on completing 81 execution of a program. It will also return to the Text Screen on encountering an INPUT statement in a program, or an Error. To stop this program and return to the Text Screen, press <CNTL><STOP>. The pixels in this screen are grouped together in rows 8 pixels long. The first pixel row ties between co-ordinates (0,0) and (8,0): [tt | Each row of 8 pixels can only contain two colours, a foreground colour and a background colour. Adjacent rows may contain two different colours. The colour resolution of this screen is therefore 32x 192. To PRINT characters on this screen, see the ADVANCED SECTION, CHAPTER 14. You cannot use the PRINT Command directly. One 8x1 Pixel Row The Multi-Colour Screen Screen 3 is the Multi-Colour Screen and can be used for low resolution graphics. Like Screen 2, it is 256 pixels wide by 192 deep, but you can't resolve each pixel on this screen: SCREEN 3 255 191 The smallest point you can plot on this screen is a block of 4x4 pixels. You can still plot point (0,0), but you will find a block has appeared on the screen whch also covers points (1,0), (2,0), (3,0) to (3,3). So plotting (3.3) would have the same effect. 82 One 4x4 Pixel Block Each 4x4 block can only contain one colour, the foreground colour. Therefore the colour resolution of this screen is 64x48. Like Screen 2, you cannot access this screen from Command Mode: you will have to freeze the screen by using an infinite loop. An INPUT Command or Error will force you back to the default Text Screen. 83 CHAPTER 22 COLOUR On switching on the computer, the colours are preset to give you a dark blue background with a cyan border. The foreground, or text, is white. You can change any or all of these colours by using the COLOR Command. Note the American spelling of the word! After receiving the command COLOR, the computer expects up to three numbers, separated by commas. The first number sets the foreground colour, the second the background and the last the border. There are 16 colours for you to use: 6) Transparent Black Medium Green Light Green Dark Blue Light Blue Dark Red Cyan Medium Red Light Red/Pink Dark Yellow Light Yellow Dark Green Magenta Grey White OMDN DOP WDNY = a nt WM +] O 84 so the default colour is: COLOR 15,4,7 The Function Key, F6, is predefined to print COLOR 15,4,7 on the screen, followed by <RETURN>. So if you accidently set the fore- ground colour to the background colour, press this key! Key F1 is set to print COLOR on the screen. You do not have to type in all three numbers for the foreground, background and border when using COLOR. Just give a colour code for the one you want to change. For example: COLOR 1 This sets the foreground to black, leaving the background and border as they were. COLOR, 15,15 The background and border are changed to white, leaving the fore- ground black. Notice that, although the foreground colour has not been specified, the comma following it must not be left out, as this tells the computer that the next number sets the background. COLOR,,10 The above command sets the border to dark yellow. You must be in Screen 1,2 or 3 to observe this, as Screen 0 doesn’t have a border. lf you change the background colour while in either of the graphics screens, you will have to execute the Command CLS to observe the efect. This clears the screen to the new background colour. The next program displays all the border and background colour combinations available on the MSX. Type this in on Screen 1: 10 REM colours 20 FOR BDR=O TO 15 30 COLOR, ,BDR 40 FOR BACKGD=0 TO 15 SO COLOR,BACKGD 60 FOR DELAY=1 TO 200 =:NEXT DELAY 70 NEXT BACKGD 80 NEXT BDR 90 COLOR 15,4,7 The DELAY loop in line 60 gives you time to see the different colour combinations; without this line, the colours would flash by too quickly to be seen properly. Notice that you can use Numeric Variables and Numeric Expressions in the COLOR statement. Numeric expressions, however, won't work here. If you use a real number in a COLOR statement, it will be truncated 85 to the nearest integer. The number must be in the range of 0 to 15. Try running this program on the other screens. You willneed to insert a CLS Command for Screens 2 and 3. Have you realised what COLOR 0 does? It sets the background or foreground to the same colour as the border. If used to set the border colour, the border becomes black. EXERCISE 1. Set one of the Function Keys to give a different foreground, back- ground and border combination. 86 CHAPTER 23 PLOTTING POINTS The first two commands dealt with in this Chapter, PSET and PRESET, are very similar. As they are Graphics Commands, they can only be used in Screens 2 and 3. PRESET can be used to set a point to the background colour. The co-ordinates of the point can be either absolute or relative. After using PRESET. you will find that the current foreground colour is the same as the background. In both Screens 2 and 3, the x co-ordinate must be in the range of 0 to 25, the y co-ordinate between 0 and 191. Absolute co-ordinates are read directly off the X and Y axes: X-AXIS 0 255 Of | a | | | Y | | | | A | | Ce R | | (100,90) | S | 1 ee. an P 191 (230,162) The point R is at the absolute x co-ordinate 100, and y co-ordinate 90. P is at absolute co-ordinates (230,162) A relative co-ordinate gives you the position of a point as 87 measured from some reference point. For example, the point P is at relative co-ordinates (130,72), measured with respect to R.. This means It is 130 points further along the X axis than R, and 72 points lower down the Y axis. in all the Graphics statements, the reference point is always taken to be the last position of the Graphics Cursor. The position of the Graphics Cursor can be set by PRESET: 10 SCREEN 2 20 PRESET (100,100) 30 GOTO 30 If you run this program, nothing appears to happen! However, the Graphics Cursor is now positioned at the point (100,100). Any Graphics command immediately following this, using relative co-ordinates, will now use the point (100, 100) as its reference point if the next Graphics statement draws a line, for example, you must specify the colour of the line, or reset the foreground colour first, with the COLOR statement. Otherwise your line will be invisible! If you specify a colour ina PRESET statement, a point of this colour will be plotted, and this colour becomes the current foreground colour. Replace line 20 by: 20 PRESET (100,100),1 The number following the PRESET statement must be one of the COLOR Codes as given in CHAPTER 22, i.e.anumber between 0 and 15. If areal number is used in this range, the decimal part will be ignored. lf your co-ordinates are off the screen, obviously you will not see anything! If the co-ordinates are out of the allowed integer range — 32768 to 32767 you will get an Error Message. The command PSET plots a point of the foreground colour at the specified co-ordinates. These co-ordinates can be either absolute or relative. If a colour is specified in the statement, then a point of this colour is plotted, so this is identical to using PRESET with a specified colour. As with PRESET, this statement also changes the foreground colour to the specified colour. The following program plots blocks of colour, at random, onto Screen 3, using the command PSET. These blocks are then blotted out by resetting the same random points to the background colour using PRESET: 88 10 REM spots 20 R=TIME 30 RDRST=RND(-R) aO SCREEN 3 SO REM plot coloured points at random 60 FOR I%=1 TO 300 70 GOSUB 1000 80 PSETC(X%,Y%),Z% 90 NEXT I% 100 REM reset points to background colour 110 RDRST=RND(-R) 120 FOR J%=1 TO 300 130 GOSUB 1000 140 PRESETCX%,Y%) 150 NEXT J% 990 REM subroutine randomise 1000 X%=RND(1)#256 1010 Y%=RND(1)#192 1020 Z%=RND(1)#16 1030 RETURN Integer Variables are used throughout this program, to speed execution up, i.e. X% is used instead of X. Each time the program is RUN, the points will be plotted at different random positions. This is because the point at which the sequence of random numbers is entered is set by the Variable TIME, and this variable will have a different value each time the program is RUN. The same random numbers are generated in the second half of the program, as in the first, so all the plotted points are blotted out. To generate the same numbers, the random number sequence is reset by using the same negative argument in the RND function in line 110 as was used pre- viously in line 30. Finally, the command POINT. This is another Graphics Command. It returns the colour code of a point, in either of the Graphics Screens. The number returned is therefore in the range O to 15, representing the colours Transparent to White. —1 is returned if the co-ordinates in the POINT statement are off the screen. Type: 10 COLOR ,2 20 SCREEN 2 30 P=POINT(100, 100) Now, back in Command Mode, type: PRINT P The number 2 will be displayed, as this is the colour code for mid-green, which is the new background colour. 89 EXERCISE 1. Try plotting a SINe curve, using: Y=A*SIN(X)+A A sets the amplitude of the curve. Vary X over the range 0 to 2"PI. 90 CHAPTER 24 DRAWING LINES AND BOXES This Chapter deals with the command LINE. This is another of the Graphics Commands and so can only be used in Screens 2 and 3. Drawing Lines When drawing a line, you may specify the start and end positions, which can be done using either absolute or relative co-ordinates, or even a mixture of the two. Type this program in: 10 SCREEN 2 20 LINE (0,0)—(255,191) 30 GOTO 30 This program draws a LINE diagonally across the screen. Notice the “—" sign between the co-ordinate pairs. Absolute co-ordinates have been used. The first pair of co-ordinates, (0,0), gives the starting point of the LINE; (255,191) gives the end position of the LINE. Change line 10 of this program, so that it Runs in Screen 3, to compare the resolution of the two Graphics Screens. Try using different co-ordinates in the LINE Command. if your x or y co-ordinate is greater than 255 or 191 respectively, the computer draws to the edge of the screen, taking x as 255 andy as 191. if, however, your co-ordinates are out of the permitted integer range, ~32768 to 32767, you get an Error Message. You do not have to specify the starting point of the LINE. If this is unspecified, the computer starts drawing from the present position of 91 the Graphics Cursor. If you have just drawn a LINE. the Cursor will be at the end position of the LINE. You could have used relative co-ordinates to draw the same LINE. Replace line 20 of the program with: 20 LINE (0,0)-STEP(255, 191) The statement STEP tells the computer that the following co-ordinates are relative. So, in this case the final x position will be 255 points further along the X axis than the starting point, and the final y position 191 points further down the Y axis putting the last point of the LINE at absolute co-ordinates (255, 191). Coloured Lines So far you have only drawn lines in the current foreground colour. It is possible to specify the colour of the LINE, however, by following the LINE Command by a comma and then a valid COLOR Code, i.e. a number between 0 and 15. See CHAPTER 22 for the COLOR Codes. Now replace line 20 of the last program with: 20 LINE — (255,191),10 This draws a diagonal yellow LINE across the screen. The next program demonstrates the colour resolution on Screen 2. 10 SCREEN 2 20 FORA=07T015 30 LINE(0+A,0)—(200+A,191),A 40 NEXTA 50 GOTO 50 You will see 16 diagonal lines drawn across the screen but their colours are rather blurred. If you now replace Line 30 by: 30 LINE(0,0+A)—(255,0+A),A You will be able to resolve the colour of each line. You get these effects because of the way the screen is stored in the computer's memory. Each horizontal line is divided up into groups, 8 pixels long. Each group may only contain two colours at any one time, a foreground and a background colour. On plotting the first diagonal line, the foreground colour is set to 0 and the line is drawn in this colour. The next line sets the foreground colour to 1, but this changes the fore- ground for the complete group of 8 pixels, so the first pixel of the first line changes colour to black. The 8th line plotted changes the foreground colour of the first group of 8 pixels for the last time; this complete group is now cyan, The vertical colour resolution is somewhat better than the horizontal 92 resolution. The colour of each vertical pixel is independent of the pixels above and below, hence the 16 horizontal lines do not interfere with each other. Try running the two programs on Screen 3. On this screen, you will see the same four lines each time, as the horizontal and vertical colour resolutions are the same. You can have one colour for every 4x4 pixel block. The colour resolution is therefore the same as the graphics resolution. The first four lines are all plotted on top of each other, so you only see the last one which is light green. Drawing Boxes It is possible to draw a box using four LINE Commands, one after the other. However, there is an easier way to draw boxes if you know the co-ordinates of the top left and bottom right hand corners. You still use the LINE Command, but now the first co-ordinate pair gives the co- ordinates of the top, left corner and the second, of the bottom right hand corner. To let the computer know you want a box and nota line, you must follow the COLOR Code by another comma and then the letter B” for Box. Type in the following program: 10 REM Boxes 20 SCREEN 2 30 FORA=0TO 90 STEP 10 40 LINE (6+A,5+A)—(255—A,185—A),A/10,B 50 NEXTA 60 GOTO60 This program will draw 10 boxes of different sizes and colours. Notice that you can use an expression for the COLOR Code, but this expression must evaluate to a number between 0 and 15. Ifit evaluates to areal number, the decimal part will be truncated. See the effect of running this program on Screen 3 by changing line 20 to: 20 SCREEN3 lf you know the lengths of the sides of the box, rather than the co-ordinates of its corners, it is easier to use relative co-ordinates for the second co-ordinate pair, i.e. to draw a box 100 pixels long in the X direction, and 50 pixels deep in the Y, you could use: 10 SCREEN 2 20 LINE (10,10)—STEP(100,50),,B 30 GOTO 30 The program draws a box of the correct dimensions in the current 93 foreground colour. The top right hand corner is situated at absolute co-ordinates (10,10). The second co-ordinates pair are now specifying the lengths of the sides of the box which are parallel to the X and Y axes. This is the best method to draw squares. For example: 10 REM squares 20 SCREEN2 30 FORA=10TO 90 30 LINE (A,A)—STEP(A,A),,B 40 NEXTA 50 GOTO50 The last program draws squares, of increasing size, diagonally across the screen. Coloured Boxes To draw a box and fill it with a specified colour, replace the letter “B” by “BF. The colour specified in the LINE statement is now used to paint the box: 10 REM black box 20 SCREEN2 30 LINE (20,20)—(100, 100), 1,BF 40 GOTO 40 EXERCISE 1. Write a program to draw a box of a random size, at random, on the screen. Fill this box with a random colour. 94 CHAPTER 25 DRAWING CIRCLES AND ELLIPSES The CIRCLE Command enables you to draw the whole, or part, of either a circle or an ellipse. You must specify the position of the centre of the CIRCLE, in either absolute or relative co-ordinates, and the length of the radius. You have the option of specifying the colour of the CIRCLE and how much of the CIRCLE is to be drawn. CIRCLE is a graphics Com- mand and can only be used on the Graphics Screens. Drawing Circles To draw a CIRCLE, you must specify the central co-ordinates of the CIRCLE, and the length of the radius. For example: 10 REM drawing circles 20 SCREEN 2 30 CIRCLE (128, 100),90 40 GOTO 40 This draws a CIRCLE centred at absolute co-ordinates (128, 100), witha radius of 90 pixels, in the current foreground colour. Obviously, if you specify a radius which is too large for the screen, you won't see your CIRCLE, or you will just see the part which fits on the screen. To specify the colour of the circle, follow the radius by a comma, then a valid Colour Code, i.e. anumber between 0 and 15. (See CHAPTER 22 for the Colour Codes.) The next program draws the same CIRCLE, but in black: 10 SCREEN 2 20 CIRCLE (196, 100),90, 1 30 GOTO 30 95 Drawing Arcs To draw just part of a CIRCLE, or an arc, you must specify the start and end angles. The next program draws three quarters of a black CIRCLE: 10 REMdrawing arcs 20 SCREEN 2 30 PI=4*ATN(1) 40 CIRCLE (128, 100),90,1,0,3*P1/2 50 GOTO 50 The last two numbers we have added give the start and end angles for the arc. The angles must be specified in radians. (If you cannot remem- ber what a radian is, See CHAPTER 19.) Line 30 only calculates Pl, so the arc is drawn between the angles 0 and 3/2 PI radians: PI/2 Pl ) 3/2 Pl All angles are measured from the X axis, in the anti-clockwise sense. As there are 2*Pi radians in a CIRCLE, the specified angles can be any- thing between 0 and 2*PI. Drawing Pies To do this put a “—” sign prefix before the start and end angles. This will result in lines being drawn from the centre of the circle to the start and end positions of the arc. RUN the following program: 10 REMdrawing pies 20 SCREEN 2 30 PI=4*ATN(1) 40 CIRCLE (128, 100),90,,—PI/2,—PI 50 GOTO 50 96 You should see the top left hand sector of the CIRCLE, drawn in the foreground colour. Drawing Ellipses To draw an ellipse, you must specify the aspect ratio. This is defined as: — VERTICALRADIUS ASPECT RATIO = ORIZONTAL RADIUS If the aspect ratio is greater than 1, then the ellipse will be elongated vertically. The specified radius is taken to be the vertical radius. If the aspect ratio is less than 1, then the ellipse will be elongated horizontally, and the specified radius is taken to be the horizontal radius. The next program draws a complete, black ellipse, as the start and end angles have not been specified: 10 REM drawing ellipses 20 SCREEN2 30 CIRCLE (128, 100), 90,1,,,1.5 40 GOTO40 As the aspect ratio is 1.5, the ellipse is elongated vertically, and the vertical radius is 90 pixels. Now replace line 30 with: 30 CIRCLE (128, 100), 90,,,0.5 You will see a horizontally elongated ellipse, with a horizontal radius of 90 pixels. Try aspect ratios 100 and 0.01. With these aspect ratios, the ellipses will appear as vertical and horizontal lines respectively, of length 180 pixels. The following program draws different coloured ellipses, with random aspect ratios, all positioned at the centre of the screen: 10 REMelliptical patterns 20 SCREEN 2 30 FORB=1T015 40 A=RND(-— TIME)*2 50 CIRCLE (128,96),90,B,,,A 60 NEXTB 70 GOTO 70 If you wish, you can draw part of an ellipse, by specifying the start and end angles, just as you did when drawing circles. The '~” sign prefix can also be used: 97 10 SCREEN 3 20 PI=4*ATN(1) 30 FORB=1TO15 40 A=B/10 50 SA=RND(—TIME)*2*PI 60 EA=RND(-TIME)*2’PI 70 CIRCLE (126,96),90,B,SA,EA,A 80 NEXTB 90 GOTOS90 The last program draws fan shapes, centred at (126, 96). The Start and End Angles are calculated using the RND function. Notice that the End Angle can be smaller than the Start Angle. The Aspect ratio is varied over the range 0.1 to 1.5. RUN the program in Screen 2, to compare the resolution. EXERCISES 1 Write a program to draw 10 small circles or ellioses on the screen at random. 2 Change your program so your circles all have different radii. 3 Draw acircle made up of four sectors, each sector being drawn ina different colour. 98 CHAPTER 26 THE GRAPHICS MACRO LANGUAGE The Graphics Macro Language simulates the motion of a pen on paper, and enables you to DRAW complicated pictures easily. Using the command DRAW, you may move the Graphics Cursor to any point on the screen, with the “pen” in either the “up” or “down” position. The command DRAW must be followed by a string. This string contains the Graphics Macro Commands, which are all represented by single characters. To DRAW aline in one of the four directions up, down, right or left, the following Graphics Macro Commands are used: U DRAWs Up D DRAWs Down Ie DRAWS Left R DRAWSs Right You must specify the length of the line after the command. This Jength is measured in pixels. The next program demonstrates how to use the Macro Language: 10 REM square 20 SCREEN 2 30 PSET (123,91) 40 DRAW “R10D10L10U10" 50 GOTO50 On seeing line 40, the computer executes the Macro Commands sequentially, i.e. it DRAWs a line to the right, of length 10 pixels, then down for 10 pixels, left 10, and finally up 10, thus completing the square. 99 Using the next four commands, diagonal lines may be DRAWn: E DRAWSs diagonally up and right F DRAWs diagonally down and right G DRAWSs diagonally down and left H DRAWSs diagonally up and left Again, each of these commands must be followed by a number. The number in this case, however, specifies the number of pixels moved in the x and y directions. For example, the relative co-ordinates of the last point of the line DRAWn by E10, would be (10,10), taking the first point of the line as the reference point. This program draws a hexagon: 10 REM hexagon 20 SCREEN 2 30 PSET(128,96) 40 DRAW “E20F20D30G20H20U30” 50 GOTO50 To DRAW a line to a specific co-ordinate position on the screen, the M command is used. This command must be followed by the x and y values of the co-ordinate position of the last point of the line. The given x and y values may be either absolute or relative. For example: M40,50 will move the graphics Cursor to the absolute co-ordinate position (40,50), while DRAWing. To specify relative co-ordinates, the prefix + or — is used: M+40,50 DRAWSs to a position +40 pixels along the X-axis and +50, along the Y-axis M+40,-50 DRAWS to.a position +40 pixels along the X-axis but —50, along the Y-axis Note that the following Commands are equivalent: M+0,-10 = U10 M+10,—10 = E10 M+0,10 =D10 M+10,10 =F10 M+10,0 =R10 M-—10,10 =G10 M-10,0 =L10 M—10,—10 = H10 In the last example program, the command PSET was used to position the Graphics Cursor, before DRAWing. In fact this can be done by using the B prefix before the M command. B stands for Blank. This prefix will cause the Cursor to Move without DRAWing. 100 30 DRAW 'BM128,.96" could be used within a program to position the Cursor at the centre of the screen. The B prefix may also be used with other Macro Commands you have met, which is to be expected as these commands are just special cases of the M command. lf you wish to move back to your starting position after DRAWing a line, use the prefix N. The next program DRAWS a star on the screen: 10 REM star 20 SCREEN 2 30 REM move cursor 40 DRAW “BM128,96" 50 REMdraw arms 60 DRAW “NUS50NDS50NR50NL50" 70 DRAW “NE30NF30NG30NH30” 100 GOTO100 Each time an arm is DRAWn, the Cursor moves back to the centre of the star. You should now be capable of DRAWing fairly complicated shapes. Once you have defined your shape, it is possible to change its orienta- tion on the screen by using the Angle Command, A. This command rotates the axes of the screen, in the anticlockwise direction, through 0, 90, 180 or 270 degrees, from the normal, default orientation. This in turn, effects the orientation of the direction Commands U, D,L, R, F, E, G, and H. The command A must be followed by an integer number between 0 and 3, where: An rotation AQ resets axes to the default orientation Al rotates default orientation anticlockwise by 90 degrees A2 rotates default orientation anticlockwise by 180 degrees A3 rotates default orientation anticlockwise by 270 degrees Once A is executed, the orientation of all the following direction commands in the program will be rotated by the set angle. To return to the default value, you must use AO within a DRAW Command. RUN the following program twice: 101 10 REM angle 20 SCREEN 2 30 DRAW “BM50, 1OONRSOND50" 40 DRAW °A1BM150,100NRSONDSO" 50 GOTO 50 The first time you will see: The second time: This is because the A command is still set to A1. If you wish to repeat the first result, you must reset the Angle of rotation to AO at the beginning of the string in line 30. So far all your DRAWings have been in the current foreground colour. To DRAW in a different colour, i.e. to change the foreground colour, you may use the C command. This must be followed by a valid Colour Code. The Colour Codes are the same for this command as for the COLOR Command which you met in CHAPTER 22. So: CO. Transparent C8 Mediumred C1 Black C9 Light red C2 Medium green C10 Dark yellow C3 Light Green C11 Light yellow C4 Dark blue C12 Dark green CS Light blue C13 Magenta Cé6 Dark red C14 Grey C7 Cyan C15 White 102 As with the Angle Command, once C has been set. the current foreground colour will stay this colour until reset by either a C or COLOR Command. The S Command changes the Scale, or size of your DRAWing. The number, n, following S sets the Scale factor. This number can be any integer between 0 and 255. The Scale factor (SF) is defined as: SF = n/4 Therefore, S1 will result in a SF of a 1/4. All the lengths specified by the following U, D, L, R, Commands etc. will be scaled down by 1/4. RUN the next program: 10 REMscale 20 SCREEN 2 30 DRAW "°BM50,50” 40 DRAW °R150D100L150U100" 50 DRAW "BM50,50" 60 DRAW "S1R150D100L150U100" 70 GOTO 70 You will see two rectangles, one a quarter of the size of the other. If you RUN the program again, both rectangles will be DRAWn on top of each other, at the reduced size. To reset the SF to 1, you may use either S4 or SO. Insert one of these at the beginning of the string in line 40 and RUN the program again. If you wish to DRAW the same shape more than once in a program, itis advisable to put the string containing the necessary Macro Commands into a String Variable. The next program DRAWS four diamonds on the screen: 10 REM diamonds 20 SCREEN 2 30 AS$="FAOGAMOHA0Ed0" a0 DRAW"BMSO,8”" SO DRAW As 60 DRAW"BM200,8" 70 DRAW AS 80 DRAW" BMSO, 10a" 90 DRAW AS 100 DRAW" BM200, 104" 110 DRAW AS 120 GOTO 120 The program can be considerably shortened by putting more than one command in each DRAW statement. To put a String Variable within a DRAW string, the X command must be used. The next program again 103 DRAWS four diamonds, but this time they are different sizes. The String Variable, A$, is used as a sub-string: 10 REM scaled diamonds 20 SCREEN 2 30 A$="FA40G40HA0EAO" 40 DRAW"S4BMSO,8XA$;3" SO DRAW"S3BM200,8XA$;" 60 DRAW"S2BMSO , 1L04XAS;5" 70 DRAW"BM200, 104XA$;* 100 GOTO 100 Notice that the String Variable Name is followed by a semicolon: this must not be omitted. if you wish to use a Numeric Variable within a DRAW string, perhaps to set the length of a line, you must precede the Variable Name by the “=” sign. Again, the name must be followed by a semicolon. The following program DRAWs 15 boxes and uses the variable |, to set the scale and colour of each box: 10 REM increasing boxes 20 SCREEN 2 30 DRAW'BM31,186" 40 FORX=1 to15 50 DRAW"S=X;C=X;U48R48D48L48” 60 NEXT X 70 GOTO 70 The left hand vertical sides of the boxes have been carefully positioned to lie at the last pixel position of a horizontal group of 8 pixels. This is to prevent smudging between the last, white, vertical line DRAWn and the 14 horizontal lines which define the tops of the other boxes. The smudge effect results if you try to have more than two colours in a horizontal group of 8 pixels. (See CHAPTER 21 for more details.) To see the smudge effect, replace line 30 with: 30 DRAW'BM25, 186" EXERCISES 1 UsetheS and C Commands to DRAW stars at random on the screen, each star a different colour and size. 2 DRAW ahouse. 104 CHAPTER 27 PAINTING The Command PAINT can be used to fillin any area which is completely surrounded by a border line, with a specified colour. The co-ordinates In the PAINT Command must be the co-ordinates of a point within the shape to be PAINTed. They can be either absolute or relative. . The constraints on the allowed PAINT colours are slightly different in the two Graphics Modes. Screen 2 Before using PAINT, you must first draw the outline of your shape using one of the Graphics Commands, LINE, CIRCLE or DRAW. In this screen, the PAINT colour must be the same as the colour specified in the Graphics Command. For example: 10 REM painted parallelogram 20 SCREEN 2 30 PRESET (100,100) a0 LINE -STEP(-50,50),6 5O LINE —-STEPC(100,0),6 60 LINE —-STEP(50,-50) ,6 7O LINE -STEP(-100,0),6 80 PAINT(110,110),6 90 GOTO 90 The co-ordinates (110,110), are inside the parallelogram. The outline of ared parallelogram is drawn, and then filled in with red PAINT. If you try to PAINT the parallelogram with any other colour, the PAINT will overflow and fill the screen. If the outline had been drawn in the current foreground colour, then it 105 would not have been necessary to specify a colour in the PAINT statement, because when no colour is specified in this statement the current foreground colour is used. For example. 10 REM painted circle 20 SCREEN 2 40 COLOR 10 50 CIRCLE(100,100),50 60 PAINT STEP(O,0) 70 GOTO 70 Line 40 sets the foreground colour to yellow. The outline of the CIRCLE is therefore drawn in yellow, and then PAINTed yellow. Notice that the co-ordinates in the PAINT statement are those of the centre of the CIRCLE. In Screen 2, care must be taken to avoid smudgy effects when using the PAINT Command. The following program DRAWs two triangles, backing onto each other, to form a square. One of the triangles is then PAINTed yellow, the other black: 10 REM triangles V1 20 SCREEN 2 30 COLOR 10 40 PSET( 48,50) SO DRAW"D9I6R96H96" 60 PAINT(60,70) 70 COLOR 1 80 PSETC70,60) 90 DRAW"R9I6D96H96" 100 PAINT(C70,60) 110 GOTO 90 When you RUN this program you will get a zig-zag effect. This is because in Screen 2, each horizontal group of 8 pixels cannot hold more than 2 colours. There is no problem when PAINTing the yellow triangle, the foreground is yellow and the background is blue (the default colour). When the outline of the black triangle is DRAWn, however, the back- ground colour is still blue but the new foreground colour is black. Most of the yellow triangle is unaffected but where a group of 8 pixels lies in both triangles, the foreground colour of the complete group is changed to black. This results in small sections of the yellow triangle being lost to the black triangle, hence the zig-zag effect. There is a simple remedy to this problem. RUN the next program: 106 10 REM triangles V2 20 SCREEN 2 30 COLOR 10 40 PSET(48,50) SO DRAW"DI6R9I6U96L96" 60 PAINT(60,70) 70 COLOR 1 80 PSET(48,50) 90 DRAW"R9I6D9I6H96" 100 PAINT(70,60) 110 GOTO 110 In this program, a yellow square is drawn and PAINTed. A black triangle is then DRAWn and PAINTed on top of this. As the square only contains one colour, there is no problem when DRAWing the black triangle. Screen 3 In this screen, you can PAINT a shape any colour and then PAINT a border around it: 10 REM painting with a border 20 SCREEN 3 30 LINE(20,20)—(100,100),7,B 40 PAINT(50,50),9,7 50 GOTO 50 Two numbers are given after the PAINT statement. The first number gives the colour of the PAINT — this may be any colour. The second number gives the colour of the border. The border PAINT must be the colour of the shape’s outline. In the above example, the box was Originally draw in cyan, so the border colour must also be cyan. If you try to use any other border colour, the PAINT will overflow and fill the screen. If the outline of the shape was drawn in the current foreground colour, then you must set the border PAINT colour to this foreground colour, if you are specifying a different colour PAINT for the rest of the shape. RUN the next program: 107 10 REM triangle 20 SCREEN 3 30 COLOR 15 dO LINE (20,20)-STEP(200,100) 50 LINE -STEP¢(~200,0) 60 LINE ~—(€20,20) 70 PAINT (50,80) ,10,15 80 GOTO 80 You will see a triangle drawn in white, which is then PAINTed yellow with a white border. If the border PAINT had not been set to white in the above program, the PAINT would have covered the screen. . lf you want a completely white triangle, however, just replace line 70 by: 70 PAINT (50,80) EXERCISES 1 Write a program which draws a black box on the screen, using the “BF” option in the LINE statement. Draw another black box next tot, using the ‘B” option in the LINE statement and PAINT. RUN this program on both Screens 2 and 3. 2 Write a program in Screen 2 which draws multi-coloured concentric CIRCLEs, each CIRCLE being filled with a different colour. Do not forget, to prevent smudgy effects start with the largest CIRCLE first. 108 CHAPTER 28 THE MUSIC MACRO LANGUAGE If you are familiar with standard music notation, the command PLAY will enable you to write, or transcribe music from a manuscript, very easily. The Music Macro Language is very similar to the Graphics Macro Language, which you have met in CHAPTER 26. The Keyword PLAY is followed by a string containing the various commands allowed in the Music Macro language. The commands A,B,C,D,E,F and G, play the corresponding musical notes. Any of these commands can be followed by a “+” or ‘—” sign to indicate whether the corresponding note is sharp or flat. The resultant sharp or flat note, however, must correspond to a black note on the piano; B+, would therefore be invalid. We have now got access to the notes making up one octave. To select the desired octave, we use the O command. The first note in the octave is C, the last, B. The Octave Command must be followed by a number between 1 and 8. The initial default octave is 04, in which the first note, 109 C, corresponds to ‘middle C” on the piano. Using O1 to 08 you have access to every note you can play on a piano. Once the octave is set, it changes the octave for all the subsequent notes inthe program. The following program PLAYs the scale of E major, over two octaves; 10 REME Major: A-G Notation 20 PLAY “EF +G+ABO5C+D+EF+G+ABO6C+D+E" If you RUN this program again, the scale will start off in the 6th octave. To prevent this, add an O04 command at the beginning of the string. Thus line 20 becomes: 20 PLAY “O4EF+G+ABO5C+D+ED+G+ABO6C+D+E" Each note within the 8 available octaves can also be PLAYED using the N command. This command must be followed by anumber between 0 and 96 inclusive. As there are 96 notes in 8 octaves, the difference between N52 and N53 for example, is one semitone. In fact, the 8 octaves available with the N command are shifted up one semitone from those available with the O command. This is because NO, which would correspond to the C in 01, is silent, and can be used as a rest. The lowest note available, using this notation, corresponds to C + in Ol. The highest note with the O command is B in O8. N96, however, is one semitone higher and plays the next C up from this. Note that “middle C's at N36. G A N91 N93 08 The next program plays the same scale of E Major, but this time using the N Command: 110 10 REME Major: N Notation 20 PLAY“N40N42N44N45N47N49N5 1" 30 PLAY ’N52N54N56N57N59N61 N63N64” If you can read music, you will undoubtedly find the A-G notation easier to use. So far all of our notes have been of the same duration or Length. Tc change the duration of the following notes, use the Command L. This command must be followed by a number, which can take any value between 1 and 64. The most useful values are listed below; Ln Note Duration L1 plays a whole note or semibreve L2 plays a half-note or minim L4 plays a quarter note or crotchet L8 plays an eighth note or quaver L16 plays a 16th note or semiquaver 132 plays a 32th note or demisemiquaver You can also use L3 etc. L3 would produce a note the third of the Length of a whole note, and can be used for PLAYing triplets. If you do not set the Length of your note, it is automatically played as a crotchet, |.e. L4 is the default value. To change the scale of E major, so that it is PLAYed in quavers, you will need to insert L8 at the beginning of the PLAY string. The L command affects all the following notes in the program. The next program PLAYs E major in quavers, using first the A-G notation, and then the N notation. Notice that the note Length only needs to be set once in the program: 10 REME Major: Quavers 20 PLAY“L8EF+G+ABO5C+D+E" 30 PLAY“N40N42N44N45N47N49N51N52" To set the length of a single note, you may leave out the letter L, and put the number, which specifies the length of the note, after the note to be played: A+16 is equivalent to L16A+. Note that you cannot do this with the N notation. To specify a Rest, the Command R is used. The default value of this command is R4, which is equivalent to a crotchet Rest. To change the 111 length of the Rest, follow the R command by a number between 1 and 64. This number sets the duration of the rest. So: Rn Rest Duration R41 Rests for a whole note R3 Rests for a third of a note R2 Rests for a half-note R4 Rests for a quarter note R8 Rests for an eighth of a note R16 Rests for a 16th of anote R32 Rests for a 32nd of a note R64 Rests for a 64th of anote If you wish to PLAY a*dotted" note or rest, follow the command for that note with a dot or full stop. Each dot after the note results in the duration of the note increasing by half its original length. For example: 10 REM dotted notes 20 PLAY'AR64A.R64A..” This PLAYs the note A as acrotchet, followed by a dotted crotchet. and finally a dotted dotted crotchet! The Rests are included so you can hear the notes individually. Without these Rests, you would just hear the note A PLAYed once. The Tempo of a piece of music can be set using the T Command. The number which follows the command can take any integer value betwee 32 and 255, and sets the number of crotchets PLAYed in a minute. Yo can therefore set the Tempo of a piece of music from a slow 32 crotchet per minute, to an invigorating 255! The default Tempo is T120. In the last example of the E Major scale, we changed all the notes to quavers, so the scale was PLAYed twice as quickly as in the previous example. An alternative way to speed the scale up would be to set the Tempo to 7240. The notes are now still PLAYed as crotchets, but each crotchet is PLAYed twice as quickly as before: nc D 10 REME Major: Faster Tempo 20 PLAY“T240EF+G+ABO5C +D+E" 30 PLAY“N40N42N44N45N47N49N51N52” The Tempo only needs to be set once in the program. The Command V, sets the volume. The following number must lie between O and 15 inclusive. VO is very soft or pianissimo; V15 is very loud or fortissimo. 112 The default value is V8. Like the other Commands, O, L and T, once this command has been set it will remain at the set value until you either reset it, or turn the computer off. If a phrase is repeated often in a piece of music, you can assign the string, containing the phrase, to a String Variable. When using the String Variable in a PLAY statement, you must precede the Variable Name by the X Command and follow it with a semicolon. if you use a Numeric Variable in a PLAY string. this must be preceded by the “=" sign and followed by a semicolon. Two String Variables, A$ and B$, are used in the next example program. The Tempois INPUT into the Numeric Variable TEMPO, so you can see the effect of altering this. Try TEMPOs of 120 and 240: 10 REM Drink to Me Only: Melody 20 INPUT “Tempo”; TEMPO 30 A%="VBAREGAR64GAB- 2R64B-O05COd4B-AGA" a0 BS="B-OSCOdFB-A2GF2-F2."* SO GOSUB 1000 60 GOSUB 1000 70 PLAY*V9IOSCREACOdGADSCF 2CREA4COdAD5CREAC2" 80 PLAY”"CD2CR64CO4B-AR64A2.G2." 90 GOSUB 1000 100 END 990 REM subroutine 1000 PLAY" T=TEMPO; XA$;"* 1010 PLAY"XB$;" 1020 RETURN When typing this program in, do be careful to type the letter O, for the Octave Commands, and not the number 0. Also, take care with the semicolons and the dots. Notice the length of the notes are all set individually. The “R64’'s are used to separate the sounds when the same note is played twice. When transcribing this tune, each phrase of music was put into one PLAY statement, to make it easier to check for errors. This can, however, result in rather long program lines; don't forget you can only have up to 255 characters in a line. The MSX computer can produce up to three sounds simultaneously. We can therefore add two more voices to our tune. The next program shows you how to do this: 113 10 REM Drink to Me Only: 3 Part Harmony 20 INPUT “Tempo"; TEMPO 30 A1S$="V9IAR6dAR6d4 AB-—2R64B-—O05COd B-AGA" 40 A2S="V7FEDCDEFEFEF” SO A3$="V703F2-G2-AGFB-A" 60 B1S$="B-OS5SCO4F B—A2GF2.F2." 70 B2$="EC2DRCO3B-RO4ACDCDC" 80 B3$="GA02BN3C2.F2-C2." 90 GOSUB 1000 100 GOSUB 1000 110 PLAY"V100SCR64COdAUCSCF 2CREACO4ADSCREAC2", "VSEF 2GF2GF2EGF", "VSB—F 2ED2EF 2GA2" 120 PLAY"CD2CR64COd4B-AR64A2.G2-", “E-DEFR64F 2R64FR64F2.E2-", "GB-~2AGO2GAB2-03C2.-" 130 GOSUB 1000 140 END 990 REM subroutine 1000 PLAY" T=TEMPO; XA1$5","T=TEMPOs XAZ$35", T=TEMPO; XA3$3" 1010 PLAY "XB1%3;",XB2%$3;","XB3$;" 1020 RETURN The strings associated with the three voices are listed in each of the PLAY statements. The strings must be separated from each other by commas. Notice that the Tempo, Volume etc. are set independently for each voice. Voice 1 PLAYs ata slightly higher Volume throughout. This is because this voice is PLAYing the melody. You have probably noticed that by the end of each phrase, the voices are slightly out of synchronisation. This is because the strings for each voice contain different numbers of commands; some strings have a lot more rests and octave changes than others. Each command takes a very small, but finite, execution time. You could try to even up the strings by inserting a few “R64’’s and octave changes at suitable places. (Note: An 04 command will do nothing if the string is already in this octave, but it will increase the overall execution time of the string.) So far all the notes you have PLAYed have sounded very even in tone. It is possible to change this by superimposing an envelope on top of the note; this envelope will control the quality of the note. There are several predefined envelope shapes you can choose from, see SECTION 3: SOUND. To select an envelope, use the S Command, followed by a number between 0 and 15. The period of the envelope, or modulation, is set by 114 the M command. This can be set to any value between 1 and 65535: the default value is 255. Type in the following: 10 PLAY*S10L1CDE” You will hear the 10th envelope superimposed on the three notes, C, D and E. Try changing the envelope number. Once you think you have got the idea of this, change the modulation by using the M Command. For example: 10 PLAY“M1000S10L1CDE" With the same envelope, try other values of M, i.e. M10, M6000 etc. Experiment with other values of M and S. The order of the S and M commands is unimportant, but they both must come before the notes to be PLAYed. Finally the Command BEEP. This is the simplest command to use if you just want the computer to make a noise. Type: BEEP <RETURN> On executing a BEEP, the computer resets all the Sound Variables. So, after a BEEP, all the Music Macro Commands have been reset to their default values. 115 SECTION 2 ADVANCED PROGRAMMING GUIDE CHAPTER 1 ADVANCED PROGRAM EDITING MSX BASIC has a full screen editor which allows you to edit a program on any part of the screen. This section goes into the technical details of editing, as well as the advanced editing features of MSX using the control <CTRL> key. HOW TO EDIT When writing a program, you often need to write a series of lines, or delete whole sections of your program. MSX BASIC has four very useful commands which can ease the tedium of editing; LIST, AUTO, RENUM and DELETE. LIST Once you have entered your program into your computer's memory, you'll want to look at it before you RUN it. To have a look at the program in its entirety, use the LIST command. It will display the program in a clear and logical order; the program can be then edited using the various editing features of MSX. The LIST command has the following variations which can be used to suit your needs: LIST lists the entire program LIST <line> lists the line specified LIST <line1 >-<line2> lists the program from line? to line2 LIST <line>- lists from the line specified to the end LIST -<line> lists from the begining to the line specified You will notice that if the program you are listing is fairly long, LIST will 119 scroll the program until it reaches the last program line, or the last line specified in the LIST command. There are two methods of getting the relevant section of the program displayed: A Use LIST <line1>-<line2> so that you only get the section you want. B Use LIST and when you see the part that you want, press <CTRL><STOP> and break out from the listing. Note that the <STOP> key only will temporarily halt the listing. If you press the <STOP> key again, listing will continue from where it left off. AUTO AUTO command will put you into the automatic line numbering mode, which will offer you a line number every time you press the <RETURN> key. It saves the bother of typing the line number for every line, and makes life a lot easier for the programmer. The line numbers will have a constant increment. The AUTO command has the following variations which can be used to suite your needs: AUTO will give line numbers from 10 in steps of 10. AUTO <line> will give line numbers from the specified line number in steps of 10. AUTO <line>, will give line numbers from the specified line <increment> number, insteps <increment>. AUTO ,<increment> will continue from the line number where you left off, with the specified increment. AUTO <line> will give line numbers from the specified line number with the same increment as the previous increment. Two of the most common ways of using oie AUTO command when writing or editing a program are: 1 When starting off a new program which has been already been planned in some form on paper, AUTO is used to get line numbers from 10 in increments of 10. This gives a fairly ordered program. 2 Let's say that you have lines 100, 110, 120, 130, 140, 150, 160 and you want to insert a subroutine between lines 150 and 160. First of all you use RENUM to create more space between 150 and 160. This can be done by RENUM 1000, 150, 100. The result is that you get lings 100, 110, 120, 130, 140, 1000, 1100. Now use AUTO 1010, 5 120 which will give you a series of line numbers starting from 1010, in increments of 5. The final result will look like this: 100, 110, 120, 130, 140, 1000, <1010, 1015, 1020, 1025 1100. There are some points to remember about AUTO command: 1 There are two ways to exit the AUTO mode: a) Press <CTRL><STOP> atthe same time. b) Press <CTRL><C> atthe same time. The line from which you exit will not be saved. 2. Ifthere is an existing line number within the program you are editing, the computer will give you a warning signal in the form of a star sign *" after the line number. If you do not want to ruin that line then press the return key, but if you are replacing that line then ignore the warning ‘*" and type in what you want. RENUM Having typed in your program, invariably you will find that the line numbers are not in fixed increments. It looks messy, and if the program has to be presented to any third party, it will certainly give them a bad impression. The cure is simple: just renumber the whole program using the RENUM command. RENUM, just by itself, will renumber from 10 in increments of 10. This is the most common form used but you can renumber from a specified line number, in various increments. RENUM will automatically change the line numbers associated with GOTO, GOSUB, RESTORE, THEN, ELSE, ON GOTO, ON GOSUB. The RENUM command has the following variations which can be used to sult your needs: RENUM Renumbers from line number 10 = in increments of 10. RENUM <iline1> Renumbers from the new line number 1 in increments of 10. RENUM <line1>,<line2> Renumbers from the old line number 2 starting with the new line number 1 in increments of 10. RENUM <line1>,<line2> Renumbers starting with the old number 2 ,<increment> from new number 1, with the given increment. RENUM ,<line2>, Renumbers from the old line number 2, from <increment> the new line number 10 in given increment. RENUM ,<line2> Renumbers from the old line number 2, from the new line number 10, inincrements of 10. 121 RENUM ,,<increment> Renumbers from the new line number 10, with the given increment. RENUM <line1>,, <increment> Renumbers starting with the new line number 1 from the old line number 10 with the given increment. RENUNM is also used to create enough line number space so that you can insert a section in a program, (such as when used with AUTO, as shown above). DELETE Sometimes you'll find a section of a program to be total garbage. If the part you want to delete is a single line, then all you have to do is to type in the line number, then press <RETURN>. However, if you want to delete a whole series of lines, then use the DELETE command. It is simple and effective. It is also permanent, so once you get rid of the lines there is no way of recovery — except to type them in again! The DELETE command has the following variations which can be used to suit your needs: DELETE <line> deletes the line specified. DELETE <line1>-<line2> deletes from line1 to line2 inclusive. DELETE -<line2> deletes up to line2 inclusive. CONTROL KEYS AND SPECIAL FUNCTION KEYS MSX has a number of special function keys and control keys which are used mainly in editing. Here is a table to show you what they do and how to access them. All control keys are accessed by pressing <CTRL> and the relevant key at the same time. HEX CONTROL SPECIAL FUNCTION CODE KEY KEY 02 * <CTRL><B> moves cursor left one word. 03 * <CTRL><C> stops execution when BASIC is waiting for INPUT. Returns to the command level. 05 * <CTRL><E> deletes everything to the right of the cursor including the character underneath the cursor. 06 * <CTRL><F> moves cursor to the right one word. 07 * <CTRL><G> beep. 122 08 09 OA *mark indicates that it will disable INSERT mode if it is ON. * <CTRL><H> <CTRL> <I> <CTRL><J> <CTRL><K> <CTRL><L> <CTRL><M> <CTRL><N> <CTRL><R> <CTRL><U> <CTRL><X> <CTRL><<> <CTRL><Y> <CTRL><>> <CTRL><* > <CTRL><_> BS TAB HOME CLS RETURN INS SELECT ESC ~ —_— t + <CTRL><DEL> DEL back spaces and deletes a character to the left of the cursor, and then pulls the right hand side of the current line to the left by one. TAB. Moves to the next TAB position. TAB is set at 8 character intervals. It will place blanks from where it has moved. line feed. Physically moves the cursor to the beginning of the next line. moves cursor to the top left of the screen. clears the screen and moves the cursor to the home position. Carriage return. Inputs the current line. moves the cursor to the end of the current line. goes into the INSERT mode. The cursor becomes half the size and allows you to insert characters to the current cursor position without deleting anything. deletes the entire current line. ignored by current version of MSX. ignored by current version of MSX. moves cursor to the right by one. moves cursor to the left by one. moves cursor up by one. moves cursor down by one. deletes the character at the cursor position and pulls the right hand side of the line by one character. 123 CHAPTER 2 CONSTANTS AND VARIABLES CONSTANTS Constants are numbers or strings which do not vary, i.e. 100 and “SPEED” are constants. There are two types of constants — strings and numerics. String Constants A string constant is a sequence of characters up to 255 long. They can be any characters, graphics or control codes which are used in MSX BASIC. They must be enclosed in double quotation marks, e.g. “LUKE SKYWALKER” “WHAT IS [T?” “TOM, DICK, AND HARRY” “$ 100,000,000" Numeric Constants These are positive or negative numbers, and there are six types. 1 integer constants Whole numbers between -—32768 and 32767 or 14491111111111111 to 0000000G00000000 in 16 bit binary. They do not contain decimal points. 2 Fixed point constants Positive or negative real numbers which sometimes contain a decimal point, e.g. 657.188 124 3 Floating point Positive or negative real numbers in constants exponential form. Formats: Single precision <integer>E<integer> —56E10 <fix-point>E<integer> 7.865E5 Double precision <integer>D<integer> — 1896243232662D50 <fix-point>D<integer> 7.8827 726266265D —5 The accuracy range is between 1E—64 to 1E64. 4 Hexadecimal Hexadecimal numbers denoted by prefix constants &H. Hex is a number system with base 16, i.e. ituses O, 1, 2,3,4,5,6,7,8,9,A,B,C,D, Ee Fe Examples: &HFF (1 byte long) &H8D1A (2 bytes long) 5 Octal constants Octal numbers denoted by prefix &O. Octal is anumber system with base 8, 1.e. ituses 0, 1,2,3,4,5,6, 7. Example: &06543 6 Binary constants Binary numbers denoted by prefix &B. Binary is anumber system with base 2, i.e. it uses 0, 1 only. Example: &B01010101 Single Precision and Double Precision One of the major features of MSX BASIC is that it has 14 digit accuracy double precision BCD arithmetic functions, which most 8 bit computers do not have. Double precision is usually offered in 16 and 31 bit systems but Microsoft has rewritten its 4.5 MS-BASIC to cater for accurate computation. You can either have single precision or double precision numeric constants but if you do not specify which, the computer will assume double precision. Accuracy: Single precision —6 digits Double precision — 14 digits Single Precision constants have the following characteristics: 1 Exponential form E 1.65E-2 2 Trailing exclamation mark. 235.77! 125 Double Precision constants have the following characteristics: 1 Any digit or number without an an exclamation mark or exponential. 1878932.2 156 2 Exponential using D. 562.765333D-06 3 Trailing # sign 387 .001# VARIABLES Variables are names to which you can assign values. These values of a variable can be changed. They can be assigned specifically by the programmer (i.e. Pl=3.14) or calculated by the computer (i.e. P!=4*ATN(1)). Variable Names Variable names can be any length. However, the computer will only recognise the first two letters of any variable name. If two variables have the same first two letters then they are assumed to be the same, i.e. VAR1 and VAR2 are the same. If you want these to be different, you should use V1 and V2. The variable names cannot contain any reserved words, i.e. BASIC Keywords, in any part of the name. For example, LENGTH% is wrong because it has LEN in it. lf a variable name starts off with FN, then it is assumed to be a user-defined function. Type Declaration Ifa variable name does not have any type declaration character trailing the variable name, then it is assumed by the computer to be a double precision numeric variable. A1=<double precision number> Special trailing characters: $ Adollar sign indicates that the variable is a string. Z$="‘string characters” ADDRESS$="21B BAKER STREET” % Apercent sign indicates an integer variable. X%=25 ZXY %= —32768 An exclamation mark indicates a single precision numeric variable. Fl=5679.7! SPRES! =4.2888E —02 # #sign indicates a double precision numeric variable. H#=3.14161718293 MAXIMUM#=5277.76525D25 126 DEF <character type> It is possible to define a character type using various DEF statements. When the type of variable is declared using one of these statements, any variable starting with that declared character will become that character type without declaring it in the variable name. This means that if you DEFSTR A, then any variable starting with A will be a string variable without having the usual dollar sign. However, that doesn’t mean that AB% will be a string variable, since the integer declaration character (i.e. %) takes precedence over DEFSTR. There are four DEF<character type> statements. These are as follows: DEFSTR statement declares that any variable name beginning with the specified range of characters is to be treated as a string variable: DEFSTRA AST="PLANET” DEFINT statement declares that any variable name beginning with the specified range of characters is to be treated as an integer. DEFSNG statement declares that any variable name beginning with the specified range of characters is to be treated as a single precision number. DEFDBL statement declares that any variable name beginning with the specified range of characters is to be treated as a double precision number. Note that the type declaration of the character by #, $, %, and ! takes precedence over these statements. ARRAY (DIM) An array is a group of variables which have acommon name. It is set up by using the DIM statement. DIM statements set aside a section of memory for the use of any array before it is used. Each element of the array has a subscript, or number label. Say, for instance, you define DIM C(3). You then get four variables with the name C(_), i.e. C(O), C(1), C(2), C(3). All array variables start with the zeroth element. This is a one dimensional array, but there is no reason why you can’t have a multi- dimensional array such as DIM A(2,2) which gives a table of 3 by 3 variables as shown below: A(0,0) A(1,0) A(2,0) A(0,1) A(1,1) A(2, 1) A(0,2) A(1,2) A(2,2) This is a two dimensional array but you can have a three dimensional array, or even ten. The maximum dimension for an array is 255 but you are likely to run out of memory if you try to create such a massive array. 127 if an array has less than 12 elements, i.e. DIM A(10), then there is no need to execute a DIM statement, as the computer automatically sets aside 11 elements of memory space. Programs such as the one shown below are perfectly legal even without the DIM statement. 10 FORI=0TO 10 20) S(I=l 30 NEXTI Itis possible to have arrays of any type as long as the type is declared. For example, DIM S$(100) will give you one hundred and one elements of a string array. When the array is initialised, all the values are assumed to be zero for numeric arrays, and null strings for string arrays. Memory Requirement Variables are stored in either DIM AREA, VARIABLE AREA or STRING AREA depending on the type (See memory map). Here is a list of the number of bytes used by these variables. Variables Integer 2 bytes Single precision 4 bytes Double precision 8 bytes String 3 plus length of content Arrays Integer 2 bytes per element Single precision 4 bytes per element Double precision 8 bytes per element String 3 plus length of content per element VARPTR To find out exactly where the data of a variable is stored in memory, you use a special function VARPTR which gives the address of a particular variable, e.g. 10 A%=100 20 PRINT VARPTR(A%) You can find out the memory location of any type of variable as long as it exists. It can be a string or even an array. String Variable Area All the contents of string variables are stored in the STRING AREA section of the computer's memory (see Memory Map). The size of this area is restricted to 200 bytes at default; therefore the maximum amount of string at any one time is limited to 200 characters. However, you can increase the size of the STRING AREA by using the CLEAR statement. For example, to increase the STRING AREA to 255 bytes, execute: CLEAR 255 128 CHAPTER 3 TYPE CONVERSION MSX BASIC can convert one type of numeric constant to another. There is a Set rule for carrying this out. 1. If a numeric constant of one type is set equal to another type, the number stored will depend entirely upon the type of variable. Example 1 10 C!=1.2345678901 The double precision number 20 PRINT C! 1.2345678901 is converted to single RUN precision when assigned to Cl, 1.23456 a single precision numeric variable. Example 2 101% =1.23E—03 The single precision number 1.23E-—03 20 PRINT 1% is rounded to the nearest integer which RUN is zero when assigned to an integer 0 variable. 2 During expression evaluation, all variables and constants are converted to the most accurate precision of the variables being used. When the expression is set to equal a variable, the result of the expression is converted to the precision type specified by the variable. If the result variable type is not specified, the result remains at the same precision level as that in which the evaluation took place. 129 Example 1 10 D=1!/3# 20 PRINT D RUN 0.33333333333333 Example 2 10 D=1!/3! 20 PRINT D RUN 333333 Example 3 10 D!—1/3 20 PRINT D! RUN 333333 1! is single precision while 3# is a double. The arithmetic was performed in double precision and returned to a double precision variable D. Both constants are single precision, so the arithmetic was carried out in single precision, but D is a double precision variable so the result is converted to double precision, but lacks the double precision accuracy. The arithmetic was carried out in double precision but the result is converted to single precision, when returned to D!, by truncation. Logical operators convert their constants and variables to 16 bit integer numbers before the operation is carried out. The result is an integer and the numbers involved must be within the integer range or an “Overflow Error” will result. Example 10 D%=156.65 AND 654 20 PRINT D% RUN 140 The numbers are changed to 156 and 654. Result 140 is returned to D% which is an integer variable. 4 When a floating point or a fixed point value is converted to an integer, the fractional part is truncated. Example 1 10 A%=123.456 20 PRINT A% RUN 123 The fixed point value 123.456 is truncated to 123 when assigned to an integer variable A%. 130 Example 2 10 A%=1.0097554D02 The floating point number 20 PRINT A% 1.0097554D02 is rounded to 100 when RUN converted to A%, an integer variable. 100 5 When a single precision number is converted to a double precision number, only 6 digits can be supplied to the double precision result. Example 10 A#=1.33333 A# is adouble precision variable but that 20 PRINT A# doesn't make the number assigned to it any RUN more accurate —- the result is still 1.33333. 1.33333 TYPE CONVERSION USING MSX BASIC FUNCTIONS CDBL, CINT, CSNG, VAR, STR$ Although MSX does type conversion automatically when it is evaluating expressions, you may program it to do type conversion, and, when necessary, using one of the type conversion functions. CDBL, CINT, and CSNG follow the same rules as above but VAR and STR$ are string to numeric and numeric to string respectively, and they follow slightly different rules. CINT CINT function canverts single and double precision numbers, variables, and expressions to integer numbers. The argument must be strictly within the integer range of —32768 to 32767. If not, then an Overflow error will result. CINT(<num-const>) CINT(<num-var>) CINT(<num-exp>) Example PRINT CINT(1234.56789) 1234 CSNG CSNG function converts the argument into a single precision number. CSNG rounds off to the nearest 6th decimal place. If an integer is converted, the accuracy will still be the same. 131 CSNG(<num-const>) CSNG(<num-var>) CSNG(<num-exp>) Example PRINT COS(0.7656) .72096670541357 PRINT CSNG(COS(0.7656)) 720967 CDBL CDBL function converts integer and single precision numbers, variables and expressions into double precision numbers. When a single precision number is converted, only 6 digits will be supplied to the double precision result. CDBL(<num-const>) CDBL(<num-var>) CDBL(<num-exp>) Example PRINT CDBL(5.666!) 5.666 VAL and STR$ These functions are used for converting strings to numbers and vice versa. STR$(<numberic>) STR$ converts the numeric argument into a string. VAL(<string>) VAL returns the value of a string whch contains a number. If a string is a string representation of a number then it can be VALed. However, it can’t work out an expression within a string. VAL can strip off spaces, tabs, and line feed in front of a number in a string, but can’t cope with other characters. VAL can recognise plus and minus signs. See SECTION 1 CHAPTER 15 for examples. 132 CHAPTER 4 EXPRESSIONS AND OPERATORS An expression may be a string, anumeric constant, or a variable, or any kind of combination of these which returns a single value, as long as it is legal. There are four types of mathematical or logical operators. 1 Arithmetic 2 Relational 3 Boolean (Logical) 4 Functional 1 and 2 are covered by this part but see the relevant sections for the others. ARITHMETIC OPERATORS The arithmetic operators in order of precedence are: 1 * — exponential A“*B 2 — negation -A 3 */ multiplication and division A*B, A/B 4 +,- addition and subtraction A+B, A-B To change the order of calculation, use brackets (). The crder of precedence is maintained within the brackets. DIV and MOD (Integer division) DIV is denoted by ¥ (yen) sign in MSX BASIC. It performs integer division. Modulus is denoted by MOD. What are DIV and MOD? Well, here is an example to illustrate their use. In usual division if you divide 10 by 4, you get 2.5. 133 Similarly, Example 13/5=2.6 A*B=C In integer division, however, you don’t get the decimal point but A*B is carried out first, then the result of A*B is compared with C. instead you get the whole number part of the division, 2, and a remainder, 3. In MSX BASIC, if you carry out 13 DIV 5 you get 13¥%5=2........ whole number part If you 13 MOD 5 you get, 13 MOD 5=3........ remainder Integer Division and Modulus arithmetic are carried out with 16 bit integers. Before the calculation takes place, any non-integers are changed to integers by truncation, i.e. all fractional parts are discarded. 16.78¥5.89 = 3 (is the same as 165 = 3) 16.78 MOD 5.89 = 1 (is the same as 16 MOD 5 = 1) MOD follows DIV (¥) in order of precedence. Example: Showing order of precedence of MOD and DIV: 1580 MOD 789 DIV 10 1 789 DIV 10 78 2 1580 MOD 78 20 i.e, 1580 MOD 789 ¥ 10 = 20 RELATIONAL OPERATORS Relational operators compare two values and return a true (—1) or false (0) answer. They are usually used in IF THEN conditions, but can be incorporated in to an expression. Relational operators are covered extensively in the IF THEN ELSE section (Section 2 Chapter 7). OPERATOR RELATION EXAMPLE = Equal to A=B <> Not equal to A<>B < Less than A<B > Greater than A>B <= Less than or equal to A<=B >= Greater thanorequalto A>=B True returns — 1 False returns 0 When there is a mixture of relational operators and arithmetic operators, arithmetic operators take precedence. 134 135 DECIMAL BINARY IN 8 BIT 0 0 00000000 1 1 00000001 2 10 00000010 3 11 0000001 1 CHAPTER 5 4 100 00000100 5 101 00000101 6 110 00000110 7 114 00000111 SURVEY OF NUMBER 8 1000 00001000 9 1001 00001001 SYSTEMS USED IN MSX 10 1010 00001010 11 1011 00001011 12 1100 00001 100 13 1101 00001101 14 1110 00001110 15 1114 00001111 Since most human beings on this planet have ten fingers on their hands, 16 10000 00010000 they use the decimal number system. Unlike homo sapiens, computers, which are basically a huge array of switches, can have only two ‘fingers’ Binary to decimal and decimal to binary conversion = ON or OFF (or in other words, 0 or 1). So computers work in binary, Pee internally, all the time. For the benefit of users, the MSX BASIC will How to convert 8 bit binary numbers to decimal equivalent. display numbers in decimal but naturally it can also handle binary BIT NUMBER CORRESPONDING VALUE numbers in users’ programs. Apart from binary, MSX can use Octal and iA a 128 Hexadecimal. All hex, binary and octal numbers are integers and must 6 2°6 64 be within the integer range, that is, between —32768 and 32767. 5 OES 32 The arithmetic functions are all calculated in Binary Coded Decimal 4 2*4 16 system (BCD) which we will explain later. 3 28 8 2 2°2 4 BINARY ul ae ; 7 Example: convert 10110001 to decimal. Definition: Number system with base 2. Uses Q and 1 to represent all 1011000 1 numbers. | eee 1x2°O0= 1 MSX BASIC: &B<binary> represents a binary number. Ox2*1= 0 BIN$ decimal to binary Ox2*2= 0 Ox2*3= 0 on . . . 1x2°4= 16 Each digit ofa binary number is called a ‘bit’. 8 bits of binary are called i AD Be a ‘byte’. One byte is the size of one address of memory in the MSX —- Ox2*6= 0 system. One byte of binary can have any value between 00000000 to 1x2*7= 128 11111111, or 0 to 255 in decimal. 1/7 136 137 binary decimal &B1011001 = 177 How to convert decimal number to binary. Example: convert 53 to binary BIT NUMBER 7 2°7 128 ¢) 6 2°6 64 0 oS 2°5 32 53-32=21 1 4 2%°4 16 21-16=5 1 3 2%°3 8 0 2 2%2 4 5-4=1 1 1 274 2 0 QO 2°0 1 1—1=0 1 Collect all binary bits and you get 53=&B00110101 In BASIC: How to convert an 8 bit binary number to its decimal equivalent. Example 10 X%=&B00101011 20 PRINT X% RUN 43 How to convert a decimal number to its binary equivalent. Example 10 X%=171 20 PRINT BINS(X%) RUN 10101011 OCTAL Definition: A number system with base 8. This uses numbers between 0 and 7 to represent all numbers. MSX BASIC: &0<octal> represents an octal number. OCT$ decimal to octal 138 DECIMAL OCTAL 0 0 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 10 9 11 10 12 11 13 12 14 13 15 14 16 15 17 16 20 Octal to decimal and decimal to octal conversion How to convert an octal number to decimal equivalent. <decimal=> = <1st digit» *8*O+<2nd digit> *8 41+<3rd digit>*8 .2+<4th digit>*8*%3....... Example Octal &05436 to decimal 6*8 * 0+3°8 * 1+4*8 * 24+5*8 * 3=2846 &05436=2846 How to convert decimal number to octal. Example Decimal 6588 to octal DIGIT DECIMAL INTEGER DIV OCTAL MODULUS 4 8%4 4096 6588 DIV 4096 6588 MOD 4096 = 2492 3 8*3 512 2492DIV512 2492 MOD 512 = 444 2 8%2 64 444 DIV64 444 MOD64 =60 1-8" 8 60 DIV8 60 MOD8 =4 0 8°0 1 4 DIV 1 So you get &014674 = 6588 (decimal) In BASIC How to convert an octal number to its decimal equivalent. 139 foi dh Wd ARNO Example 10 X%=8&06517 20 PRINT X% RUN 3407 How to convert a decimal number to its octal equivalent. Example 10 X%=171 20 PRINT OCT$(X%) RUN 253 HEXADECIMAL Definition: Anumber system with base 16. It uses numbers between 0 and 9 and additionally A, B, C, D, E, and F to represent all hex numbers. MSX BASIC: &H<hex> represents a hexadecimal number HEX$ decimal to hexadecimal. DECIMAL HEXADECIMAL 0 0 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 10 A 11 B 12 C 13 D 14 E 15 F 16 10 Hex to decimal and decimal to hex conversion How to convert a hex’ number to its decimal equivalent. <decimal> = <1st digit>*16 *0+<2nd digit>*16 * 1+<3rd digit>*16 *2+<4th digit>*16 *3....... 140 Example Hex &HFF to decimal (&HF)*16 * 0+ (&HF)*16 * 1=(15)*1 +(15)"16=255 &HFF=255 How to convert decimal number to its hexadecimal equivalent. Example: Decimal 7752 to hex DIGIT DECIMAL INTEGER DIV HEX MODULUS 3 16%3 4096 7752 DIV 4096 1 7752 MOD 4096 = 3656 2 16% 256 3656 DIV 256 14(E) 3656MOD 256 = 72 1 16% 1 16 72 DIV1I6 4 72 MOD16 =8 O 16% 1 8 DIV 1 8 So you get &H1E48 = 7752 (decimal) In BASIC How to convert a hexadecimal number to its decimal equivalent. Example 10 X%=&HEFA3 20 PRINT X% RUN 4189 How to convert a decimal number to its hexadecimal equivalent. Example 10 X%=171 20 PRINT HEXS$(X%) RUN AB NOTES 1 Binary is often used when designing sprites because is is easy to see the on and off of the bit pattern. 2 Hex is used in describing memory locations and in machine codes. 141 Summary of Decimal, Binary, Octal and Hexadecimal. DECIMAL BINARY OCTAL HEXADECIMAL 0 0 0 0 0 1 1 1 2 10 2 2 3 11 3 3 4 100 4 4 5 101 5 5 6 110 6 6 7 111 7 7 8 1000 10 8 9 1001 11 9 10 1010 12 A 11 1014 13 B 12 1100 14 C 13 1101 15 D 14 1110 16 E 15 1114 ie F 16 10000 20 10 BINARY CODED DECIMAL (BCD) SYSTEM AND MSX MSX features up to 14 digit accuracy double precision BCD arithmetic functions. This means arithmetic operations no longer generate strange rounding errors which are invariably associated with many 8 bit computer systems. All arithmetic operations and functions are calculated with this accuracy unless specified otherwise by the user program. Why use binary coded decimal? What is its advantage and why is it more accurate? To start with, let's have a look at how numbers are worked out in the binary system which is most suitable for computers. in binary, if you try to express anumber smaller than 1 (i.e. one with a decimal point) you sometimes get errors. There are certain numbers which can be expressed in decimal but not in binary. Say we want to express 0.6525 in binary: 0.625 = 1/4+1/8 = 1/(2 * 2)+1/(2 * 2) 0.625 = 2 *(—2)+2 * (-3) So you can convert 0.625 to binary without any trouble. However when it comes to a number like 0.1, you can't do this. There is no way of expressing 0.1 in binary exactly. The following shows why: 0.1 ~ 1/164+1/382+1/256+ 1/512 + 1/4096+4 1/8192+ 1/65536 .... ~ 0.09999.... 142 Ot = 40a 12” Sei *8) 41/2 * 9/41/02 °'13)41/ (2° 16)+ ...... ~ 0.099999... You can't get a 0.1 in binary; instead you obtain a series. Instead of turning a decimal into binary completely, MSX uses binary coded decimal. In the binary coded system, a 4 bit binary represents a digit in decimal. This means that 0001 corresponds to 1, and 0010 corresponds to 2, and so on up to 1001 which is 9. 4 bit binaries between 11 and 15 i.e. 1010, 1011, 1100, 1101, 1110, and 1111, are not used in the binary coded decimal! system. Instead 10 (decimal) carries to the next 4 bit — therefore 10 is 0001 0000 in BCD. 1 byte (8 bit) represents two digits in decimal O 0000 90000 10 0001 0000... 90 1001 0000 1 0000 0001 11 0001 0001 ... 91 1001 0001 2 0000 0010 12 0001 0010... 92 1001 0010 3 0000 0011 13 0001 0011... 98 1001 0011 4 0000 0100 14 0001 0100. .... 94 1001 0100 5S 0000 0101 15 0001 0101... 95 1001 0101 6 0000 0110 16 0001 0110... 96 1001 0110 7 Q000 0111 17 0001 O111 .... «97 1001 0111 8 0000 1000 18 0001 1000 .... 98 1001 1000 9 Q000 1001 19 0001 1001... 99 1001 1001 8 byte double precision numbers in BCD MSX stores numerical data in double precision unless it is programmed to otherwise. It has an accuracy of 14 digits, and is stored in 8 byte binary coded decimal system. The first byte stores the exponent, and the other bytes give the 14 digits, i.e. 2 digits per byte. To see how the numerical data Is stored, first find the memory location of that data using VARPTR function, and then peek the content and display it in hex. 10 P=3.1415926534890 20 ADDRESS=VARPTR(PI) 30 FOR|I=ADDRESS TO ADDRESS+7 40 PRINT HEX$(PEEK(1)),” 50 NEXT | 60 PRINT RUN 41 31 41 5926 53 58 97 OK Let's analyse the above; 143 41 31 41 59 26 53 58 97 a ee 7 Byte BCD for 14 digits 0 1000001 7 Bit for exponent. then 65 subtract 65 to get E00 positive or negative indicator +=0 —=1 The header byte gives the positive or negative sign and the exponent. To get the exponential value, first ignore the 7 bit and calculate the digital value for the rest, i.e. 6th to Oth bits; then subtract 65. This gives a number with size between 1E-64 and 9.99999999999999E€é2. Advantages and Disadvantages of the Binary Coded Decimal System Advantages 1 BCD gives no rounding off errors, and can give numbers which are not possible in binary such as 0.1. 2 |t is extremely easy to display a BCD number because there is no need to convert binary to decimal. This speeds up the process of printing to the screen. Disadvantage 1 Because it is not in ordinary binary, the computer takes more time when it comes to the arithmetic computation. However, the Z 80 CPU has a special intruction DAA, Decimal Adjust Accumulator, to make it easier to process. The DAA converts the accumulator content into packed BCD following add or subtract, with packed BCD operands. 144 CHAPTER 6 BOOLEAN ALGEBRA (LOGICAL OPERATORS) INTRODUCTION However small your MSX may be, it is still a computer, and a computer is basically made of millions of switches. These can be either ON or OFF, but combinations of these create a working computer. The logic of a computer, or this huge bunch of switches if you like, is governed by aset of rules which come under Boolean Algebra. It is mainly used in multiple condition testings, in IF/THEN statements, with ANDs and ORs, but you can use it in bit manipulation which is important in machine code. MSX LOGICAL OPERATORS MSX provides proper Boolean (logical) Operators, unlike some 8 bit micros which seem to miss these out. They are very important in advanced programming. Here is a list of them in order of precedence. Computer Jargon NOT Complement AND logical AND OR logical OR XOR Exclusive OR EQV Equivalent iMP Implication AnhwWN- NOT, AND and OR are the important operators since they are the most commonly used logical operators; and besides the others, i.e. XOR, EQV, and IMP, can be derived from combinations of the NOT, AND and OR operators. Logical operations are always carried out on integer numbers. Any 145 argument that is not an integer is converted to 16 bit, signed, two's complement integer. We shall now see how they work, and what they can be used for step by step, starting with NOT. NOT NOT is a denial or negation. It gives true for false, and false for true, i.e. 1 for O and 0 for 1. In other words, it negates one argument. Here is the truth table for NOT. NOT X NOT X 0 1 1 e) NOT X 100 0000000001 100100 —101 1111111110011011 AND The Boolean algebraic operations of AND can be listed in a truth table as shown below: AND xX oY X AND Y Oo 0 0 1 0 0 fe) 1 0 1 1 1 Therefore, say 100 AND 50 can be calculated as follows: X 100 0000000001 100100 AND Y 50 0000000000110010 32 0000000000100000 EXAMPLE: USE OF AND Logical AND can be used for bit testing a particular number. For example, if Y= 2 * n, where nis the test bit between O and 7, then X AND Y = Y ifthe test bitnis 1 and X ANDY = Oif the test bit n is O. BIT TEST the 5th bit (n=5) for X=117. Y=2 *5=32 X 117 0000000001 110191 AND Y 32 0000000000010000 32 0000000000010000 Therefore the 5th bit of number 117 is 1, i.e. true. OR When either X, Y or both are true, OR will return true. Logical OR is sometimes called logical sum. 146 The truth table for OR is: OR Y X ORY 7a0+-+0x =— = O © 0) 1 1 1 Example 34 OR 67 X 34 0000000000100010 OR Y 67 000000000100001 1 99 0000000001 10001 1 EXAMPLE: USE OF LOGICAL OR Logical OR can be used for converting an upper case character to a lower case character, since the difference between the ASCII code of upper case and lower case is constant, i.e. 32 (or 2 * 5). Example: change character A (ASCII code 65) to a (ASCII code 97). A 65 0000000001000001 OR 2.5 32 0000000000100000 a 97 0000000001 1000001 This method is quite flexible, because if you tried to convert a lower case letter to a lower case letter, nothing happens. Try this for yourself. XOR XOR, Exclusive OR returns true (1) when X and Y are different. XOR X YX XOR Y oO 860 0 1 0) 1 6) 1 1 1 1 6) EXAMPLE: 100 XOR 50 X 100 0000000001 100100 XOR Y 50 00000000001 10010 86 0000000001010110 XOR can be calculated using NOT, AND and OR. Here is how you do it. X XOR Y =(X AND (NOT Y)) OR ((NOT X) AND Y) So, in plain English, XOR is Either X AND NOT Y ORNOT XANDY. X XOR Y =(X AND (NOT Y)) OR ((NOT X) AND Y) 147 Proof x Y NOT X NOT Y 0 ) 1 1 1 ) ¢) 1 0 1 1 0) 1 1 ) ) Xx (NOT Y) =XAND (NOT Y) 0 1 6) 1 1 1 0 0 0) 1 ) 6) (NOT X) Y (NOT X) AND Y 1 0) 0 ) 6) O 1 | 1 0 1 6) (X AND (NOT Y)) ((NOT X) AND Y) (X AND (NOT Y)) OR ((NOT X) AND Y) ) 6) @) 1 ) 1 ) 1 1 0 ¢) @) Therefore X XOR Y=(X AND (NOT Y)) OR ((NOT X)) AND Y) EXAMPLE: USE OF XOR XOR has the unique property that if you XOR a value with another value twice, you get the original value back. X XOR Y XOR Y =X Let's try 100 XOR 50 XOR 50 as an example: 100 XOR 50 X 100 0000000001 100100 XOR Y 50 00000000001 10010 86 0000000001010110 (100 XOR 50) XOR 50 X 86 0000000001010110 XORY 50 00000000001 10010 100 0000000001 100100 Therefore 100 XOR 50 XOR 50 = 100 148 EQV This is the logical equivalence function. EQV returns true whenever X and Y are equal, and false when they are different. EQV X Y XEQVY 0 0 1 1 0 6) Oo 1 ) 1 1 1 Example 51 EQV 75 X 51 00000000001 10011 EQV Y 74 0000000001001010 —122 1111111110000110 The following relation is true. X EQV Y=(X AND Y) OR ((NOT X) AND (NOT Y)) Proof X Y NOT X NOT Y ) 0) 1 1 1 0) 0) 1 ) 1 1 0) 1 1 0) 0) X Y (XANDY) 0 6) 0) 1 @) 0) fe) 1 0) 1 | 1 (NOT X} (NOT Y) ((NOT X) AND (NOT Y)) 1 1 1 ) 1 0) 1 6) 0) ¢) 0) 0) (X AND Y) ((NOT X) AND (NOT Y)) (X AND Y) OR ((NOT X) AND (NOT Y)) ) 1 1 ¢) 0) @) 0 0) ) | 0 1 Therefore X EQV Y=(X AND Y) OR ((NOT X) AND (NOT Y)) 149 IMP IMP Truth table. IMP x Y X IMP Y 6) 0) 1 1 0) 0 6) 1 1 a] 1 1 100 IMP 50 can be calculated as follows: X 100 0000000001 100100 IMP Y 50 00000000001 10010 —69 1111111110111011 The following relation is true. X IMP Y = (NOT X)ORY Proof x Y NOT X ) 0) 1 1 0 ) ) 1 1 1 1 ) (NOT X) Y (NOT X) OR Y 1 ) 1 ) ) ) 1 1 1 0) 1 1 Therefore X IMP Y = (NOT X) ORY Well, that is about it as far as MSX BASIC goes but Boolean algebra does not stop there. There are many other useful functions such as NOR and NAND which are not included in MSX but can be simulated using NOT, AND, and OR. Remember that these three are the basics of Boolean algebra, and they can be combined to give the others. Take a look at the table below. Two binary variables, X and Y, each can have a value of either 1 or 0. They can have four different combina- tions of O and 1, since 2* 2=4. There are 16 combinations of these combinations (or functions) since 2 * 2 * 2=16. What you get is a list of the all possible combinations or truth tables. You can recognise the following: f2 is NOT {5 is AND f9 is OR f13 is XOR f11. is IMP f14 is EQV These are all in MSX BASIC but what about the others, e.g. f1 and f3 etc.? In this section we will look into non-MSX logical operators. These can't be used in MSX as they are, so an equivalent NOT AND OR expression will be given. fO Always false function whatever the combination of X and Y Is. Not used in computing. fi Always X whatever the value of Y. f2 NOT X (see NOT). {3 Always Y whatever the value of X. f4. NOT Y (see NOT). f5 XANDY (see AND). f6 XAND(NOT Y). f7 (NOT X)ANDY. f8 NOR function. Its MSX equivalents are (NOT X) AND (NOT Y) or alternatively, NOT (X ORY) ....... see De Morgan's Law. {9 XORY (see Logical OR). f10 ((NOT X) OR Y) Similar to IMP. f11. XIMP Y (see IMP). Similar to f10. Alternatively X OR (NOT Y). f12 NAND function. It is negation of AND (NAND is short for NOT AND). NOT (X AND Y) (NOT X) OR (NOT Y) X NAND Y = (NOT X) OR (NOT Y) = NOT (X AND Y). See De Morgan's law later on. f13 X XOR Y. Exclusive OR (see XOR). Alternatively (X AND (NOT Y)) OR ((NOT X)) AND Y) f14. XEQVY. Logical equivalence. X EQV Y=(X AND Y) OR ((NOT X) AND (NOT Y)) f15 Always true whatever the value of X and Y. Not used. NAND and NOR, both not included in MSX BASIC, are useful func- tions to remember. 151 Some Useful Logical Relations There are a variety of logical relationships which can be used in Boolean algebra. They can be used in shortening long winded logical expres- sions by way of minimisation. USING LOGICAL OR X ORO=X X OR X=X X OR X OR X ORX OR... =X X OR (NOT X)=1 XOR 1=1 X OR (X AND Y)=X X OR ((NOT X) AND Y)=X ORY NOORWNDY = USING LOGICAL AND 1 X AND 0=0 2 X AND X=X 3 X AND XANDXANDX...... =X 4 X AND (NOT X)=0 5 X AND 1=X 6 X AND (X AND Y)=X AND Y 7 X AND ((NOT X) AND Y)=0 Examples Try out some of the following examples on your computer and see if the above relations are really true. ANSWERS PRINT 145 ORO 145 PRINT 167 OR 167 167 PRINT 167 OR 167 OR 167 167 PRINT 133 OR (133 AND 222) 133 PRINT 111 ANDO QO PRINT 111 AND 111 111 PRINT 111 AND 111 AND 111 AND 111 111 PRINT 234 AND (NOT 234) ie) PRINT 234 AND (234 OR 54) 234 De Morgan’s Rules Apart from the above relationships, Boolaen algebra has two special tricks to reduce the length of logical expressions. 152 1 Negation of Logical OR The negation (NOT) of a logical OR is equivalent to the logical AND of negations of the variables making up the logical OR. This may be expressed as: NOT (X OR Y)=(NOT X) AND (NOT Y) 2 Negation of Logical AND The negation of a logical AND is equivalent to the logical OR of the negations of the variables comprising the logical AND. This may be expressed as: NOT (X AND Y)=(NOT X) OR (NOT Y) Laws of Rearrangement In mathematics, there are a number of laws which allow you to rearange and simplify complex expressions. They are known as the commutative, associative, and distributive laws. Boolean algebra has similar laws as follows: 1 Commutative Laws a. XORY = YORX b. XAND Y = Y AND X 2 Associative Laws a. X AND (Y AND Z) = (XK AND Y) AND Z = X ANDY AND Z b. XOR(Y ORZ) = (XORY)ORZ = XORY ORZ 3 Distributive Laws a. X AND (Y OR Z) = (XK AND Y) OR (X AND Z) b. X OR (Y AND Z) = (XK ORY) AND (X OR Z) Summary of MSX Logical Operator's Truth Table X Y AND OR XOR EQV IMP O 0) 0) 0) 0) 1 1 | 0) 0) 1 1 0) 0) O 1 O dl ] O | 1 1 1 1 O 1 1 153 CHAPTER 7 BOOLEAN II: THE IF/THEN/ELSE When you are writing a computer program, it is very likely that the computer will have to make decisions by itself while running your program. To make decisions on certain conditions, use the IF statement to first test the conditions, and then carry out certain tasks depending on the outcome. IF is always used in conjunction with THEN, and sometimes ELSE key words. It has basically two types of format: IF <conditions> THEN <statements> and IF <conditions> THEN <statements> ELSE <statements> IF is followed by the <conditions> to be tested. If the <conditions> turn out to be true, then the computer executes the <statements> after the THEN word. If the computer finds that the <conditions> are not true then, if there is no ELSE word in that line, it will carry on from the next line without executing the <statements> after the THEN. If, however, there is an ELSE word within that line, then if the <conditions> is found to be false, the <statements> after ELSE are executed. In this case the <statements> between THEN and ELSE are completely ignored. To begin with, let’s look at the <conditions> part of the IF THEN ELSE structure, <CONDITIONS > This bit follows all the rules and regulations described in the EXPRESSIONS AND OPERATORS chapter, and BOOLEAN ALGEBRA chapter. This section shows the practical side of what we said in the last two chapters. 154 Numerical Comparison The simplest condition compares two values, e.g. they can test whether one number is larger than another. They use the relationed operators, such as <,>,<>,=,<=,>=. Examples 1 15<8 returns _O (false). 2 a%=67 returns —1 (true) if a%=67 O false if a% is not 67. In an IF THEN format: 1 IF D%=100 THEN PRINT “D% IS 100” 2 IF V<=G THEN PRINT “V<=G" These are the simplest forms but you may also use arithmetic expressions on both sides of the relational operator. 1 IF F%*H%=9876 THEN PRINT “TRUE” 2 IF 2 * 8=N%—19*Y% THEN PRINT “TRUE” Also, there is no reason why you can't use BASIC functions such as COS and INSTR, etc. In fact, you can even incorporate a user defined function FN, too. 1 IF INSTR(AS,B$)=5 THEN PRINT BS 2 IF COS(0.6242)>= TAN(D%) THEN PRINT “TRUE” 3 IF FNA(D% *I%— 100) =&HFF THEN PRINT “FF” Numerical Condition without Comparison It is not strictly necessary to have a relation as a <condition>. In fact, it can be a single numerical variable or just a simple expression Take a look at this example: IF X% THEN PRINT “X%=TRUE™ In this case the computer takes the condition to be true when the variable X% is non-zero, and false when X%=0. In general: IF <expression> THEN <statement> <expression>=TRUE when <expression= returns non-zero. <expression>=FALSE when <expression> returns 0. You can use this in a variety of situations. Here are some examples to illustrate this point. 1 IF A<>0 THEN PRINT “A NOT 0” ... canbe IF A THEN PRINT “A NOT 0° 155 2 Tocheck if there is a certain character within a string. IF INSTR(S$,"a") THEN PRINT “S$ CONTAINS CHARACTER ‘ah String Comparison You may compare two strings using relational operators in much the same way. The comparison is made by taking one character at a time from each string and comparing the ASCII codes. If the ASCII codes differ, the lower case code number precedes the higher. For which character corresponds to which ASCII code, see the ASCII code table. If during string comparison the end of one string is reached, the shorter string Is said to be smaller. String comparison can be used to alphabetize strings. In the following example, all the conditions are true. “abc’= “abc” “ABC"<"“abc” “ABC"<"“ABD” “ab"<“abc” °4234"<'42345” OBRWNM — Multiple Condition Testing using Boolean Operators It is possible to test multiple conditions in one go, using Boolean logical operators. There are 6 logical operators in MSX BASIC but only 3 of them, NOT, AND, and OR, are relevant. They follow all the rules governing Boolean algebra, such as De Morgan’s Laws. 1. Simple use of AND IF <condition-- 1> AND <condition-2>> THEN <statements > In this case, only if both conditions are true, will the computer execute the <statement> after THEN, e.g. IF X=0 AND Y$="YES” THEN PRINT “FINE” 2. Simple use of OR IF <condition—1> OR <condition—2> THEN <statements > In this case, if either one or both conditions are true, then the computer will execute the <statements> after THEN, e.g. IF X=0 OR Y=0 THEN PRINT “ONE OF THEM IS ZERO” 3. Simple use of NOT If NOT <condition> THEN <statements> If the condition is true then NOT <condition> returns the opposite. 1.€, false, and vice versa, e.g. IF NOT(V=U*8) THEN PRINT “V NOT U*8" 156 Now, you may have noticed that it is pointless using NOT in the above relation when you can rewrite it as IF V<>U"*8 THEN PRINT “V NOT U*8” Here is a list of such relations which have the same meanings. NOT(X=Y) X<>Y NOT(X<>Y) X=Y NOT(X<=Y) X>Y NOT(X>=Y) X<Y NOT(X>Y) X<=Y NOT(X<Y) >= So, don’t waste your computer's memory. What NOT can do is reverse the result of an expression such as: IF NOT INSTR(AS,“y”) THEN PRINT “y is NOT IN STRing A$” The above example is self-explanatory. Minimisation — De Morgan’s Law NEGATION OF LOGICAL OR The following three have exactly the same effect. IF NOT(X=9 OR Y= 10) IF NOT(X=9) AND NOT(Y = 10) IF X<>9 AND Y<>10 NEGATION OF LOGICAL AND The following three have exactly the same effect. IF NOT(X=9 AND Y=10) IF NOT(X=9) OR NOT(Y=10) IFXK<>9ORY<>10 If in doubt, try them out and see for yourself. IF/THEN/ELSE STRUCTURES ELSE is a part of the IF/THEN/ELSE structure. ELSE tells the computer that if the <condition> in the IF is not satisfied (i.e. false), then skip the Statements after THEN, and execute the statements after ELSE. Multiple IF/THEN/ELSE statements can be set up as well as nested IF/THEN/ELSE structures. They are only limited in the length of one line, which is 255 characters. lf there are less ELSE clauses than THENs, then each ELSE is matched to the nearest THEN. 157 IF THEN (IF THEN ELSE) —e | IF THEN (IF THEN (IF THEN ELSE) ELSE) ELSE fe x] | Multiple IF/THEN/ELSE IF THEN ELSE IF THEN ELSE. Las ec The second IF statement is executed after the first IF statement is found to be false. Syntax IF... THEN GOTO <line> format can be abbreviated into two forms: 1. IF...THEN <line> 2. IF...GOTO <line> Also ELSE GOTO <line> can be abbreviated: 1. JF...THEN....ELSE <line> 2. IF...THEN....GOTO <line> Points to Remember iIF/THEN/ELSE statements tend to get too long to fit on a single line, because you can have multiple statements after THEN and ELSE. If this is the case, then it is a good idea to use subroutines using the GOSUB statement. It is advisable that you avoid nested IF/THEN/ELSE statements, as they can get into a bit of muddle. You can create a spaghetti program if you don't watch out. 158 CHAPTER 8 PRINT USING INTRODUCTION MSX has a comprehensive print format control, using the PRINT USING statement. The PRINT USING statement allows you to print strings or numbers in various formats. Syntax PRINT USING <string-exp>;<list of items> There are two parts to the argument of PRINT USING — the format specifier string, and the list of items to be printed. They are separated by a semi-colon. The <list of items> can be numbers or strings. <string- exp> contains the special formatting string which determines how the items are to be printed. It can be either a string or a string variable. CHARACTER EXPLANATION ! The exclamation mark specifies that the first character of the string is to be printed. Here is an example program, which prints out only the initial of each of the first names and surnames given in the data statements. 10 FOR I = 1 TO 3 20 READ FIRST$,SUR$ 30 PRINT USING “!";3FIRST$,".",SURS 40 NEXT I 5O DATA JOSEPH,MINTER 60 DATA MAX ,RAYNE 70 DATA DEREK, MCNALLY 159 Inserts a specified string into the position given by @. 10 A$="QWERTY” 20 PRINT USING “THIS IS A @ KEYBOARD” :A$ THIS IS A QWERTY KEYBOARD The “#"” sign indicates a digit to be printed. For example: PRINT USING “#.###":1.3 prints 1.300 i.e. it formats the number to be displayed. As you can see in the above example, you may include a decimal point in the formatting string, between the #” signs. If the number has fewer digits than specified by the formatting characters, it will be right-justified with preceding spaces. 10 PRINT "984634 . 9698 EE" 20 PRINT USING "#444 . 4444 #44" 5 ATNC 1) #4 ee . 3-1459265 If the formatting characters specify more decimal places than the number to be displayed, the empty spaces will be filled with zeros. 10 PRINT "444. #8" 20 PRINT USING “#4. ###HHH" 599-999 set. HEH 99 «999000 Numbers will be rounded if they do not fit in the specified field. 10 PRINT "#4. ###" 20 PFINT USING "##~.###"3510-2367 HH. HHH 10.237 You may print two or more numbers with the same format in one PRINT USING statement. 10 PRINT USING "##.##" ,9.866,18.7 9.97 18.70 If the number to be printed is negative, then the negative sign will be displayed but it will occupy one digit space in the specifier. 10 PRINT “#44. #488" 20 PRINT USING "###.####" 3-27.63 HH. HE -Z-6300 if the number to be printed does not fit into the field specified by the format characters, a % percent sign is printed in front of the 160 ak $$ **§ number. This also happens when the rounded number exceeds the size of the field. PRINT USING "#.###" 59.9999 %10-000 A plus “+” sign at the beginning or the end of the formatting string will result in the sign of the number being printed, i.e. a+" or" at the position specified. PRINT USING "+44. ##4##" 5-O-127d22,10-986 -0-1234d2 +10.98600 PRINT USING "“####.#+"310.71,100,-200.5 10.-7+ 100-0+ 200.5- The double asterisk sign causes the leading spaces in the numeric field to be filled with “*"’s. Also, they represent two more digit spaces in the field. 10 PRINT "#4. HEHEHE 20 PRINT USING "se. #HHEEEHH S ATNC 144 30 PRINT USING “+###. ##4#8888" 5-ATNC1) #4 RHEE. HHH HHH ###3 2140159265 ##-3.14159265 The double dollar sign puts a dollar sign in front of a number. $$ specifies two more digit spaces in the field, one of which will be the $ dollar sign. 10 PRINT “SS#. ##HHHHHH" 20 PRINT USING “SS#. ######88" 5 ATNC 1) #4 ZO PRINT USING "SS#. #448488" 5—-ATNC 1) #4 SSH. HEHEHE SD $3-14159265 -$3.14159265 Note that you can't use the exponential format in conjunction with $$ sign. This sign is a combination of ** and the $$ sign. All blank spaces are filled by * and the number is preceded with a $ sign. **$ specifies three more digit positions, one of which is the $ sign. 10 PRINT "#858. ###EHHHH 20 PRINT USING “##$#. #####8H8" 5 ATNC 1) 44 3O PRINT USING "##$#. ###H#88H" 53-ATN(C 1) 44 HSH. HHHHHHHE ##$3210159265 #-$3 210159265 161 ; Acomma is placed to the left of the decimal point in the formatting string causes a comma to be printed every third digit to the left of the decimal point. 10 PRINT "“###0##88HH , . 9" 20 PRINT USING "34st , . #4" 5 1090382-888 3O PRINT “SS#HHHHHHHHEH, 8" 40 PRINT USING "SS#####8448# , -##" 5 1090382-88% HHHHHHHHHH , 448 1,090,382 -88 SSHHHHHHHHHEH , - 48 $1,090,382-85 Acomma at the end of a format string prints acomma at the end of the number. PRINT USING “##.##, "512.567 12-57, AAAA This is the exponent formatting specifier. This allows space for E+xx. You can also specify the position of decimal points. Significant digits are left justified, and the exponent is adjusted. PRINT USING "##.##°°"*"3200.00 2-00E+02 PRINT USING "“+#.##7777"3200.00 +2-00E+02 PRINT USING "#. ##°°""+"3-200-00 2-00E+02- Notes If the number of digits specified exceeds 24, an illegal function call error will result. LPRINT USING LPRINT USING is exactly the same as PRINT USING but this writes to the printer and not to the screen. PRINT# USING PRINT# USING Is exactly the same as PRINT USING but this writes to the device specified by the file number after the # mark. You must execute an OPEN statement to open a channel to the device you want to write to, before using this statement. Syntax PRINT#<file number>,USING <string-exp>;<list of expression> 162 CHAPTER 9 EVENT HANDLING AND INTERRUPT BY BASIC INTRODUCTION The MSX computer is equipped with various event handling statements. Event handling is interrupt driven. That is, to say, that while the computer is executing a program it is also looking out for a particular event to happen. If that event occurs, it interrupts the proceedings by immediately jumping to a predetermined user subroutine. The MSX can handle interrupts for the following events: 1 Set Time Interval ON INTERVAL GOSUB 2 Function key press ON KEY GOSUB 3 <CTRL><STOP> press ONSTOP GOSUB 4 Trigger press ON STRIG GOSUB 5 Sprite collision ON SPRITE GOSUB 6 Errors ON ERROR GOSUB 1) TIME INTERVAL INTERRUPT Statements used: ON INTERVAL=<=<time>> GOSUB <line> INTERVAL ON/OFF/STOP The MSxX has its own internal clock, which starts when you switch on the computer. The time can be found using the TIME function. TIME is incremented every 1/50th of a second, every time the Video Display Processor makes an interrupt. Using the ON INTERVAL GOSUB statement, you can specify a time interval for the interrupt to occur. Since the TIME is incremented every 1/SO0th of a second, to set the time interval for the interrupt to 1 second 163 you have to equate the INTERVAL to 50, i.e. ON INTERVAL =50 GOSUB <line>. ON INTERVAL=<time> GOSUB also sets the subroutine which the BASIC should jump to once the interrupt occurs. The computer will jump to this subroutine as soon as the interrupt occurs, regardiess of which line in the main program the computer is in. The interval interrupt is enabled with INTERVAL ON. This tells the computer to start counting the clock. Unless this statement is executed, the ON INTERVAL GOSUB statement will not operate. Once the time interrupt occurs, an automatic INTERVAL STOP Is executed. This prevents the time interrupt occurring during the current time interrupt subroutine. However, it remembers if an interrupt has happened during the current subroutine and the computer will immediately execute the subroutine again, if this has happened, unless the current subroutine disables the time interrupt altogether by executing INTERVAL OFF. After it leaves the interrupt subroutine, the computer will automatically execute INTERVAL ON to enable the interrupt, unless INTERVAL OFF is executed within the subroutine. You may disable the time interrupt any time you like using INTERVAL OFF. Time interrupt is disabled as soon as the BASIC gets out of a program, i.e. no time interrupt in direct mode is possible. Also, it is disabled in any error trapping subroutines. EXAMPLES A short program to simulate a digital watch with beeps: 10 ON INTERVAL=50 GOSUB 60 20 INTERVAL ON 30 CLS 40 S=0:M=0 50 GOTO 50 SS REM INTERVAL SUBROUTINE 60 IF S=60 THEN S=O:M=M+1=2LOCATE10O 911 2PRINT "MIN" 3M: BEEP 70 S=S+1 BO LOCATE 10,102PRINT "SEC" 5S 90 BEEP 100 RETURN 164 LINE 10 SETS INTERVAL TO 1 SEC SUBROUTINE AT 60 LINE 20 ENABLES INTERVAL TRAPPING LINE 30 CLEARS THE SCREEN LINE 40 INITIALISESSANDM LINE 50 INFINITE LOOP TO WAIT FOR AN INTERRUPT LINE 60 iF 60 SECONDS THEN 1 MINUTE LINE 70 S+1SECOND LINE 80 PRINTS SECONDS LINE 90 BEEPS LINE 100 RETURNS TO WHEREVER IT LEFT MAIN ROUTINE: IN THIS CASE ALWAYS LINE 50 SEC Sa GIVES A BEEP EVERY MIN 3 SECOND AND A DOUBLE BEEP EVERY MINUTE 2) FUNCTION KEY INTERRUPT Statements used: ON KEY GOSUB <line>,<line>.,...... KEY (<key number>) ON/OFF/STOP itis possible to set up interrupts for the 10 function keys using ON KEY GOSUB and KEY () ON/OFF/STOP STATEMENTS. ON KEY GOSUB is followed by a list of line numbers, specifying the interrupt subroutine for each function key. If there is no corresponding subroutine for a particular function key, then the line number can be omitted. KEY (<key number>) ON statements enable function key interrupts for each function key. Unless this statement is executed, the computer will not look out for an interrupt for the function key. When there is an interrupt, the computer jumps to the subroutine specified by ON KEY GOSUB. Once the function key interrupt occurs, an automatic KEY (<key number>) STOP is executed. This prevents the interrupt by the same function key occurring during the current interrupt subroutine. However, it remembers if that function key has been pressed during the current subroutine, and the computer will immediately go to the same subroutine again, if this has happened, when it has finshed the current subroutine, unless the current subroutine disables the KEY interrupt altogether, by executing KEY (<key number>) OFF. 165 After it leaves the interrupt subroutine, the computer will automatically execute KEY ON to enable the interrupt, unless KEY (<key number>) OFF is executed in the subroutine. KEY interrupt is disabled when the program is not running, and also during error trapping routines. EXAMPLE 10 ON KEY GOSUB 100,120 140,160,180, 200 20 KEY (1) ON 3O KEY (2) ON 40 KEY (3) ON SO KEY (4) ON 60 KEY (5) ON 70 KEY (6) ON 80 GOTO 80 90 REM SUBROUTINES HERE 100 PRINT “FUNCTION 1" 110 RETURN 120 PRINT “FUNCTION 2" 130 RETURN 140 PRINT "FUNCTION 3" 150 RETURN 160 PRINT "FUNCTION 4" 170 RETURN 180 PRINT “FUNCTION 5" 190 RETURN 200 PRINT "FUNCTION 6" 210 RETURN LINE 10 SETS SUBROUTINES FOR F1, F2, F3, F4, F5 AND F6 LINE 20 ENABLES FUNCTION KEY F1 LINE 30 ENABLES FUNCTION KEY F2 LINE 40 ENABLES FUNCTION KEY F3 LINE 50 ENABLES FUNCTION KEY F4 LINE 60 ENABLES FUNCTION KEY F5 LINE 70 ENABLES FUNCTION KEY F6 LINE 80 INFINITE LOOP TO WAIT FOR A FUNCTION KEY LINE 100 ROUTINE FOR F1 LINE 110 RETURNS TO WHEREVER IT CAME FROM LINE 120 ROUTINE FOR F2 LINE 130 RETURNS TO WHEREVER IT CAME FROM 166 LINE 140 ROUTINE FOR F3 LINE 150 RETURNS TO WHEREVER IT CAME FROM LINE 160 ROUTINE FOR F4 LINE 170 RETURNS TO WHEREVER IT CAME FROM LINE 180 ROUTINE FOR F5 LINE 190 RETURNS TO WHEREVER IT CAME FROM LINE 200 ROUTINE FOR F6 LINE 210 RETURNS TO WHEREVER IT CAME FROM 3) <CTRL><STOP» INTERRUPT AND BREAK PROOFING Statements used: ON STOP GOSUB <line> STOP ON/OFF/STOP A BASIC program is very easy to break into. All you need to dois to press <CTRL><STOP>. This halts the program immediately and puts you back into command mode. However, you will sometimes find that you need to ‘break-proof’ a program to prevent users of the program seeing the contents. This is done by using trapping <CTRL><STOP> with ON STOP GOSUB. The ON STOP GOSUB statement sets the subroutine for the <CTRL><STOP® interrupt. Within this subroutine, you can either put a message saying ‘Don’t break out” or just a RETURN statement so the program can immediately carry on even if there was a <CTRL><STOP> interrupt. STOP ON/OFF/STOP enables/disables <CTRL><STOP> trapping. When STOP ON is executed, the BASIC starts to check if <CTRL><STOP> has been pressed every time it executes a new statement. Ifa <CTRL><STOP> has been detected, then the BASIC is diverted to the subroutine specified by the ON STOP GOSUB statement executed earlier on in the program. Once the interrupt occurs, an automatic STOP STOP is executed. This stops the interrupt occurring during the current subroutine. However, it remembers if <CTRL><STOP> is pressed again during the current subroutine, and the computer will immediately go to the subroutine again once it has left the current subroutine, if this has happened, unless the current subroutine disables the <CTRL><STOP> interrupt altogether by executing STOP OFF. After it leaves the interrupt subroutine, the computer will automatically execute STOP ON to enable the interrupt, unless STOP OFF is executed within the subroutine. The only way to break out of a break-proofed program is to RESET the computer. Therefore you should remember to save your break-proofed program before you RUN it. 167 EXAMPLES Here is a ready-made ‘break-proofing’ routine to add on to your program. - 10 ON STOP GOSUB 10000 | 20 STOP ON ADD YOUR PROGRAM HERE 9999 REM CTRL-STOP SUBROUTINE 10000 RETURN LINE 10 STOP SUBROUTINE SET TO LINE 10000 LINE 20 SWITCHES ON THE STOP DETECTOR LINE 10000 RETURNS TO WHEREVER THE INTERRUPT WAS DETECTED Note that ON STOP trapping does not prevent the STOP key halting the program. It just prevents you from breaking into a program by <CTRL><STOP>. If you press the STOP key only, you will find that it will halt your program and display the cursor. If you press the STOP key for the second time, the program will continue. The <CTRL><STOP> interrupt is disabled when the program is not running and also during error trapping routines. 4) JOY STICK TRIGGER INTERRUPT Statements used: ON STRIG GOSUB <a list of line numbers> STRIG(<n>) ON/OFF/STOP The trigger number, <n>, for each trigger is as follows: O=Space bar 1=trigger 1 of joystick 1 2=trigger 1 of joystick 2 3=trigger 2 of joystick 1 4=trigger 2 of joystick 2 Every MSX_ compatible joystick has two trigger buttons. The <space> bar is also considered as a trigger. ON STRIG GOSUB sets trigger interrupt subroutines for these 168 triggers. You can define subroutines for all the triggers at once by listing the line numbers for each trigger. If there is no subroutine for a particular trigger, then you may omit it and just place acomma. For example: ON STRIG GOSUB 1000,200,,, will interrupt to a subroutine when a) trigger 0 space bar or b) trigger 1 on joystick 1 is pressed but does nothing if any of the other triggers are pressed. STRIG(<n>) ON actives the trigger trapping for the specified trigger. You can only enable one trigger at atime. Once the interrupt occurs, an automatic STRIG(<n>) STOP is executed. This stops the interrupt occurring during the current trigger subroutine. However it remembers if a trigger is pressed during this subroutine and the computer will immediately goto the corresponding subroutine for that trigger once it has left the current subroutine, unless the current subroutine disables trigger interrupt for that trigger altogether by executing STRIG(<n>) OFF. After it leaves the trigger subroutine, the computer will automatically execute STRIG(<n>) ON to enable the interrupt. The STRIG interrupt is disabled when the program is not running and also during error trapping routines. EXAMPLES Here is a very short program which demonstrates how the trigger trapping works by testing each of the joystick triggers. The program is written so that both joysticks can be tested as well as the space bar as a trigger. If you press the <s> key, the program will end. 169 ON STRIG 160,180 STRIGCO) STRIGC1) STRIGC2) STRIGCS) STRIG(4) IF INKEY$="s" GOTO 70 GOSUB 100,120,140, ON ON ON ON ON THEN END REM TRIGGER ROUTINES PRINT "SPACE BAR RETURN PRINT "TRIGGER 1 RETURN PRINT "TRIGGER 1 RETURN PRINT "TRIGGER 2 RETURN PRINT "TRIGGER 2 RETURN PRESSED" JOYSTICK JOYSTICK JOYSTICK JOYSTICK. SETS ALL THE TRIGGER SUBROUTINES SWITCHES THE TRIGGER 0 ON SWITCHES THE TRIGGER 1 ON SWITCHES THE TRIGGER 2 ON SWITCHES THE TRIGGER 3 ON SWITCHES THE TRIGGER 4 ON IF <s> IS PRESSED THEN END LOOPS BACK TO LINE 70 TRIGGER ROUTINES 5) SPRITE COLLISION INTERRUPT See the Advanced Graphics Sprite section for detailed explanation. 6) ERROR DETECTION See the Error Handling section for more detail. 170 CHAPTER 10 ERROR HANDLING ERROR MESSAGES When MSX BASIC encounters an error while executing a program or a command, it displays an error message and halts the execution. Here is a summary of error messages. CODE 1 MESSAGE NEXT without FOR A variable in a NEXT statement does not match with a variable in a previous FOR statement: ora NEXT statement is encountered without a previous FOR statement. Syntax error Spelling mistakes in the Keyword or wrong punctuation. Return without GOSUB A RETURN statement is encountered without a matching GOSUB statement. Out of DATA A READ statement is executed when all the data in DATA statements is used up; or there are no DATA statements to READ from. \Ilegal function call An illegal paramater is passed to a mathematical or string function. Overflow The number is too large for the variable type, function or statement to handle. DESCRIPTION 171 18 19 20 Out of memory Ran out of memory due to a program that is too large, too many FOR, too many GOSUB, too many functions or variables. Undefined line number Non-existent line number referred to in a GOTO, GOSUB, IF/THEN/ELSE statement or a DELETE command. Subscript out of range A subscript of an array element is out of the range specified in the DIM statement: or the number of the subscript is wrong. Redimensioned array Dimensionalising an already existent array using DIM. Division by zero A division by zero in an expression has been encountered. Illegal direct A statement that is not allowed in direct mode. Type mismatch A string variable is assigned to anumber or vice versa. Out of string space Ran out of string work area in systems RAM. String too long An attempted creation of a string of more than 255 characters long. String formula too complex A string expression is too complex for the computer to handle. Can't continue Attempted CONTinue on a program that has ended with an error, or has been edited or ENDed. Undefined user function Calling an FN function without first defining it with DEF FN statement. Device |/0 error An input/output error on a cassette, printer etc. It is a fatal error with no way of recovery. Verify error CLOAD? cassette tape verification has failed. 172 | 22 23 24 25 26-49 50 51 52 53 54 55 56 of No RESUME An error trapping routine has no RESUME statement when one is expected by the computer. RESUME without error An unexpected RESUME statement is encountered outside an error trapping routine. Unprintable error The error is encountered but there is no message. Missing operand An expression contained an operator with no operand following it. Line buffer overflow An entered line has too many characters. Unprintable error Reserved for future expansion. Field overflow AFIELD statement is attempting to allocate more bytes than specified for the record length of a random file in the OPEN statement; or the end of the FIELD buffer is encountered while doing sequential 1/0 (PRINT#, INPUT#) to a random file. Internal error Machine malfunction. Bad file number A statement or command referring to anon-existent file, or out of MAXFILES range. File not found A LOAD or OPEN statement referring to a non-existent file. File already open Attempt is made to OPEN an already OPENed file. Input past end An INPUT statement is executed after all the data in the file has been exhausted. Use EOF to prevent this happening. Bad file name An illegal file name given in LOAD, SAVE, or other |/0 statements. Direct statement in file A direct statement is encountered while LOADing. LOAD Is terminated. 173 58 Sequential 1/0 only A statement to random access is issued for a sequential file. 59 File not OPEN The file referred to is not yet OPENed. 60-255 Unprintable error No error codes. These may be defined by the user. ERROR HANDLING MSX BASIC contains many functions and statements to help you debug your program. It is usual practice to incorporate an error handling or trapping routine within your own program while it is still under development. This will help you to spot any mistakes you may have made while doing a trial RUN of your program. Before we move on to how to write such error trapping routines, let's have a look at each of the BASIC keywords used in error handling. ERL means error line ERL is a reserved variable which contains the line number of the line containing the error found in the program just RUN. ERR means error code number ERR contains the error code number after the computer has detected an error in a program. It has a value between 1 to 255. The meanings of errors associated with the ERR number are listed as in the previous page. ERROR There are basically two ways of using the ERROR statement. 1. To simulate the occurrence of an error. An ERROR statement with an integer argument will fool the computer into thinking that there is an error and will terminate the program and print out the error message with the line number. 2. To create user defined errors. An ERROR statement together with the use of an ON ERROR GOTO error trapping function will let you create your own customised errors (more on this later). ON ERROR GOTO <line> ON ERROR GOTO enables error trapping, and specifies which line to goto once an error has been detected. Once the computer is told to trap errors it will jump to the specified line, to execute the error handling subroutine, if an error occurs during a program or in direct mode, i.e. outside the program. 174 RESUME RESUME means resume BASIC operation after an error handling procedure has taken place using the error trapping statement ON ERROR GOTO. After the error has been dealt with, RESUME tells the computer to continue the execution according to the syntax below: RESUME or RESUME 0 Resumes execution from the statement which caused the error. RESUME NEXT Resumes execution from the statement following the one which caused the error. RESUME <line> Resumes execution from the line number specified Error Handling Routines To incorporate an error handling routine into your program you must have the following: 1. ONERROR GOSUB statement right at the beginning of the program to enable the error trapping from then onwards. After this statement has been executed the computer will look out for an error, and divert to the error handling routine whether you are in direct mode or indirect mode. 2. Anerrorhandling subroutine placed right at the end of the program. A simple program with an error handling routine looks like this. ft ~ 1 10 ON ERROR GOTO 100 20 PRINT 10 30 PRONT 20 40 PRINT 30 270 END 99 REM ERROR HANDLING ROUTINE 100 ON ERROR GOSUB 0 LINE 10 ENABLES ERROR TRAPPING AND SETS THE ERROR SUBROUTINE TO LINE 100 LINE 30 MISTAKE HERE LINE 50 END OF PROGRAM LINE 99 ON ERROR GOSUB 0 CAUSES THE BASIC TO STOP AND PRINTS OUT THE ERROR WHICH HAS CAUSED THE TRAP When the above is executed the following will happen: 175 RUN 10 Syntax error in 30 The error in line 30 was detected and the program diverted to line 100. ON GOSUB O has an effect of displaying the error which caused the trap and halted the program. It also disables the error trapping. So far, all this can be done without having the error trapping at all. If there was no ON ERROR GOSUB statement the computer will still detect the error in line 30, and give the “Syntax error in 30” message. The error trapping routine, however, does have other features. For instance, the error handling subroutine could contain an error recovery procedure, to put you back into the program. To do this, you need to know what kind of errors you want the recovery to cope with, and how the computer should RESUME operation. In this example there are too few data items in the DATA statement in line 60. This causes an out of DATA error (code 4) when the computer is on its fourth loop. The error handling routine is equipped with a recovery procedure for such an error in the form of line 100. This line finds out if the error was really the out of DATA error by checking the ERR error code, and then executes a RESTORE statement so that the data can be used again. It RESUMEs operation from the line where the error was detected. 10 ON ERROR GOTO 100 20 FOR I= 1 TO 5 30 READ At 40 PRINT A&% 30 NEXT I 60 DATA ONE, TWO, THREE 70 END 90 REM ERROR ROUTINE 100 IF ERR=4 THEN RESTORE: SPRINT “out of DATA": RESUME 110 ON ERROR GOTO O LINE 10 oe ERROR TRAPPING AND SETS THE ERROR SUBROUTINE TO LINE 1 LINE 20 LOOP FIVE TIMES LINE 30 READS DATA LINE 40 PRINTS A$ LINE 50 NEXT LOOP LINE 60 ONLY THREE DATA !ITEMS HERE 176 LINE 70 END LINE 100 !F ERR WAS OUT OF DATA (CODE 4) THEN RESTORE AND RESUME OPERATION FROM WHERE IT HAS LEFT OFF LINE 110 PRINTS OUT ERROR MESSAGE IF ANY OTHER ERROR IS ENCOUNTERED RUN ONE TWO THREE out of DATA | ONE TWO With some errors you wouldn't want the computer to RESUME operation from the same line. Errors such as Syntax error can't really be remedied within a program, so they have to be edited. However, if you want the computer to skip the line with the error while giving you a message, and then follow the program through to the end, use the RESUME NEXT statement which resumes operation from the next line after the one in which the error was detected. r pearets See ee Batts SS, 10 ON ERROR GOTO 100 20 PRINT 10 30 PRoNT 20 40 PRINT 30 30 END 99 REM ERROR HANDLING ROUTINE 100 PRINT “Error code “;ERF; "at line";ERL 110 RESUME NEXT ie NE 10 ENABLES ERROR TRAPPING AND SETS THE ERROR SUBROUTINE TO LINE 100 ANE 30 MISTAKE HERE LINE 50 END OF PROGRAM LINE 100 DISPLAYS ERROR CODE AND THE ERROR LINE LINE 110 THEN RESUMES OPERATION FROM NEXT LINE RUN 10 Error code 2 at line 30 30 Note that you can specify which line to goto in the RESUME statement, so Inthe above example, RESUME 40 has the same effect. It is possible to display the line which contains the mistake using an error handling routine. it will be much easier to correct the mistake if the line with the error is displayed straight after the error has been detected. To do this you must end the error recovery routine with the LIST. (list dot) command. LIST. has the effect of listing the last line referred to by the computer. 10 ON ERROR GOTO 100 20 PRINT 10 30 PRONT 20 40 PRINT 30 50 END 99 REM ERROR HANDLING ROUTINE 100 PRINT “Error code "sERF; “at line";sERL 110 LIST- ie a = anes | LINE 10 ENABLES ERROR TRAPPING AND SETS THE ERROR SUBROUTINE TO LINE 100 LINE 30 MISTAKE HERE LINE 50 END OF PROGRAM LINE 100 DISPLAYS ERROR CODE AND THE ERROR LINE LINE 110 THENLISTS THE LINE WITH THE ERROR [ RUN 10 Error code 2 at line 30 30 PRoNT 20 1 Now you can correct the line 30 to: 30 PRINT 20 178 HOW TO CREATE YOUR OWN CUSTOMISED ERRORS An ERROR statement together with the use of an ON ERROR GOTO error trapping function will let you create your own customised errors. There are about 36 errors between error code numbers 1 and 60 inthe present version of MSX BASIC. Code numbers 61 to 255 can be used by the programmer. To program your own error, first of all you must put the computer into the error trapping mode by executing the ON ERROR GOTO <line> statement at the beginning of the program. Then, if in the middie of the program, a condition arises such that you want to call the error trapping routine, you can do so by: IF <condition> THEN ERROR <error code>. ON ERROR GOTO will be activated and the computer will immediately start executing the error trapping subroutine. Within the subroutine you must have a line saying: IF ERR=<error code> THEN PRINT “Error message” In the following example all the commands inputted with KILL will be treated as a new error (No. 255), and are deait with in the error trapping routine using the ERR function. je a 10 ON ERROR GOTO 100 20 INPUT “MY LORD, WHAT IS THY COMMAND" ;A% 30 IF INSTRCA$,"KILL") THEN ERROR 255 40 PRINT "OK." 30 END 100 IF ERR=255 THEN PRINT"KILLING IS ILLEGAL IN THIS ADVENTURE-"“:RESUME 20 110 END =a a LINE 10 ACTIVATES ERROR TRAPPING LINE 20 INPUT LINE 30 CHECKS FOR INVALID WORD LINE 100 ERROR MESSAGE RESUME FROM 20 LINE 110 ENDIF ERROR !S NOT 255 RUN MY LORD, WHAT IS THY COMMAND? KILL HIM KILLING IS ILLEGAL IN THIS ADVENTURE MY LOPD, WHAT IS THY COMMAND? GOTO EAST OK 179 It is usual practice in MSX BASIC, when you are defining your own error codes, to work your way down from 255, so that you can avoid clashes with any future enhancements to the error messages made by the manufacturers of MSX computers. Notes on Error Handling The ERROR statement can also be used to simulate the occurrence of an error. For example: ERROR 2 will give: Syntax error lf the error code has no previously defined error message, then, when the computer encounters this error in the program, it will print “Unprintable Error”. No error trapping is carried out during an error trapping routine. If there is an error within the error trapping routine, then it will give the error message and halt the program. Not all errors can be dealt with by error trapping routines. It is therefore recommended to end the error trapping routine with ON ERROR GOTO 0 which will stop the BASIC program execution and display the error message. ON ERROR disables all event handling traps such as ON INTERVAL and ON STRIG. 180 CHAPTER 11 CASSETTE SAVING AND LOADING INTRODUCTION There are several ways of saving and loading programs and data to a cassette tape recorder.This section deals with more sophisticated use of the cassette tape recorder. Statements and Functions Associated with Cassette Saving and Loading Here is a list of statements and functions which are associated with the display screen. Saving and loading BASIC programs (see the Introduction to MSX BASIC for details). CLOAD Loads a BASIC program from cassette CSAVE Saves a BASIC program to cassette CLOAD? Verifies a BASIC program on tape to the one in the computer's memory MOTOR Switches on/off cassette motor. Saving and Loading and Merging using the ASCIli format SAVE Saves a BASIC program to the specified device, in an ASCII file. LOAD Loads a BASIC program saved in an ASCII file from the specified device. MERGE Merges a BASIC program saved in an ASCII file to a BASIC program in the computer's memory. 181 Saving and Loading a section of the computer's memory as machine code program or data. BSAVE Saves a section of memory. BLOAD Loads a section of memory. Setting the BAUD rate of cassette saving SCREEN CSAVE THE CASSETTE SAVE BAUD RATE The baud rate is the rate of data transmitted to the cassette tape recorder. The MSX uses the FSK format which gives you the option of two baud rates: 1,200 baud or 2,400 baud. This can be set by either the SCREEN staternent or the CSAVE command. The default Baud rate is 1200. Their syntaxes are as follows: SCREEN ,,,1 1200 baud SCREEN ,,,2 2400 baud CSAVE “<name>” 1200 baud CSAVE “<name>",1 1200 baud CSAVE “<name>"’,2 2400 baud The recorded baud rate can be either 1200 baud or 2400 baud but the CLOAD, LOAD, and BLOAD will automatically sense which rate the recording is in and load accordingly. SAVING AND LOADING IN THE ASCII FORMAT When an ordinary BASIC program is recorded on cassette tape, itis ina tokenised format so that the program can be saved in an efficient format. However, if you need to record your program in ASCII codes, you must use the SAVE statement. To load in a program saved in ASCII code you must use the LOAD statement. In both SAVE and LOAD, you must specify the device you are saving from and loading to, in the form of <device-descripter>. However, only the cassette is supported in the current MSX BASIC, so the <device-descriptor> should always be “CAS:”. Here is a list of syntax: SAVE '<device-descriptor> <program-name>" Saves a Basic program in an ASCII file. LOAD “'<device-name>” Loads the first BASIC file found, saved in ASCII format. LOAD “<device-name> ioads a BASIC file with the <file-name>" specified name. LOAD “<device-name> loads a BASIC file with the specified name, <file-name>",R then runs it. 182 Note that the R option in LOAD will auto-run the BASIC program just loaded. The main use of saving in ASCII code is that the program can be merged to another, which is what we will look into next. MERGING TWO BASIC PROGRAMS Let us say that you have two Basic programs, program 1 and program 2 and you want to merge program 2 with program 1, such that program 2 comes after program 1 when it is eventually merged. Let’s say that both are initially saved on a cassette tape. PROGI 10 REM PROGRAM 1 20 PRINT "WELCOME" 30 PRINT" TO" a0 PRINT"MSX" PROG2 10 REM PROGRAM 2 20 FOR I = 32 TO 58 30 PRINT CHR#CI) a0 NEXT I a First, load program 2 and renumber it so that all line numbers are greater than those in program 1. CLOAD "“PROG2” RENUM 50 LIST SO REM PROGRAM 2 60 FOR I = 32 TO 58 70 PRINT CHRSCI) 80 NEXT I Then save it in an ASCII file using SAVE: SAVE “CAS:PROG2” Load program 1 from the cassette: CLOAD “PROG1" 183 After you have loaded program 1 from cassette, rewind the program to where PROG2 is recorded. Type in: MERGE"CAS:PROG2” and play the cassette. The computer will add program 2 to the end of program 1. 10 REM PROGRAM 1 20 PRINT"WELCOME" 30 PRINT"TQ" 40 PRINT"MSX" 5O REM PROGRAM 2 60 FOR I=32 TO 58 70 PRINT CHRSCI) r ‘ 71 SO NEXT I Le. Please note the following when you are merging two programs: 1) If a line number occurs in both the initial program (1), and the second program (2), the resulting MERGEd program will contain the line from the second program (2). 2) MERGE needs the device-name in its syntax. MERGE ‘‘<device-name><file-name>” <device-name>=CAS: for cassette. 3) if the file name is omitted in the MERGE statement, then the next ASCII file encountered on the tape will be merged. SAVING AND LOADING A SECTION OF THE COMPUTER’S MEMORY To save a section of memory, at a specified memory address, on to cassette, use the BSAVE command. You must specify the start address and the end address of the memory you want to save. This command is used for saving machine code programs and data in the form of bytes. lf you are saving a machine code program, you have an option to specify the execution address. When you are loading the machine code in with BLOAD, if the BLOAD specifies to auto-execute the machine code program, it will execute the program from the execution address specified in BSAVE. Anything that is saved with BSAVE must be loaded using BLOAD. if the execution address is omitted when the code is BSAVEd, then the computer assumes the execution address is the start address when the program is BLOADed with the R option to auto-execute the program. 184 Here is a list of syntax: BSAVE “<device-name>:<name:=",<<start address>>,<end address> BSAVE ‘‘<device-name>:<name>”’,<start address>,<end address>,<execution address> BLOAD “<device-name>:<name>",R <name> and R are optional. R option automatically executes the machine code program just loaded. R stands for RUN. BLOAD ‘‘<device-name>:<name>",<num-const> When the off-set <num-const> is specified the file will be entered in memory at the position specified by <num-const>. <device-name>=CAS: for cassette The addresses can be given in hexadecimal numbers, e.g. BSAVE "CAS:PROG’” &HF300,&HF380,&HF30A 185 CHAPTER 12 ADVANCED GRAPHICS | CHARACTERISTICS OF EACH SCREEN MODE MSX has one of the most versatile graphics chips, the TMS 9929A, developed by Texas Instrument Inc. in the United States. It is one of the most comprehensive graphics chips available, and provides you with a variety of facilities, such as 16 colour high resolution graphics and sprite animation. It can even manage its own 16K RAM, which means that the central processor is free from providing memory for the graphics screen. MSX BASIC has been written especially with these strong graphics capabilities, and easy programming for beginners, in mind. It is easy to use once you have experimented with it, and it can produce some spectacular pictures if you try hard enough. Let us then start from the beginning by first explaining the display modes available on the MSX. SCREEN MODE MODE | TEXT/GRAPHICS ; RESOLUTION | COLOUR | INPUT GRAPHIC] SPRITE ) 40x24 textmode | 40x 24char | 2outof16} yes no no 1 32x24 textmode | 32x 24char | 2 outof 16 yes no yes 2 Hi-res graphic 256 x 192 char 16 no yes yes 3 Mutt colour 64x 48 biock 16 no yes yes NOTES: INPUT —use or not of the INPUT statement in the given screen mode. You can’t INPUT while in a graphics mode. GRAPHICS — use of graphics statements, for example DRAW, in the given screen mode. Generally you can’t use graphics statements in text modes (0 and 1). 186 Statements and Functions Associated with Screen Modes Here is a list of statements and functions which are associated with the display screen. More details are given in BASIC reference section. Statements associated with ALL screen modes SCREEN Sets the screen mode. CLS Clears the screen. COLOR © Sets foreground, background and border colour. Statements and functions associated with TEXT MODES ONLY WIDTH — Sets the width of text screen. LOCATE Sets the position of the cursor on text screen. TAB Sets the horizontal position of the cursor on current line. CSRLIN Returns the current vertical position of the cursor. POS(Q) — Returns the current horizontal position of the cursor. Statements and functions associated with GRAPHICS MODES ONLY CIRCLE Draws acircle. DRAW Draws according to graphics macro language. LINE Draws lines, squares and boxes. PAINT Paints with current foreground colour. PSET Plots a point. PRESET Unplots a point. POINT Returns the colour of the pixel referred to. Statements associated with the VIDEO DISPLAY PROCESSOR VPOKE Poke to Video RAM. VPEEK Peek from Video RAM. VDP Returns values of VDP register. BASE Returns the base address of Video RAM tables. MODE 0: 40 x 24 TEXT MODE SCREEN 0 <——__-——— maximum width of 40 characters —-—---__> (0,0) =o 24 char. WIDTH—1,23 ee tes ( A ~<-——— width set by WIDTH statement —->> 187 MODE 0 gives 40 characters per line, which is the maximum number of characters you Can have in one line for MSX computers. In this mode you may write and edit your program. You will find that the display ofa program in this mode is easier to read. MODE 0, however, has a number of disadvantages. For instance, the characters are displayed in a compressed format, i.e. 6 by 8 instead of 8 by 8 and hence some graphics characters will appear cut off. The 2 rightmost pixels of each character are not displayed at all! That does not affect alphanumeric characters, however. The default width of the screen in MODE 0 is 37 characters. You may increase the display width to 40 characters using the WIDTH statement. The coordinates of the top lefthand corner are (0,0) while the bottom righthand corner is (WIDTH—1,23), or when initialised it is (38,23). You may position the text cursor using TAB and LOCATE. To find out where the cursor is located, the function POS(0) and CRSLIN are used. In MODE 0, you cannot use SPRITES, and also you can only use 2 out of 16 colours, although the combination of two colours is entirely your choice. The default colour is the same as MODE 1, white foreground colour and blue background. Note that MODE 0 does not have any border: it simply disappears from the display, so setting the border colour in MODE 0 is meaningless. Note that the colour of the display will change immediately upon execution of the COLOR statement. s eee graphics mode, you are free to use INPUT statements and Looe key list is displayed at the 23rd line unless it is disabled by All graphics statements and functions in this mode will be treated as an ‘Illegal function call’, so watch out for that. MODE 1:32 x 24 TEXT MODE SCREEN 1 <_—_—--—— maximum width of 32 characters aera a, ry 24 char. (WIDTH — 1,23) ~«<— width set by WIDTH statement ——» 188 Mode 1 has aresolution of 32 by 24 characters but no graphics. Inthis mode you may write programs and edit them. It can display 8 by 8 characters without any ‘cut-offs’ (as in the case in MODE 0.) The default width of the screen is 29 characters. You may increase the display width to 32 characters using the WIDTH statement. The reason for the smaller default width is because some TVs and monitors cannot display the entire screen. The coordinates of the top lefthand corner are (0,0), while the bottom righthand corner is (WIDTH — 1,23), or when initialised, it is (28,23). You may position the text cursor using TAB and LOCATE. To find out where the cursor is located, use the POS(0) and CRSLIN functions. You can only ‘use 2 out of 16 colours, although the combination of two colours is entirely your choice. Note that the colour of the display will change immediately upon execution of the COLOR statement. Unlike the graphics mode you are free to use INPUT statements, and the function key list is displayed at the 23rd line unless it is disabled by KEY OFF. All graphics statements and functions in this mode, except those involved in sprites, will be treated as illegal function calls. MODE 2: 256 x 192 HIGH RESOLUTION GRAPHICS MODE : SCREEN 2 << —_—_—__—_——- 256 PIXELS -———____-_ (0.0) — (255,0) A 192 PIXELS (0,191) (255,191) | y MODE 2 is the most widely used graphics mode which provides the user with a high resolution graphics screen, with 16 colours. In this mode you may use sprites and graphics macro language through DRAW statements. This is the mode which is used in most games programs. Its horizontal resolution is 256 pixels, while the vertical resolution is 192 pixels. The colour resolution, however, is somewhat different, its colour resolution is 32 by 192. This means you can only select two colours per 8 by 1 horizontal! block of pixels. You can select the back- ground and foreground colours for each 8 pixel blocks, but if you plot a 189 point with a third colour, the pixels plotted in the previous colour within that block will change to the new foreground colour automatically. Hence you can obtain smudgy pictures if you are not careful with where you are plotting and with what colour. However, if you draw carefully with 8 pixel colour resolution in mind, you can produce good pictures. More on advanced drawing techniques later. You may use sprites in this mode. The Sprites are on a different screen plane from that of the main screen plane. This meansithat whatever sprites are displayed, they do not affect what is plotted on the main screen in any way. You will see that when a sprite leaves a particular part of the screen, the main screen is totally undisturbed. in graphics MODE 2, the function key list cannot be displayed. You will also find that ordinary PRINT statements will not work in this mode. To PRINT onto the graphics screen, you must first OPEN a file to the screen and use PRINT#. This will be explained in more detail in the Graphics mode printing section. The COLOR statement in this mode does not change the display colour immediately. To change the colour of the entire scren you must execute a CLS statement. You cannot use the INPUT statement in this mode, because it would force the screen back to the previous text mode used. MODE 3: 64 x 48 MULTI-COLOUR LOW RESOLUTION GRAPHICS MODE: SCREEN 3 << 64. BLocKg > (0,0) (255,0) | A 48 BLOCKS (0,191) (255,191) | Multi-colour MODE 3 gives a low resolution of 64 by 48 blocks but the individual blocks can have their own colour (unlike MODE 2). You won't get the smudged effects as in MODE 2 but it is in low resolution, so there are few applications of this mode. Its coordinates are the same as in MODE 2, but 4 by 4 pixel co- ordinates represent one block. This allows you to draw and plot in the same coordinates system as in MODE 2, which is advantageous. 190 You may use sprites in this mode in much the same way as in MODE 2 In multi-colour MODE 3, the function key list cannot be displayed. You will also find that the ordinary PRINT statement will not work inthis mode To print onto the graphics screen you must first OPEN a file to the screen and use PRINT#. The size of the printed characters will be 4 times as large as in MODEs 1 and 2. This will be explained in more detail in the Graphics mode printing section. The COLOUR statement in this mode does not change the display colour immediately. To change the colour of entire screen, you must execute a CLS statement. You cannot use the INPUT statement in this mode because it would force the screen back to the previous text mode used. 191 CHAPTER 13 ADVANCED GRAPHICS II COLOUR IN HIGH RESOLUTION MODE 2 This section deals exclusively with how colours are used in MODE 2. The COLOR statement sets the colour of the current background, fore- ground and border. However, the background colour of the whole of the screen will not change unless you clear the screen with a CLS state- ment. After the COLOR statement is executed, any point, line, or shape drawn will be in the new foreground colour, unless another colour is specified within the drawing statements. The colour resolution of MODE 2 is 32 by 192. This means that you can only select two colours per 8 by 1 horizontal block of pixels. You can specify the background and foreground colours for each 8 pixel block but if you try to plot a point with a third colour to a block which already has two colours, the pixels plotted in previous foreground colour within that block will change to the new foreground colour automatically. This can cause smudgy pictures if you are not careful. The short program below will demonstrate this point. 10 SCREEN 2 20 COLOR 4,15,15 30 CLS 40 LINE (0,0)-(4,191),4,BF 50 IF NOT STRIGCO) THEN SO 60 LINE (6,0)-(6,191),10 70 GOTO 70 192 —_— ; = 1 LINE 10 SELECTS MODE 2 LINE 20 SETS BACKGROUND TO WHITE LINE 30 CLEARS THE SCREEN TO WHITE LINE 40 DRAWS A RECTANGLE IN BLUE LINE 50 WAITS UNTIL SPACE BAR IS PRESSED LINE 60 DRAWS ALINE iN DARK YELLOW BESIDE THE BLUE RECTANGLE. LINE 70 LOCKS GRAPHICS SCREEN Do run this program. First, you will see a blue rectangle drawn on white background, at the edge of the screen. Then, if you press the <space> bar you will see a yellow line being drawn and the blue rectangle will change to yellow as well. The program did not plot any yellow rectangle where the blue one was. Why did the colour of the rectangle change just by plotting a yellow line near it? To understand what has happened, let us see how the colour attrib- utes are stored in the video RAM. lf you take the top left hand side section of the screen and magnify it, it may be represented as the table given below. (B means blue and W means white.) Within the MSX’s Video RAM, the picture above is held in binary code as shown below. Each bit corresponds to a point on the screen, 1 = foreground colour, and 0 = background colour. In this case, 1 repre- sents blue, 0 white. 3 ee ee es 8 9 7 o {1111111 ]o Jo ololololo 1 E A: 1lolo | 010 ]o = 2 |4 iii 1 [0 0 lo olololo 4 x 1/4 ki 0 ae o/ololo 4 193 This table of memory, however, does not reveal what the colour of each point is. This is because there is a Separate colour attribute table in the VRAM which contains the foreground and background colour for each of 8 by 1 pixel blocks. Let us take the top left hand corner block and see how its pattern and colour is stored in the memory. PATTERN TABLE COLOUR ATTRIBUTE TABLE QO 4 2 3 4 5 As you Can see from the above, while the pattern table gives which pixel is in foreground and which one is in background colour, the colour attribute table gives which colour is the foreground colour and which ts the background. This means that it is physically impossible to plot more than 3 colours in one block. So, when you try to draw a yellow line in column 6, first the foreground colour in the attribute table is changed to the colour code of yellow i.e. 10, and then pixel 6 Is set to this new foreground colour. PATTERN TABLE COLOUR ATTRIBUTE TABLE 1 0) Pixels 0, and 1, 2, and 3 stay as foreground colour but now the fore- ground colour itself has changed from blue to yellow, and hence with it the colour of O, 1, 2, and 3 pixels have changed. The explanation is simple when you see it in terms of the memory table, isn't it? So, what do we do about this problem? There Is only one answer: try not to plot more than 3 colours in one block. For example, the program above can be changed so that the yellow line is not plotted in the same block as the blue rectangle, by moving the whole picture left by 2 pixels. The improved program, which will plot a blue rectangle and yellow line side by side, will look like the following: 194 10 SCREEN 2 20 COLOR 4,15,15 30 CLS 40 LINE (2,0)-(6,191),4,BF 50 IF NOT STRIG(O) THEN 50 60 LINE (8,0)-(8,191),10 70 GOTO 70 LINE 10 SELECTS MODE 2 LINE 20 SETS BACKGROUND TO WHITE LINE 30 CLEARS THE SCREEN TO WHITE LINE 40 DRAWS A RECTANGLE IN BLUE LINE 50 WAITS UNTIL THE SPACE BAR PRESSED LINE 60 DRAWS ALINE IN DARK YELLOW BESIDES THE BLUE RECTANGLE LINE 70 LOCKS THE GRAPHICS SCREEN Note the changes which have been made in the coordinates specifiers in lines 40 to 60. NOTE: if you want to know more about how the screen is stored inside the video RAM, turn to the Video RAM section. 195 CHAPTER 14 ADVANCED GRAPHICS lil HOW TO PRINT TO THE GRAPHICS SCREEN USING FILES PRINTing to the graphics screen is not as straight forward as PRINTing inatext mode. You must open a file to the graphics screen, and then use the PRINT# statement. In fact, you can open a file to a text screen if you like, but this is not necessary. Here is a demonstration program, which will print to all four screen modes, using files. Line 30 OPENs the file according to which screen mode (M) ts selected. It executes OPEN “CRT:” AS #1 when M Is less than 2. This means OPEN a file to the text screen "CRT:” AS, file number #1. When M is more than or equal to 2, then the statement executes OPEN “GRP:” AS #1 which means OPEN a file to the graphics screen “GRP:" AS, file number #1. A file OPENed shouid be CLOSEd once Its function has finished. This is done in line 60: CLOSE#1. ZO SCREEN M 30 IF M<2 THEN OPEN "CRT=" AS #1 ELSE OPEN "GRP:" AS #1 d0 PRINT#1,"THIS IS MODE "5M SO FOR G=1 TO 1000 60 CLOSE#1 7O NEXT M L Z J _ 1 | 10 FORM =0 7103 LINE 10 LOOPSOTO3 LINE 20 SELECTS SCREEN MODE LINE 30. IF TEXT MODE THEN OPEN “CRT: ELSE OPEN “GRP: GRAPHICS SCREEN LINE 40 PRINTS A MESSAGE LINE 50 DELAY LINE 60 CLOSE FILE LINE 70 NEXT MODE In the program above you will see the screen change to each different mode while printing the message “THIS IS MODE . . .”. Note that in the high resolution screen the PRINTing is basically in the same format as in text MODE 1, 1.e. 32 by 24 characters. !n the multi-colour MODE 3, the text is four times the size of the normal text mode. This is because the multi-colour mode is in low resolution, and in order to PRINT characters it has to be done in block graphics. How to Print in High Resolution Graphics Mode 2 When printing in graphics MODE 2 you cannot use the LOCATE state- ment to position the characters to be printed. This is because the computer does not use the text cursor in the graphics mode. Instead, it uses the graphics cursor, and therefore the computer PRINTs to the last graphics cursor position. To position the characters to be displayed, use the DRAW command to move the graphics cursor with graphics macro language's BM (blank move). Example program: PRINT # and PRINT# USING in graphics mode 2 r ; : : _ ; ss 7] 10 SCREEN 2 20 OPEN "GRP" AS #1 30 DRAW "BM100, 100" 40 PRINT#1,"HELP" 50 DRAW "BM102,102" 60 PRINT#1, "HELP" 70 DRAW "BM100,120" SO PRINT#1, USING" #44. ####8##”" SATNC 1) 4d 390 GOTO 90 £10 SELECTS SCREEN MODE 2 NE 20 OPEN FILE TO GRP NE 30 POSITIONS THE GRAPHICS CURSOR AND PRINT# THERE NE 50 REPOSITION THE CURSOR r | bs E 60 PRINTS HELP AGAIN E 70 REPOSITIONS CURSOR AGAIN NE 80 PRINT# USING E 90 LOCKS GRAPHICS MODE Pee ie ee 197 In the above example you will see two “HELP” messages overlapping each other. This arises because in the graphics mode the computer does not erase before printing. This can create some interesting effects, but it is often a nuisance. The only way to erase what is printed on a part of the screen is to use the LINE statement and overwrite it with a rectangle filled with the background colour. If you include the line: 45 LINE (100, 100)—STEP(32,8),4,BF this will erase the “HELP” message written in line 40, thus avoiding the overlap. As you can see in the program above, you can also use PRINT USING, in the form of PRINT# USING. All the various syntax of PRINT USING are the same for the graphics modes, as for the text mode. Note that ATN(1)*4 = Pi = 3.14159... How to Print Using Colour in Graphics MODE 2 One of the advantages of printing in a graphics mode is that you can use all 16 colours for printing characters as well as for the graphics. To print in a specified colour, all you have to do is to specify the current fore- ground colour before PRINT#. Here is an example to illustrate 16 colour printing. 10 COLOR 1,15,15 20 SCREEN 2 30 OPEN "GRP=" AS #1 40 FOR I = 0 TO 15 50 COLOR I 60 DRAW “BM+8,0" 70 PRINT#1,"COLOUR CODE";1 80 NEXT I 90 GOTO 90 LINE 10 BACKGROUND COLOUR IS WHITE LINE 20 SCREEN MODE 2 LINE 30 OPENS FILE TO GRP AS #1 LINE 40 LOOP LINE 50 COLOUR OF FIRST LOOP LINE 60 OVES GRAPHICS CURSOR LINE 70 GIVES MESSAGE LINE 80 NEXT LOOP LINE 90 LOCKS THE SCREEN The program above will result in the following display. 198 COLOUR CODE O TRANSPARENT COLOUR CODE 1 BLACK COLOUR CODE 2 EDIUM GREEN ~ COLOUR CODE 3 LIGHT GREEN COLOUR CODE 4 DARK BLUE COLOUR CODE 5 LIGHT BLUE COLOUR CODE 6 DARK RED COLQURF CODE 7 CYAN COLOUR CODE 8 MEDIUM RED COLOUR CODE 93 LIGHT RED COLOUR CODE 10 DARK YELLOW COLOUR CODE 11 _IGHT YELLOW COLOUR CODE 12 DARK GREEN COLOUR CODE 13 MAGENTA COLOUR CODE 14 GREY COLOUR CQGDE 15 WHITE — ath How to Print in Multi-Colour Graphics MODE 3 In the multi-colour MODE 3, the text is four times the size of normal mode because the graphic display is in low resolution. Hence in order to PRINT characters, block graphics must be used, and the resulting Characters are huge. Here is a demonstration program. 10 SCREEN 3 20 OPEN "GRP:" AS #1 30 DRAW "“BMO,O0" 40 PRINT#1,"PI" SO DRAW "BMO,65" 60 PRINT#1, USING" #. ####8##" SATNC 1) #4 70 GOTO 70 LINE 10 SELECTS MULTI-COLOUR MODE LINE 20 OPENS A FILE TO THE SCREEN LINE 30 POSITIONS THE CURSOR LINE 40 PRINTS "PI" LINE 50 REPOSITIONS THE CURSOR LINE 60 PRINTS THE VALUE OF Pi LINE 70 LOCKS THE GRAPHICS SCREEN 199 CHAPTER 15 ADVANCED GRAPHICS IV SPRITE GRAPHICS The MSX's video display processor has a facility to produce sprites. Sprites are user-definable characters or shapes, which can be dis- played on the screen without affecting the backdrop graphics screen. This is because the sprites are displayed in different sprite screen planes. They can be made to hide behind one another and move about without causing flicker. It gives you power to create arcade games without too much trouble with the graphics. Sprite Size There are four sprite sizes you can use, but only one sprite size may be used at once. The size is determined using the SCREEN statement. SIZE SCREEN ,O 8by8 pixel unmagnified SCREEN ,1 8by8 pixel magnified to 16 x 16 SCREEN ,2 16by16 _ pixelunmagnified SCREEN 3 16by16__ pixel magnified to 32 x 32 Note that magnification gives you 2 x 2 pixel resolution. Defining Sprites: SPRITES You can define your sprite using SPRITES. You can have up to 256 sprite patterns when you are using sprite size 0 or 1 (8 by 8 pixel), or 64 of them in sprite size 2 or 3 (16 by 16 pixel). 200 Syntax Sprite$ (<interger>)=<string> <integer> = sprite pattern number <integer> = 0 to 255 when sprite size is 0 or 1 <integer> = 0 to 63 when sprite size is 2 or 3 The SPRITES is a string. The length of the sprite string is fixed to 32 bytes (or characters), but for small 8 by 8 sprites you only need to define the first 8 bytes. Each byte in the sprite string represents an 8 pixel row and is as- signed to the sprite string by adding a character with the ASCII code equal to the byte. This can be done simply by using the CHR&() function. For example, to assign an 8 by 8 sprite: SPRITE$(<integer>)=CHR$(<1st row>)+CHR§$(<2nd row>) +CHR$(<3rd row>)+CHR$(<4th row>)+CHR$(<Sth row>) +CHR$(<6th row>)+CHR$(<7th row>)+CHR$(<8th row>) The CHR§() function can be replaced by an actual character if you know what Is. To define a 16 by 16 sprite, you must define all 32 bytes in the sprite pattern. Here is a Standard way of defining an 8 by 8 sprite. First plot a drawing of your sprite on a piece of paper. 128 64 32 16 8 4 2 1 BINARY DECIMAL — To... =) 6&B00010000 = 16 1 1 . = &B00110000 = 48 a FP oh A . = &B01110000 = 112 1 4 40404 40°04 ~ °4 = &B11111111 = 255 1 1 4 40°49 ~° 4064 °~°400: = &B11191111 = 255 1 11 . = &B01110000 = 112 1 1 . = &B00110000 = 48 1 . = &B00010000 = 16 Then you can form a string expression as follows: SPRITES(0)=CHRS$(16) + CHRS(48) + CHRS(112)+CHRS$(255)+ CHR$(255) + CHRS(1 12) +CHR$(48) + CHR$(16) 201 EXAMPLE LINE 10 HIRES SCREEN, NORMAL SIZE SPRITE Using the binary numbers is the easiest way of defining a sprite because LINE 20 DEFINE SPRITE 0 it is easy to see what the sprite is going to look like in the program. LINE 70 PUTS SPRITE 0 AT x=100,y=100 COLOUR WHITE IN SPRITE PLANE 0 How to define a 16 by 16 sprite. SCREEN 2,0 128 64 32 16 8 4 2 1128 64 32 16 8 4 2 FOR I=1 Ta & 2 tS Wy. ye a, Fate a a) 1 READ At i 70) a. he Yl i: thea a 4 S$=S$+CHRS$(VAL C"&E"+AS) D 1 ey ee a NEXT I 1 Ah be a SPRITE#(0)=S% 1 1 14 PUT SPRITE 0,(€100,100),15,0 1 Gata 30 ae ane ee Sahoo es are — — 4 a =i: ek Re, eo REM BINARY DATA DATA 00010000 DATA 00110000 DATA 01110000 DATA 11111111 DATA 11111111 et ht aor ek Geet Cae ee at a 1 4 So ins te an, iis spate st Sa hy DATA 01110000 OR we eM ce ae De, 4 DATA 00110000 : : ‘ ; . ae 1 1 1 DATA 900010000 ; : . . ; ee ad 4 LINE 10 HIRES SCREEN, NORMAL SIZE SPRITE The diagram above in binary and decimal. LINE 30 READS THE BINARY NUMBER AS A STRING LINE 40 S$ISA TEMPORARY STRING 00011111 = 3 11111000 = 248 LINE 60 DEFINES SPRITE 0 00111111 = 63 11111100 = 252 LINE 70 PUTS SPRITE OAT x=100,y=100 COLOUR WHITE IN SPRITE PLANE 0 01100000 = 96 000001 10 = 6 11000111 = 199 11100011 S22: Result: you see an arrow in the middle of the screen. 11001000 = 200 00010011 = 19 F ; ; 01101000 = 104 00010110 = 22 We deal with the PUT SPRITE statement in detail, later. 01100100 = 100 00100110 = 38 If you could be bothered to calculate the decimal values of each byte, 00110014 = 5] 11001100 = 204 then the program can be written in a much shorter version. The above program can be reduced to the following few lines. ze 00110100 = 52 00101100 = 44 10 SCREEN 2,0 Set oa, een oe 20 SPRITE#(O)=CHR$¢(16)+CHRE(48) i 00001100 — 40 00110000 ss 48 +CHR# (112)+CHR$ (255) +CHR$ (255) ‘ 60007 100 ae 00110000 — 48 +CHR$(112)+CHR#(48)+CHRE( 16) 00000110 a6 01100000 = "96 70 PUT SPRITE 0,(100,100),15,0 00000011 _ 3 11000000 ~ 492 80 GOTO 80 00000001 =a 10000000 = 128 202 203 16 by 16 SPRITE patterns are stored in Video RAM in the following way: A 16 x 16 SPRITES = “Sprite quadrant 1” + “Sprite quadrant 2” + “Sprite quadrant 3” + “Sprite quadrant 4" Here is a fairly efficient program which defines and displays the 16 by 16 bit sprite we have designed above. The data is stored in DATA statements and are added one by one to the SPRITES(0) in the FOR/TO/ NEXT loop. 10 SCREEN 2,2 20 FOF I = 1 TQ 32 30 READ BZ 40 S#=S%+CHFE$ (BZ) 50 NEXT I 60 SPRITE$(O)=St 70 PUT SPRITE 0,¢€¢100,100),15,6 20 GOTO so 90 DATA 31,63,96,199,200,104,100,51 100 DATA 52,27,24,12,12,6,3,1 105 DATA 242, 252,6,227,19,22, 38,204 110 DATAd4, 216, 24,48,48,96,192,128 LINE 10 HIRES SCREEN MODE 2 WITH 16 BY 16 SPRITES SIZE LINE 20 LOOP TO READ IN 32 BYTES LINE 30 READ LINE 40 ADDS DATATO S$ LINE 60 DEFINE SPRITE STRING 0 LINE 70 PUTS SPRITE O ON PLANE 0 IN WHITE 204 LINE 90 DATAFOR QUADRANT 1 LINE 100 DATAFOR QUADRANT 2 LINE 105 DATAFOR QUADRANT 3 LINE 110 DATA FOR QUADRANT 4 How to put a Sprite onto the Screen: PUT SPRITE There are 32 sprite planes in front of the text/graphics plane, with the sprite plane No 0 right at the very front. This means that the smaller the plane number, the higher the priority. If there are two sprites over- lapping, the one with the smaliest sprite plane number will be displayed in front, with the other hidden behind. You may define up to 256 sprite patterns, but only one sprite can be displayed on any particular plane. This limits the maximum number of sprites displayed at any one time to 32. Also, you can only place a maximum of four sprites per horizontal line. To place a sprite on to the screen, use the PUT SPRITE statement. It has the following syntax: PUT SPRITE <sprite plane number>[,<coordinates specifier>] [,<colour>][,<pattern number>] The <sprite plane number> has a range of 0 to 31. The <coordinates specifier> tells the computer where to place the top left corner of the sprite. There are two <coordinates specifier> formats: 1) (<x-—coordinate>,<y—coordinate>) These specify an absolute position on the screen. 2) STEP (<x—offset>,<y—offset>) These coordinates are of a point relative to the last point referred to. <x-—offset>, <y—offset>, <x--coordinates>, <y—coordinates> can be variables, or expressions, as well as just simple numerical constants. This means that the sprites can be controlled by variables allowing you to move them smoothly. If <coordinates specifier> is omitted, then the computer will PUT SPRITE at the last point referred to. Examples PUT SPRITE 0,(50,50),1,1 PUT SPRITE 1,STEP (10,10),12,2 205 DISPLAY SCREEN SPRITE SCREEN 4) The Sprite Screen 240°". 2G ; 255 Stead er toe X ACTUAL DISPLAY SCREEN 206 The sprite screen is slightly larger than the actual display screen. Ithas a range between —32 and 255 in the X-coordinate and between —65 and 248 in the Y-coordinate. The actual screen coordinates ranges are smaller, so if the sprite is placed outside the actual screen display, then that sprite will be either partially or completely hidden. The sprites are provided with a complete screen wraparound facility. This means that if a sprite goes beyond the sprite screen, it will appear again on the opposite side of the screen. How to Move Sprites By altering the values of the coordinates specifier in the PUT SPRITE statement, you can move the sprites to anywhere on screen. When the PUT SPRITE statement is executed, the previous sprite in the old loca- tion is erased automatically. This example program displays a square Moving slowing from right to left, going beyond the screen to the left, and emerging from the right. 10 SCREEN 2,0 20 SPRITE$(0)=STRING$(8, CHR$( 255) 30 FOR X=200 TO —200 STEP -1 40 PUT SPRITE 0,¢X,100),1,0 50 FOR D=1 TO SO:ZNEXT 60 NEXT 70 END — LINE 10 HIRES SCREEN, NORMAL SIZE LINE 20 DEFINES A SQUARE SPRITE LINE 40 PUTS THE SPRITE AT X.100 LINE 50 DELAY Sprite Colour You can use any of the 16 colours. However, you can only use one colour per sprite, so it is not possible to have a multi-coloured sprite. To simulate a multi-color sprite, place two or more sprites with different colours on top of one another in different sprite planes and move them together. Note that the background colour of a sprite is always transparent so that you can see through the gap in the sprite. This example defines two sprites, 0 and 1. The sprite number 0 is displayed in white, and the sprite 1 is in dark yellow. On the display you will see each of the sprites separately, as well as the combination of the two which simulates a multi-colour sprite. 207 10 SCREEN 2,0 20 SPRITE$(O)=CHR$( 16)+CHR$ (48) +CHR$(112)+CHR$(255)+CHR$( 255+ CHR$(112)+CHR$(48)+CHRS(16) 3O SPRITE$(1)=CHR$ (224 )+CHR$(192) +CHR$ (128)+CHRS (0) +CHRS$ (0) +CHRS ¢ 128)+CHR$(192)+CHR$ (224) 40 PUT SPRITE 0,(20,20),15,0 50 PUT SPRITE 1,(40,40),10,1 40 PUT SPRITE 2,(60,60),15,0 dO PUT SPRITE 3,¢€60,60),10,1 80 GOTO 80 been | LINE 10 HIRES SCREEN, NORMAL SIZE LINE 20 DEFINES SPRITE 0 LINE 30 DEFINES SPRITE 4 LINE 40 PUTS SPRITE OIN WHITE LINE 50 PUTS SPRITE 1 INDARK YELLOW LINE 40 PUTS SPRITE 2 AND 3 ON THE SAME COORDINATES USING 2 COLOURS IN DIFFERENT SPRITE PLANES How to Hide Sprites If you give a special number to the y-coordinate, you can disable sprites: Y=208 If 208 (&HDO) is given to the Y coordinate, ail sprite planes with higher sprite plane numbers are disabled until a value other than 208 is given to that plane. 10 SCREEN 2,0 20 SPRITES(CO)=STRINGS$ (8, CHR#( 255) > LINE 50 Y=208 DISABLES SPRITE PLANE 2 ONWARDS. SPRITES AT LINE 60 AND 70 ARE NOT DISPLAYED If 209 (&HD1) is specified to Y, then that sprite disappears from the screen. | 10 SCREEN 2,0 | 20 SPRITE$(0)=STRING$(8, CHR$( 255) ) 30 PUT SPRITE 0,(20,20),15,0 40 PUT SPRITE 1,(40,40),15,0 50 PUT SPRITE 2,(60,209),15,0 60 PUT SPRITE 3,(80,80),15,0 | 20 PUT SPRITE 4,(100,100),15,0 80 GOTG 80 LINE 10 HIRES SCREEN, NORMAL SiZE LINE 20 DEFINES A SQUARE SPRITE LINE 30 SQUARE SPRITE ON PLANE 1 LINE 40 SQUARE SPRITE ON PLANE 2 LINE 50 Y=209 DISABLES THIS SPRITE LINE 60 SQUARE SPRITE ON PLANE 3 LINE 70 SQUARE SPRITE ON PLANE 4 The ‘Fifth’ Sprite Rule There is a maximum limit of four sprites which can be displayed on one horizontal tine. If this rule is violated, then only the four sprites with the latest sprite plane numbers will be displayed normally. The rest will not be displayed on that line. The sprite number of the violating fifth sprite can be found from the VDP status register using the VDP function (see VDP function section). Example PUT SPRITE PUT SPRITE PUT SPRITE PUT SPRITE PUT SPRITE GOTG 80 0,¢20,20),15,0 1,(40,40),15,0 2,(60,208),15,0 3,(80,80),15,0 4,(100,100),15,0 PUT SPRITE PUT SPRITE PUT SPRITE PUT SPRITE PUT SPRITE 10 SCREEN 2,0 | SPRITES(O)= STRING$(S, CHR$(255) ) 0,€20,100),15,0 1,(40,100),15,0 2,(60,100),15,0 3,(80,100),15,0 4,(100,104),15,0 LINE 10 HIRES SCREEN, NORMAL SIZE LINE 20 DEFINES A SQUARE SPRITE 208 80 GOTO 80 209 LINE 10 HIRES SCREEN, NORMAL SIZE LINE 20 DEFINES A SQUARE SPRITE LINE 30 SQUARE SPRITE ON LINE 100 LINE 40 SQUARE SPRITE ON LINE 100 LINE 50 SQUARE SPRITE ON LINE 100 LINE 60 SQUARE SPRITE ON LINE 100 LINE 70 SQUARE SPRITE ON LINE 104 In the above example, you will see four square sprites, on the same horizontal line, 100. The fifth sprite is on line 104 so it only shows half the square, the other half being hidden due to the fifth sprite rule. If it was on line 108, the entire fifth sprite would be visible. How to Animate your Sprites Simple sprite animation can be achieved with a minimum of fuss with the MSx. All you have to do is to swap two sprite patterns rapidly so the characters appear to be changing. Here is a simple program which shows a space invader with moving legs. The SPRITE$(O) contains the invader with legs closed, and SPRITE$(1) contains the one with legs open. = 10 SCREEN 2,1 20 SPRITE$(O)=CHR$(60)+CHRS$( 126) +CHR$(129)+CHR$(219)+CHR$(126) +CHR$ (36) +CHR$ (36) +CHRS$ (36) 30 SPRITE$(O)=CHR$(60)+CHR$( 126) +CHR$ (129)+CHR$ (219)+CHR$ (126) +CHR$(36)+CHR$( 60) +CHR$ (129) 40 PUT SPRITE 0,(100,100),11,0 50 FOR I=1 TO 5SOO:NEXT 60 PUT SPRITE 0,¢(100,100),11,1 70 FOR I=1 to SOO:NEXT 80 GOTO 4o LINE 20 INVADER WITH LEGS CLOSED LINE 30 INVADER WITH LEGS OPENED LINE 40 SPRITE O (LEGS CLOSED) YELLOW LINE 50 DELAY LINE 60 SPRITE 1 (LEGS OPENED) YELLOW LINE 70 DELAY 210 Demonstration Program In this example you will see two planets orbiting a Sun. The whole system drifts from the top left-hand-side of the screen. 10 SCREEN 2,0 20 COLOR 15,1,1 30 CLS 40 SPRITE$(O)=CHR$(126)+STRINGS$(6, CHR$( 255) >+CHR$ (126) 50 SPRITE$(1)=STRING$(3,CHR#(O) + CHR$ (24 )+CHR$ (24 )+STRING$(3, CHR «€O)) 60 FOR I=0 TO 6-28 STEP 0-2 7O X=X+1-5 80 Y=Y+1 90 X1=30#COS(I1) 100 Y1=30*SINCI) 110 X2=15*COS¢(I) 120 Y2=15*SINCI) 130 PUT SPRITE 0,(X,Y),11,0 140 PUT SPRITE 1,¢(X1+X,Y1+Y),9,1 150 PUT SPRITE 2,(X2+X,Y2+Y),15,1 160 NEXT 170 GOTO 60 | LINE 10 HIRES SCREEN, NORMAL SIZE BLACK BORDER AND BACKGROUND, WHITE FOREGROUND LINE 30 CLEARS SCREEN LINE 40 SPRITE OIS THE SUN LINE 50 SPRITE 11S A PLANET LINE 60 LOOP 70 MOVE SUN IN X BY 1.5 80 MOVE SUN INY BY 14 90 X1 = X-COORDINATE FOR PLANET 1 100 Y1=Y - COORDINATE FOR PLANE 14 110 X2 = X-COORDINATE FOR PLANET 2 Y2 = Y—-COORDINATE FOR PLANET 2 130 PUT SUN AT NEW POSITION 140 PUT PLANET 1 AT NEW POSITION 150 PUT PLANET 2 AT NEW POSITION 160 NEXT LOOP 170 LOOP AGAIN f Z2aZze222 mmmmmmmm mmm nmo = nN oO f LLL 211 How to Detect Sprite Collisions The TMS 9918/9929 VDP can detect sprite collisions. This is a useful facility for those arcade games which involve shooting or avoiding the enemy. There are two BASIC statements which deal with sprite collision detection — ON SPRITE GOSUB and SPRITE ON/OFF/STOP. The ON SPRITE GOSUB statement sets the sprite collision interrupt subroutine. It tells the computer that the sprite collision subroutine is at the specified line number. The SPRITE ON statement switches on the actual collision detection, which diverts the program to the subroutine specified by the ON SPRITE GOSUB statement when two sprites collide. You must execute the SPRITE ON statement to start tne detection. lf any one pixel of two sprites overlaps, that is considered as a collision. This happens even if the overlapping pixels are transparent. Although the computer can detect a sprite collision, it Cannot tell you which two sprites have collided or even where the collision took place. You must program the computer to work this out. Once the interrupt occurs, e. two sprites collide, an automatic SPRITE STOP is executed. This stops the computer calling the sub- routine again, if another collision occurs while the computer is executing the current sprite collision subroutine. However, it remembers if there is another sprite collision during the current Subroutine, in which case the computer will immediately go to the subroutine again after it has left the current subroutine, unless the current subroutine disabies the sprite collision detector altogether by executing SPRITE OFF. After it leaves the sprite subroutine the computer will automatically execute SPRITE ON to enable the interrupt, unless SPRITE OFF is executed in the subroutine. The SPRITE OFF statement completely stops the computer from detecting any sprite collision. Example In this example, you will see two square sprites, one yellow and another white, approaching each other from opposite sides of the screen. When they collide, ON SPRITE GOSUB will come into effect and make the BASIC jump to the sprite collision subroutine. The SPRITE OFF in the sprite interrupt routine prevents any further detection of sprite collisions. This is included because the two sprites will still be overlapping after the computer executes the subroutine. If SPRITE OFF was not there, the computer will eternally loop back to the sprite subroutine. (Try this routine without SPRITE OFF, and see what happens.) 212 10 ON SPRITE GOSUB 110 20 SCREEN 2,0 30 SPRITE#(0)=STRING$(8, CHR$( 255) ) 40 SPRITE#$(1)=STRING$(8, CHR$(255) > 5O SPRITE ON 60 FOR I=10 TO 240 70 PUT SPRITE 0,¢1,100),11,0 80 PUT SPRITE 0,(250-1,100),15,0 90 NEXT I 100 GOTG 50 105 REM SPRITE COLLISION ROUTINE 110 SPRITE OFF 120 BEEP | 150 RETURN 10 SETS TRAP SUBROUTINE TO 110 20 SETS GRAPHICS SCREEN 30 SPRITE 01S A SQUARE 40 SOISSPRITE 1 50 SPRITE COLLISION DETECTOR ON 60 LOOP SPRITE (YELLOW) MOVES FROM LEFT 80 SPRITE (WHITE) MOVES FROM RIGHT 90 NEXTLOOP 100 RESTART THE LOOP 110 KILLS TRAP (ONCE IS ENOUGH) 120 BEEP INDICATOR 130 GO BACK TO THE LOOP [ Zz mmammmmmmmmmmm ~ Oo f Zaz Notes on Sprite Collisions The computer looks for a sprite collision every time the computer executes a statement. Therefore one cannot say that the computer looks for a sprite collision in any particular time interval. The VDP function can give you the status of the Sprite collision flag in register 8 of the Video Display Processor. (See the VDP section.) 213 CHAPTER 16 ADVANCED GRAPHICS V HOW TO ACCESS THE VIDEO DISPLAY PROCESSOR VDP Function The VDP TMS 9929A has 8 write-only registers, and 1 read-only register. You can access these VDP registers from BASIC, using the VDP function. For the write-only registers, VDP(O) to VDP(7), you can only assign values to them, i.e. VDP(1)=2. There is no way of directly reading these registers from the VDP. However, the computer keeps a copy of these registers in its systems work area. They are stored in memory location between ROGSAV and RG7SAV. (See the section on System RAM in the BIOS reference section for details.) VDP(8) is the read-only status register. You may read this register as below: PRINT VDP(8) You should remember that if you assign a wrong number to these registers, the screen can sometimes break up. If this does happen, you should RESET/SWITCH OFF the computer to initialise the VDP. Always make sure you know what you are doing before you use the VDP function. VDP Registers WRITE ONLY REGISTERS RegisterO....... VDP(0) The register 0 contains two VDP option control bits, BIT 1 (M3) and BIT 0 (EV). BIT 2 to BIT 7 must be zero. 214 BITOEV: external VDP. It enables/disables input from an external VDP. EV=0 for most MSX. O = disable 1 = enable BIT 1 M3: mode bit 3. M3 is used to select the screen mode and is used in conjunction with register 1. (see register 1) Register1..... VDP(1) Register 1 contains 7 VDP option control bits. BIT 2 is reserved and should always be zero. BITQMAG: magnification option for sprites. OQ = unmagnified 1 x 1 pixel resolution | = magnified 2 x 2 pixel resolution BIT 1 SIZE: size selector for sprites 0 = 8 x 8 pixel sprites 1 = 16 x 16 pixel sprites BIT 2 Reserved = 0 BIT 3 M2: mode bit 2. M2 is used to select the screen mode. (See Bit 4) BIT 4 MI: mode bit 1. M1 is used to select the screen mode. Com- bination of M1, M2, and M3 gives the screen mode. M1 M2 M3 6) 6) 0 Text mode 1 re) 0 1 High resolution graphics mode 2 6) 1 0 Multi-colour graphics mode 3 1 6) 6) Text mode 0 BIT 5 \E: Interrupt enable OQ = disables VDP interrupt 1 = enables VDP interrupt BIT 6 BL: Blank enable/disable. This allows you the switch off the display. O = causes the screen display to blank and show the border colour only 1 = enables the screen display 215 BIT 7 VRAM: selects what kind of video RAM is used. (VRAM=1 for most MSX) 0 = 4K RAM 1 = 16K RAM R1 Register2..... VDP(2) The register 2 defines the base address of the Name Table. Its range is between 0 and 15. This register represents the upper 4 bits of the 14 bit Name Table address. Therefore, you should write in the address divided by &H400. R2 B7 B6 B5 B4 B3 0 0 0 NAME TASLE ADDRESS NOTE: R2 * &H400 = NAME TABLE ADDRESS Register3..... VDP(3) The register 3 defines the base address of the Colour Table. Its range Is between 0 and 255. This register represents the upper 8 bits of the 14 bit Colour Table address. Therefore, you should write in the address di- vided by &H40. gaoOooomo COLOUR TABLE ADDRESS NOTE: R3 * &H40 = COLOUR TABLE ADDRESS Register4..... VDP(4) The register 4 defines the base address of the Pattern, Text or Multi- colour Generator Table, depending on the screen mode. Its range is between 0 and 7. This register represents the upper 3 bits of the 14 bit Pattern/Text/Multi-colour Generator Table address. Therefore, you should write in the address divided by &H800. 216 PAT GEN ADD NOTE: R4 * &H800 = PATTERN/TEXT/MULTI-COLOUR GENERATOR TABLE ADDRESS Register5..... VDP(5) The register 5 defines the base address of the Sprite Attribute Table. Its range is between 0 and 127. This register represents the upper 7 bits of the 14 bit Sprite Attribute table address. Therefore, you should write in the address divided by &H80. B1 SPRITE ATTRIBUTE TABLE BASE ADD 4 NOTE: R4 “&H80 = SPRITE ATTRIBUTE TABLE ADDRESS Register6..... VDP(6) The register 6 defines the base address of the Sprite Pattern Generator Table. Its range is between 0 and 127. This register represents the upper 3 bits of the 14 bit Sprite Pattern Generator Table address. Therefore, you should write in the address divided by &H800. R6 | B7 | B6 BS B4 B3 a2 | B1 | BO | 0 0 0 6) 0 SPR PAT GE.ADD NOTE: R6 *&H800=SPRITE PATTERN GENERATOR TABLE ADDRESS Register7..... VDP(7) The register 7 is divided into two halves. The upper 4 bits contain the foreground colour of text mode, while the lower 4 bits contain the background colour for text and graphics modes. Range for the foreground colour is between 0 and 15. Similary, for the background colour. R7 - ~.Foreground colour> * &H10 + <background colour> 217 R7 FOREGROUND COLOUR BACKGROUND COLOUR READ ONLY REGISTER The Status Register... .. VDP(8) The Status flag contains the interrupt pending flag, the sprite collision flag and the fifth sprite flag. Interrupt flag F : BIT 7 The Interrupt flag is set to 1 at the end of raster scan of the screen display. It is reset to a O after the Status register is read, or when the VDP is externally reset. Sprite Collision Flag C : BIT5 The VDP checks if two or more sprites overlap each other by more than one pixel. This is done every 1/50th of a second. 1 = sprite collision 0 = No overlap Fifth Sprite Flag 5S : BIT 6 5S ts set to 1 when there are more than 4 sprites on a horizontal line (lines 0 to 192). Note that the VDP cannot cope with more than 4 sprites on a single line. If there are 5 or more sprites, the fifth sprite will not be displayed. Fifth Sprite Number if there are more than 4 sprites on a horizontal line and 5S is set to 1, then the sprite number of the violating fifth sprite is given in the Fifth Sprite Number. R2 F 58 FIFTH SPRITE NUMBER 218 CHAPTER 17 ADVANCED GRAPHICS VI THE VIDEO RAM The MSX has 16K of Video RAM which is separate from the main programming memory. It is managed and refreshed by the Video Dis- play Processor and therefore, it is independent of the Z-80 central processor. (Dynamic RAMs used by MSX must be refreshed so that they don't loose their stored data.) This gives the following advantages: 1) The Z-80 processor does not have to refresh the RAM for the screen display thus saving time. 2) The graphics and display does not use up precious memory from the main RAM which is managed by the Z-80. The Video RAM is divided up into various sections, each with its unique function. It is quite possible to access the Video RAM from BASIC using the following statements and functions: BASE Gives base address of various tables in Video RAM VPEEK Reads content of Video RAM VPOKE Writes to the Video RAM The BASE function gives the current base address position of the various screen tables in the Video RAM. (See BASE for more explicit details.) 219 TABLE NAME AND TEXT GRAPHICS |GRAPHICS Sprite attribute table Sprite generator table LOCATION IN HEX MODE 0 | MODE 1 Pattern name table &HO000 | &H1800 | &H1800 &HO800 Colour table &H2000 | &H2000 Pattern generator table] &HO800 | &HOO0O &HOO000 | &HOO00 &H1BO00 | &H1B00 &H3800 | &H3800 &H1B00 &H3800 1 220 CHAPTER 18 ADVANCED SOUND EFFECTS USING PSG MSX’s sound generator processor, AY-3-8912, has been around for quite a long time. It was developed by General Instruments in the late ‘70s, and it is still the most popular sound chip around. It is used widely in small 8 bit computer systems and arcade games machines. The reasons for its success are its flexibility, and its ability to generate up to 3 channels of periodic sound and one white noise channel simul- taneously. Also, because it is a microprocessor on its own, when the sound is generated the processor does not halt the central processor's operation. Although it is now slightly out of date, there is alot you can do with this sound generator. This section deals with the workings of the AY-3-8910 sound genera- tor and the use of the SOUND statement. As an example, try out the following short program which generates a weird flying-Saucer noise. 10 SOUND 0,87 20 SOUND 7,62 30 SOUND 8,16 40 SOUND 11,179 50 SOUND 12,45 60 SOUND 13,14 Wave Forms generated by PSG Basically, the PSG can only generate two types of wave form: a periodic square wave and white noise. 221 periodic wave example white noise example 10 SOUND,7,62 10 SOUND 7,55 20 SOUND 8,15 20 SOUND 8,15 30 SOUND 1,1 40 SOUND 0,28 . However, these sound waves can be shaped and modified into complex noises so that they do sound life-like. To do that you have to control the frequency, loudness, tone and envelope of each sound channel using the SOUND statement which accesses the PSG. PSG Sound Registers AY-3-8910 PSG has 14 registers which you can write to, using the SOUND statement. The SOUND statement has the following syntax: SOUND <register-number>,<value to be written> By writing suitable values to these registers, you can select channels, sound waves, tone frequency, noise, and loudness. There are 14 registers, and they have the following functions: Register Description Fine tune frequency on channel A Coarse tune channel A Fine tune frequency on channel B Coarse tune channel B Fine tune frequency on channel C Coarse tune channel C Frequency of the noise generator Mixer switches on/off channels A, B and C. Volume control for channel A Volume control for channel B 10 Volume control for channel C a Envelope period (low) 12 Envelope period (high) 13 Envelope pattern CHOOSING SOUND FOR EACH CHANNEL: REGISTER 7, THE MIXER This is the register which switches the sound on/off for all three chan- nels. It can select the source of the sound wave, from the noise genera- tor, or from the individual tone generator for each channel. Bits 5, 4, and 3 of register 7 switch on the noise generator for C, B, and Achannels respectively. 1 means off and 0 means on. Bits 2, 1, and 0 of register 7 switch on the tone generator for channel C, B, and A respec- tively. So, for example, if you want channel C to emit a tone, bit 2 must be set to 0 OMOnNANFWM-OC 222 OFF | OFF | OFF | ON | OFF | OFF i L The SOUND statement format for the above Is: SOUND 7,&B00111011 , or in decimal, SOUND 7,59 The tone of channel A depends on registers 0 and 1, but more on this later, if you want channel B to emit white noises, and C and Aas above, then the register 7 changes to: The SOUND statement format for the above is: SOUND 7,&B00101011 , or in decimal SOUND 7,43 You can make a channel emit both white noise and tone if you switch both on. Say you want all channels to emit both tone and noise, then you get: 223 / / C B A C B A $+ te) 0) 0 6) QO 0) 0 0 L ON ON ON ON ON ON The SOUND statement format for the above is: SOUND 7,&B00000000 , or in decimal SOUND 7,0 NOTE: Bit B7 and B6 are not used, and therefore should be set to 0. TONE GENERATOR: REGISTERS 0,1,2,3,4,5 To set the frequency of the tone generator for each channel, you write to registers 0,1,2,3,4,5. Each channel has two registers for fixing the frequency of the tone to be generated. Register 0 and 1 for channel A. Register 2 and 3 for channel B. Register 4 and 5 for channel C. Register Range Description 0-255 Fine tune frequency on channel A O-15 Coarse tune channel A 0-255 Fine tune frequency on channel B 0-15 Coarse tune channel B Q-255 Fine tune frequency on channel C 0-15 Coarse tune channel C 0-31 Frequency of the noise generator Frequency step (f) is calculated as follows: f = (Register 0) + (Register 1) x 256 This formula does not give you the frequency in Hz., but the higher the frequency number the lower the tone generated. Therefore, the highest frequency is generated by SOUND 0,0: SOUND 1,0 which gives f=0; and the lowest frequency is given by SOUND 0,255: SOUND 1,15. To calculate the frequency generated in Hz. use the following formula: 178900 178900 is the clock frequency f = ———— of PSG (which may be different (16 x Hz) for the U.K MSX model) NoOkRWNM-O 224 if you want to be exact about what frequency you want the PSG to generate, you must do the above calculation. No! not with a calculator; use the computer instead. 10 INPUT “FREQUENCY IN HERTZ";HZ 20 F=INT (1789800#/(16*HZ)) 30 PRINT “FINE TUNE REGISTER 0, OR 2, OR 4";F MOD 256 40 PRINT “COURSE TUNE REGISTER 1, OR 3, OR 5 "sINT(F/ 256) The above program should do the trick. Example: Generate 2000Hz sound wave from channel A. f=1789800/(16*2000)=55.9 10 SOUNDO,55 20 SOUND 1,0 30 SOUND 7,62 mixer 40 SOUND 8,10 loudness NOISE GENERATOR FREQUENCY: REGISTER 6 To change the frequency of the white noise generator, you write to the PSG register 6. The range is between 0 and 31: the lower the value, the higher the frequency. Example: 10 SOUND 7,55 channel A noise on 20 SOUND 8,15 loudness on channel A set to 15 30 FORF=0TO 31 40 SOUND 6,F 50 FORI=1to200:NEXT Delay 60 NEXTF You should hear a gradual decrease in the frequency of the white noise generated. LOUDNESS: REGISTERS 8, 9, 10 The loudness of the sound emited from channels A, B, and C are controlled by registers 8, 9, and 10 respectively. nosound maximum volume OQ <------------------ >15 REGISTER 8 = CHANNELA REGISTER 9 = CHANNEL B REGISTER 10 = CHANNEL C 225 When 16 Is written to these registers, the envelope is enabled. This means that the loudness changes according to the envelope shape and frequency. The envelope Is determined by registers 11, 12, and 13. You should remember that the volume output from a television speaker depends on the volume control on the TV. If the volume control is turned off, you won't hear anything. The volume of sound from the MSX's audio output also depends on the external amplifier. HOW TO USE ENVELOPES: REGISTERS 11, 12, AND 13 When envelope is not used, the volume of sound generated stays constant at the level fixed by the registers 8, 9, and 10. This is rather boring, as all you hear is a constant pitch of noise. However, Ay-3-8910 PSG has an enveloping facility which can change the voiume period- ically according to one of the envelope shapes available. Envelope is enabled by setting volume control registers 8, 9 or 10 to 16. When the envelope is switched on. the volume of sound changes according to registers 11, 12, and 13. ENVELOPE OFF ENVELOPE ON CHANNEL AREGISTER 8 0-15 16 CHANNEL BREGISTER 9 0-15 16 CHANNEL C REGISTER 10 0-15 16 The shape of the envelope pattern can be selected by setting register 13 to anumber between 0 and 15. However, the PSG duplicates some of the patterns, so you only get 8 different enveiopes. The rate of change of volume or envelope period, can be changed using registers 11 and 12; register 12 gives a coarse period change, and register 11. a fine period change. range register11 O-255 register 12 0-255 Period = (register 12) + 256 * (register 11) Note that you can have only one envelope at a time, but all three channels may use the same envelope. See SOUND in the BASIC reference section for the envelope shapes. 226 CHAPTER 19 HOW TO USE FILES In MSX BASIC it is possible to save data, such as the contents of a string array, to cassette, and conversely to read in data. This set of data is called ‘a file’. There are a set of BASIC statements and functions which enable you to write to and read from the device enabled. Maxfiles Before you actually use a file, it is a good idea to increase MAXFILES, which is the function used to set the maximum number of files allowed at any one time. The default value of MAXFILES is 1 but this is usually not enough since you can only use the cassette as a filing device when MAXFILES is 1. The maximum number for MAXFILES is 15. Technically speaking, MAXFILES has the effect of increasing the size of the File Control Block in the memory. The File Control Block is used as a work area by the MSX BASIC ROM when you are using files. (See Memory Map. CHAPTER 20.) OPEN To tell the computer that you are starting to read or write files, you must use the OPEN statement to declare a specified device opened. It allocates a buffer in the File Control Block and sets the mode of |/O operation for the buffer. You need to execute OPEN before the following statements: PRINT# PRINT# USING INPUT# LINE INPUT# INPUTS(#) EOF OPEN °<device name>[<file name>] [FOR <mode>] AS [#] <file number> 227 There are four devices supported in the current version of MSX. They are: CAS: cassette CRT: text screen GRP: Graphic screen LPT: line printer Of the above, only CAS: and GRP: have any real use. CAS: Is speci- fied when you are using a cassette tape recorder to store or retrieve data, and GRP: is specified when you are trying to write to a Graphics screen. It is not possible to PRINT in screen MODE 2 and 3, so you have to send information to the Graphics screen as a file. A detailed descrip- tion of how to use files to print to a graphics screen is given in CHAPTER 14. There are three modes of I/O: OUTPUT Specifies sequential output mode, i.e. writing to a device using PRINT# and PRINT# USING. INPUT Specifies sequential input mode, i.e. reading from a device using INPUT#, etc. APPEND Specifies sequential append mode. <file number> is an integer whose value is between one and MAXFILES, the maximum number of files. File numbers are given so that when quoted in PRINT# statements, for exampie, the computer knows which device to read or write to. The <file number=> is associated with the file as long as it is OPENed. It must not be one which is already being used in the program. Example Let us say we want to OPEN a file to a cassette to write information to it. The OPEN statement will be as follows: OPEN “CAS:INFO” FOR OUTPUT AS #1 CLOSE The opposite function of OPEN is performed by the CLOSE statement. It is a good idea to close a file as soon as you have finished with it. PRINT # To write to various devices, we use a special print statement, PRINT#. The “#" sign is followed by the file number which is specified in the OPEN statement. When used with the cassette as the device, PRINT# has an effect of recording the information given. 228 PRINT# USING This ts a file version of PRINT USING and has the same effect. Its main use Is in PRINTing to the Graphics screen in a specified format. INPUT# To read in data from devices other than the keyboard, such as a cassette, we use the INPUT# statement. It has the same effect as the INPUT statement but instead of typing in the information required from the keyboard, it reads in from the device specified by the OPEN state- ment; in most cases the device is the cassette. During the input for numeric variables, leading spaces, carriage returns and line feeds are ignored. The same goes for string variables. Quotation marks are also ignored. INPUT$(#) LINE INPUT# These are the file versions of INPUT$ and LINE INPUT statements, respectively (see INPUT#). EOF EOF returns whether the end of file has been reached. If the end of file is detected EOF returns —1, otherwise it is always 0. Example Program Here is a demonstration program which should give you a rough idea of how to use files. The program requires you to input five names, and it stores these into a cassette file. Then it asks you to load it back again. 229 DIM N#$C5)> MAXFILES = 4 FOR I= 1 T05 INPUT "PLEASE INPUT A NAME";N% NEXT I PRINT "IS CASSETTE ON? (y=O0K)>" IF INKEY$<>"y" THEN 70 GPEN "CAS:" FOR OUTPUT AS #2 FOR I =1 7T05 PRINT#2,N$(I) NEXT I CLOSE#2 PRINT “READY TO LOAD? (y=OK)" IF INKEY$<>"y" THEN 140 OPEN "CAS:" FOR INPUT AS #3 FOR I = 1 7T05 INPUT#3,NSC(I) NEXT I CLOSE #3 FOR I=1 TO 5 PRINT N$C(I) NEXT I ARRAY FOR NAMES INCREASE THE MAXIMUM FILE NUMBER (INPUT FIVE NAMES SO WE CAN DO SOMETHING WITH THEM CASSETTE ON (RECORD) WAITS FOR CASSETTE TO BE ON OPENS FILE TO CASSETTE OUTPUT TO CASSETTE USING PRINT# 120 THEN CLOSE FILE WHEN FINISHED 130 REWIND CASSETTE SO WE CAN LOAD THE DATA SAVED. (PLAY) 150 OPEN FILE FOR INPUT 160 READS THE CASSETTE FILE USING INPUT# 190 CLOSE FILE WHEN FINISHED 200 PRINTS OUT THE RESULTS 230 RUN PLEASE PLEASE PLEASE PLEASE PLEASE INPUT INPUT INPUT INPUT INPUT A A A A A IS CASSETTE ON READY TO LOAD Cy=OK) TOM PAUL BELLA JANE CHRIS Ok NAME 7 NAME 7 NAME 7 NAME ? NAME ? Cy=0N) TOM PALIL BELLA JANE CHRIS PRESS RECORD REWIND AND THEN PRESS PLAY 231 CHAPTER 20 MEMORY MAP &HFFFF : SYSTEM WORK AREA &HF380 FILE CONTROL BLOCK STRING AREA STACK AREA FREE AREA DIM SPACE VARIABLE BASIC PROGRAM AREA &H8000 &HCO0O00 for 16K MSX MSX BASIC ROM &HOO000 232 Work Area The WORK AREA is situated between memory location &HFFFF and &HF380 and is used by the MSX BASIC ROM for its internal operations. This area holds the various system variables, a list of which is given in the BIOS reference section. File Control Block This area is reserved for the input/output operations using files. State- ments such as PRINT#¥ and INPUT#¥ use this block. The size of the File Control Block is altered using the MAXFILES function. The upper limit of the File Control Block is &HF380. However, you can lower this limit to create a memory area to store machine code programs and data. This is done by executing a CLEAR statement which lowers this limit. Example: Create a users’ machine code area between &HF380 and &HFOOOO: CLEAR ,&HFOOO FF pane SYSTEM WORK AREA &HF380 USERS’ MACHINE CODE AREA SHEGOO FILE CONTROL BLOCK String Area This area stores the contents of the string variables. The default size of the String Area is 200 bytes. This sets the limit of the maximum length of a string to 200 characters. However, the size can be increased using the CLEAR statement. Example: Increase the String Area to 255 bytes: CLEAR 255 Stack Area This stores the stacks for FOR/NEXT loops and GOSUBs. It stores the addresses of where the BASIC must return to at a NEXT or RETURN statement. 233 Free Area This is the unused area. Its size can be found using the FRE function as follows: PRINT FRE(O) 28815 for 32K or 64K MSX systems at the power up. 12431 for 16K MSX systems at the power up. Note that this area decreases as you increase the size of BASIC pro- gram and variables. DIM Space This area stores the DIM array variables and is increased upon execu- tion of aDIM statement. If an array is a string array, pointers to the string area will be stored. Variable Area This area stores numerical variables and a string descripter pointer to the string area will be stored here. The address of where the data of a particular variable is stored can be found by using the VARPTR function. (See VARPTR in BASIC REFERENCE section.) BASIC Program Area Your BASIC program will be stored here. 234 CHAPTER 21 USR FUNCTION AND MACHINE CODE To execute machine code subroutines from BASIC, use the USR func- tion. The USR function calls a machine code program at a location specified in the DEF USR statement. You can use up to 10 machine code routines simultaneously. It is possible to have more than ten of course, by redefining the USR function. How to Define the USR Function To call a machine code routine, the computer must know where the start address of the routine is. Use the DEF USR statement to assign the address to a USR function. Example: say a machine code subroutine starts at &HFOOO DEF USRO=&HFOOO (you may abbreviate 0 so DEF USR=&HFOO00) How to Execute a Machine Code Routine The USR function can be used in the same manner as any ordinary function. This means that they can be in an expression, such as below: X=USR(99) Y$="M/C” +USR9(“abx") PRINT USR7(0) The above executes the subroutine and returns a value depending on what the machine code does. The numerics or strings in the brackets are parameters to be passed to the machine code. Hence the way you use the USR function really depends on your machine code routine. You 235 can have an argument in it, or just adummy parameter; it can also return a parameter or nothing. If you just want to execute machine code without entry parameters or output, then use a dummy such as below: DUMMY =USRO0O(0) where DUMMY is a dummy variable and (0) is adummy parameter. How to Pass a Parameter to a Machine Code Routine from BASIC You may pass any type of parameter to machine code from BASIC using USR<number>(<parameter>). The parameter must be enclosed in a set of brackets. When MSX executes the function it checks the kind of argument used, i.e. whether the parameter is integer, single precision number, double precision number, or a string. if there is a parameter, the machine code must know what type it is and where it is located in memory. MSX BASIC sorts out the parameter type before the machine code is executed. To find out what type of parameter was passed, look at location &HF6E63: &HFE63 = 2 = 00000010 integer parameter &HF663 = 4 = 00000100 single precision parameter &HF663 = 8 = 00001000 ~— double precision parameter &HF663 = 3 = 00000011 string Location of the parameter passed. Integer (&HF663=2) &HF7F8 = <lower byte> &HF7F9 = <higher byte> <integer parameter> = <lower byte>+256*<high byte> Single Precision Number (&HF663=4) - &HF7F6 = ] &HF7F7 = ] data stored in binary &HF7F8 = ] coded decimal from &HF7F6 &HF7F9 = ] Double Precision Number (&HF663=8) &HF7F6 = ] &HF7F7 = &HF7F8 = ] &HF7F9 = ] data stored in binary coded &HF7FA = ] decimal from &HF7F6 &HF7FB = ] &HF7FC = ] &HF7FD = ] 236 String (QHF663=3) &HF7F8 = ] low | . &HF7F9 = ] high ].... <address1> address of string description block. <address1> = string length <address1>+1 = low |] <address1>+2 = high ] ....<address2>, string location. How to Return a Parameter from Machine Code To return a parameter from machine code to BASIC, set address &HF663 to the appropriate value according to the type of data and store your parameters into the relevant addresses before you exit machine code. The BASIC picks up whatever is in the parameter store and transfers it to a variable after leaving the machine code routine. 237 CHAPTER 22 MSX MEMORY MANAGEMENT AND CARTRIDGE SLOT MECHANISM Introduction MSX’'s CPU, the Z80 microprocessor, has a 16 bit address bus which can address up to 64K bytes of memory at any one time. 64K byte of memory is divided into a 4 16K ‘pages’ of memory. MSX can expand its memory by switching to different ‘slots’ for each page. A ‘slot’ is basi- cally a64K memory space and can be thought of as a bank of separate memory. Physically, a slot can be a cartridge slot or a hardwired bank of memory like the MSX BASIC ROM with its system's RAM. To select which slot is used for which page, the slot-select register has to be set. More details on this later. Usually a basic MSX has the following memory configuration. 238 T ; SLOT 1,20R3 SLOT O CARTRIDGE SLOT SYSTEM SLOT TO BE PLUGGED IN oa ~| 16 K RAM for 16 K MSX computer _ | ROM cartridge for games page 16 K RAM for software 8 or 16 K ROM, 2 32 K MSX computer or 16 K expansion RAM for 16 K MSX system page Used for BASIC expansion 1 MSX BASIC ROM ROM or disc operating ROM or other languages. | MSX BASIC ROM (BIOS) 4 Cartridge All MSX computers have at least one MSX standard cartridge slot where you can plug in a variety of devices. Here is a list of possible cartridge devices: 1. Expansion RAM. Usually to expand a 16K machine to 32K. 2. Games program cartridge ROM. The cartridge contains either a machine code or a BASIC program. 3. BASIC extension ROM. This ROM contains routines to expand the MSX BASIC. The ROM routines are accessed using CALL state- ments. Some cartridges have their own RAM as a work area. 4. |/0 device cartridge. This can be a floppy disc controller or printer interface or light pen cartridge. It usually comes with its own utility ROM to control the input/output. Any of the above can be plugged into any slot. MSX has a slot-select mechanism so it knows which cartridge to access. Basic Slot Arrangement MSX can have four basic slots, slot number 0 being the one with the MSX BASIC ROM, and is called the system slot. Each of these 4 slots can be expanded to 4 expansion slots. Therefore, the total number of slots expandable is 16. These 16 slots can all be 64K of RAM giving you 1M 239 byte of memory space. This is the maximum RAM MSX can handle. However you should be reminded that you cannot access all 1 M byte from BASIC! You need either MSX-DOS or machine code program to make effective use of memory space larger than 64K. Basic slot configuration SLOTO SLOT1 SLOT2 SLOT3 FFFF : | PAGES | | Paces | paces | | PAGES | C000 t meh BFFF PAGE2 PAGE2 PAGE2 PAGE2 | 8000 7FFF PAGE1 PAGE1 | PAGEI | PAGE1 116K 4000 3FFF 16K 0000 Expanded slot configuration SLOT O - EXPANDED ~ SLOT EXPANDED a nN Slot Selector ee es) i ee SLOT 2 EXPANDED ie) es 5 eed | ees ee SLOT 3 EXPANDED _j]-+ a= ee | RERTRIES | BEE! Fe ee ae |) RD en § eae Yi MSX can have several of slots or banks of memory but unless they are controlled by some means, they can interfere with one another. In order 240 to sort out which slot is to be used for which page, MSX has a special mechanism to select slots. At any one time, the CPU can access 64k or 4 pages of RAM. These 4 pages can each be in different slots. The pages in slots are selected using the 8255 PPI's 8 bit output port A (see slot selector circuit next page). PAO and PA1 give the slot number for page 0. PA2 and PAS give the slot number for page 1. PA4 and PAS give the slot number for page 2. PA6 and PA? give the slot number for page 3. The signal from the PPI is sent to the slot selector 74LS153. This chip also receives signals from the 280 CPU for the current page the CPU is reading from and writing to. A15 signal A14 signal: The page number the CPU is reading and writing. 0 0) page 0 O 1 page 1 1 fe) page 2 1 1 page 3 The selected slot number is then passed on to the 2 to 4 decoder, 74LS139, which gives the slot select signal, SLTSL, to the slot con- cerned for the 4 pages. SLOT SELECTOR CIRCUIT DIAGRAM Numer | “4.5183 7aisi30] stot ee SELECTOR DECODER SGN. saci! 1Y | ———.» YO |}—~* SLTSLO PAGE 1 gy ve eee ene PAGE2 >| C2 y2 |—= SLTSL2 PAGE3 >| C3 Y3 |—* SLTSL3 CHOOSES THE SLOT CHOOSES INDICATE THE PAGE WHICH A15 PAGE CPU 880 How to Select and Enable a Slot To select slots for each page from software, output to I/O address &HA8, which is the output to port A of the PPI. The value to be output is an 8 bit number. BitNo. 7 6 5 4 3 21 0 | oo slot number of page O slot number of page 1 slot number of page 2 slot number of page 3 Example: Say you want to use slot 0 for page 0, 1, and 3, and slot 1 for page 2. BitNo. 76543210 00010000=16 You can output 16 to 1/0 address &HA8. However a better method is to use the BIOS call ENASLT (&H0024) from machine code. Expansion of Slots MSX has the option of having up to 4 basic slots, from 0 to 3. Not all of them are provided in every machine, but each basic slot can be ex- panded using an expansion box. The maximum number of slots, there- fore, is 16. Since all can be filled with up to 64K RAM, the maximum RAM MSX can handle is 1M bytes. These expansion slots cannot be expanded any further. They are always expanded from a basic slot, so there is no use plugging in an expansion box to an already expanded slot. To select an expansion slot, you must first select the basic slot where the expansion slot is plugged in, using port A of 8255 PPI. The slot select register is at location &HFFFF in the expanded slot. This determines whether the selected page in that slot is used or not. To find out whether a basic slot has an expansion box plugged in or not, write to location &HFFFF (POKE &HFFFF, &HOF). If you then read &HFFFF (PRINT PEEK (&HFFFF)) and get a complement of what you have written, this means there is an expansion box there. Expansion Slot Buffer Basic slots do not need buffer but the expansion slots do. Therefore, a cartridge slot expansion box requires a two-way data bus buffer within its Circuits, The buffer changes its direction depending on whether aread or write operation is being performed. The control signal which controls the direction of these buffers is BUSDIR. Those devices that send signals to the CPU, such as an |/0 cartridge, have to send the BUSDIR signal to 242 change the direction of the buffer from the expanded slots to the CPU. Those cartridges which contain either ROM or RAM only, do not have to manage the BUSDIR signal. Ram Search Procedure When the MSX computer is switched on it Searches all the slots to select the system RAM. 1. It first searches for available RAM in page 2, from &H8000 to &HBFFF, then enables the siot with the largest RAM in page 2. If there is more than 1 slot with the same largest RAM space. then the slot with the smallest slot number is selected. 2. Then it does the same for page 3, from &HCOO00 to &HFFFF. Again, the largest available RAM with smallest slot number is selected. 3. Lastly it checks if the RAM is continuous from &H8000 to &HFFFF, and sets the system variable BOTTOM (&HFC48) to the address of the lowest available RAM. 16 K EXPANSION TO 32K MACHINE SYS SLOT CARTRIDGE FFFF RAM C000 BFFF RAM | | | 16K ] PAGE 2 8000 = JEFF BOTTOM = &H8000 MSX 4000 3FFF MSX | | ROM PAGE 0 0000 CARTRIDGE ROM SOFTWARE ROM Search Procedure Having selected its system RAM, the MSX then searches for ROM cartridges from &H4000 to &HBFFF, page 1 and page 2. It looks for a valid ID at the beginning of each page from slot 0 to slot 3, and the expansion slots when expanded. 243 The ID area Is located at the beginning of a page and may have one of the following formats: Top of page &HOO ID ‘AB’ ID... 2 byte code ‘AB’ to + indicate that there is a ROM cartridge here. INIT INIT. holds the address of the initialisation routine for that cartridge. STATEMENT . holds the address of the expanded statement + &HO6 handler. BEXMES DEVICE . holds the address of +&HO02 +&HO4 STATEMENT! the expanded device handler UBBERE -eres TEXT routine. TEXT . holds the start address of +&HOA the BASIC program contained in that cartridge. reserved +&H10 Note: ID, INIT, STATEMENT, DEVICE, TEXT will contain zeros if not applicable. MSX BASIC takes the following action for the cartridge search procedure. 1. Checks the ID area and finds out what type of routine it has. It then passes on the information collected to the relevant work area. 2. Executes the INIT routine if there is one. 3. Executes the BASIC program in that cartridge if there is one. Note: STATEMENT and DEVICE are not executed at this point because they are only used when the user requires the expanded statement or device. INIT: Initialisation Routine INIT holds the address of the initialisation routine specific for the cartridge. It usually initialises 1/0 devices connected to that cartridge, or sets up a hook (more on this in Hook section), or reserves a work area for the cartridge software. The initialisation routine can change all registers except the stack pointer [SP]. It returns to BASIC upon the Z80 RET command. The INIT does not have to be an initialisation routine. It can be a machine code program to be executed immediately upon power up. It could be agame. If there is no initialisation routine then INIT contains O's. 244 TEXT: BASIC Program A ROM cartridge does not necessarily have to be a machine code program: it could be written in BASIC. TEXT holds the start address of the BASIC program contained in the cartridge. When programming a cartridge software in BASIC, the following notes should be remembered: 1. When there is more than one cartridge with BASIC software plug- ged in, the machine will only execute the one in the slot with the smallest slot number. 2. The BASIC program in the cartridge is stored in tokenised format. 3. The cartridge must be placed at page 2, &H8000 to &HBFFF. That means that the maximum memory of BASIC cartridge is 16K. 4. RAM placed at page 2 in any other slot is disabled and thus cannot be used from the BASIC cartridge program. 5. The address pointed to by the TEXT entry must contain a zero 0. 6. The line numbers for GOTO and GOSUB etc. should be changed to pointers for faster execution. If there is no BASIC program in the cartridge, then TEXT contains zeros. STATEMENT: Expanded Statement Routine Using the CALL statement in MSX BASIC, you may use expanded statements called from outside the MSX BASIC ROM. A cartridge con- taining an expanded BASIC statement must contain the start address of the first expanded statement routine in the STATEMENT. The cartridge must be placed at page 1, but can be in any slot except for the system slot where BASIC resides. The syntax for the expanded statement is: CALL <statement-name> CALL <statement-name> (arguments) CALL can be abbreviated by an underline character —’. When BASIC encounters a CALL statement, it stores the expanded statement name in the system work area, PROCNM. PROCNM is located at &HFD89 onwards and is 16 bytes long. The statement name ends with a zero when stored in PROCNM, so the maximum length of an expanded statement is 15 characters. Before the expanded statement is executed, the text pointer, the HL register, points to the address after the STATEMENT name. The STATEMENT procedure in the cartridge then checks the content of PROCNM, and executes the routine which matches the statement name. 245 After it has executed the expanded statement, the carry flag is cleared and the text pointer [HL] is pointed to the position of the next statement. If there is no expanded statement to match the name in PROCNM, then the carry flag and text pointer remain as they are, and return to BASIC with the ‘Syntax Error’ message. If there is no expanded statement routine in the cartridge, then STATEMENT contains zeros. DEVICE: Expansion Device Handler Routine MSX BASIC has the ability of plugging in an expansion I/O device to a cartridge slot. The DEVICE contains the start address of the device handler routine. The following notes should be remembered about DEVICE routines: The cartridge must be placed at page 1, &H4000-&HBFFF. 2. One cartridge (16K) can have up to 4 logical devices connected. 3. Device name is stored in the system area PROCNM, as with STATEMENT. PROCNM is located at &HFD89 onwards and is 16 bytes long. The device name ends with a zero when stored in PROCNM, so the maximum length of an expanded statement is 15 characters. 4. When BASIC encounters a device name, in an OPEN statement, and others, which is not known by the MSX ROM, then it calls device entry with &HFF in register A. If there is no handler which matches the device name, the carry flag is set. If there is a device, ID (from 0 to 3) should be returned in register A and the carry flag reset. All registers can be altered in the device routine. 5. Real 1/0 operation is carried out when a DEVICE entry is made with one of following vaiues in the A register: 0 Open 2 Close 4 Random 1/0 6 Sequential output 8 Sequential input 10 LOC function 12 LOF function 14 EOF function 16 FPOS function 18 Back up character The system variable DEVICE should be set to the Device ID number (0-3). If there is no DEVICE routine in the cartridge, then DEVICE should contain zeros. a 246 DESCRIPTIONS OF SYSTEM VARIABLES CONCERNING THE SLOT MECHANISM Status of each Slot EXPTBL — indicates which slot is expanded. Location &HFCC1 — 4 bytes long EXPTBL &HFCC1 for slot 0 &HFCC2 for slot 1 &HFCCS3 for slot 2 &HFCC4 for slot 3 &H80 indicates expanded slot &HOO not expanded SLTTBL — indicates what value is currently output to the expansion slot select register. This is valid only when the corresponding EXPTBL holds &H80, i.e. when the slot is expanded. Location &HFFC5 — 4 bytes long SLTTBL &HFCC5 for siot O &HFCCE for slot 1 &HFCC7 for slot 2 &HFCC8 - for slot 3 Status of each Page SLTATR — holds what is in each page for all possible pages. Location &HFCC9 —- 64 bytes long SLTATR &HFCCY for basic slot 0 expansion slot 0 page 0 &HFCCA for basic slot 0 expansion slot 0 page 1 &HFDO7 for basic slot 3 expansion slot 3 page 2 &HFDO8 for basic slot 3 expansion slot 3 page 3 BitNo. 76543210 | 14_1—1_1_______. = Q unused Statement expander inside when = 1 device expander inside when = 14 BASIC text inside when = 1 SLTWRK — work area for each page. The usage depends on what's in the page but 2 bytes per page allowed. Location & HFDO9 — 128 bytes long; 2 bytes per page 247 SLTWRK &HFDOS &HFDOA &HFDOB &HFDOC &HFD85 &HFD86 &HFD87 &HFD88 for basic slot 0 expansion slot 0 page 0 for basic slot 0 expansion slot 0 page 0 for basic slot 0 expansion slot 0 page 1 for basic slot O expansion slot 0 page 1 for basic slot 3 expansion slot 3 page 2 for basic slot 3 expansion slot 3 page 2 for basic slot 3 expansion slot 3 page 3 for basic slot 3 expansion slot 3 page 3 248 CHAPTER 23 PERIPHERAL DEVICES CASSETTE The following statements and functions are associated with the use of a cassette (detailed descriptions are given in the BASIC reference section). CLOAD CSAVE MOTOR SAVE LOAD BSAVE BLOAD Connector: 8 pin DIN plug, type 45326. Signal table SIGNAL DIRECTION OUTPUT INPUT OUTPUT OUTPUT { 2 3 4 5 6 vA 8 PRINTER lf your MSX computer is equipped with a printer interface, then you can use any type of printer which is equipped with a Centronics interface. If your computer does not have a built-in printer interface, then you must buy a printer interface cartridge which plugs into the cartridge slot. 249 Some printers are equipped with a set.of MSX standard characters. These MSX compatible printers are available from some of the manu- facturers, and give you the advantage of being able to print out all of the MSX graphics characters. A non-MSxX printer will not be able to print out the graphics characters; it will instead print out a space for each graphics character. If you are using an MSX printer, then you must tell the computer that you are using one by using the SCREEN statement: SCREEN ,,,,0 selects MSX printer SCREEN ,,,,1 selects non-MSxX printer LPRINT LPRINT USING To write to the printer, you must use the LPRINT and LPRINT USING statements, instead of PRINT and PRINT USING. Their syntax is virtually the same except LPRINT and LPRINT USING will not give any screen display. If you want something to be printed on the screen as well as the printer, you must use both PRINT and LPRINT in your program LLIST lists the program to the printer. Its syntax is the same as the LIST command. The LPOS function returns the position of the printer head, and is analogous to POS function, which gives the cursor position on the text screen. More details on the printer will be provided to you when you buy your printer. Invariably, the manual will include demonstration programs, written in Microsoft BASIC which is compatible with MSX BASIC. Connector: Centronics type, 14 pin Aphenol PIN NO. | SIGNAL NAME } PSTB PDBO PDB1 PDB2 PDB3 7 6 3.2 ~=«41 PDB4 + +———+- + +——~+——| PDB5 14 13 12 11 10 9 8 PDB6 PDB7 NC BUSY NC NC GND OONODNABWNH 250 JOYSTICK PORT Most MSX computers have two joystick ports, while some just have one: the BASIC can support up to two. Apart from joysticks, you can plug in other devices such as touch pads and games paddles. Connector: AMP 9 pin Joystick Joysticks are mainly used for playing games. !f your computer has two joystick ports but you only have one, then make sure you plug It into the port marked JOYSTICK 1, as most commercial programs assume that when you are using one joystick it is plugged into this port. To read the status of the joystick from BASIC, use the STICK and STRIG statements. STICK returns which direction the joystick is in, and STRIG tells you the status of joystick trigger button. The ON STRIG GOSUB statement provides you with a trigger in- terrupt. See the Event handling and Interrupt section. Games Paddle You may plug in up to 12 games paddles, 6 to each of the joystick ports. The status of the paddle can be read using the PDL function (see PDL.) Touch Pad You can plug in one touch pad per joystick port. PAD returns the status of touch pad (see PAD in BASIC reference section). SIGNAL DIRECTION 1 2 3 4 5 6 7 8 9 251 SECTION 3 REFERENCE SECTION CHAPTER 1 BASIC KEYWORDS EXPLANATION BASIC KEYWORD FORMAT This chapter contains all the MSX BASIC keywords in alphabetical order, and everything you need to know about them. The descriptions of the keywords are given in a standard way, making it easy to find their meaning. The description of each keyword is broken down as follows: KEYWORD This is sometimes followed by a few words on its derivation. DESCRIPTION This section explains what the KEYWORD does in simple terms. SYNTAX Every possible syntax is listed with explanations on the actions and the differences of each syntax format. The following symbols will be used: <num-const> is anordinary number such as 100 or 1.414. <num-var> is anumeric variable such as X or AMOUNT. <num-exp > means an expression which results in a number such as 2*PI*R*COS(Y). <num- const-> and <num-var> can be considered as simple expressions and can be used whenever an expression is required. <numeric> can be any of the above. 255 <string-const> indicates a string of characters enclosed in quotation marks, e.g. ‘’MSX COMPUTER 64K". <string-var> means a string variable such as B$ or WORD$. <string-exp > means an expression which results in a string e.g. B§+'’LONDON BRIDGE", ’'4", STR$(1234) <variable> can be either 1) <string-var> or 2) <num-var> <condition> indicates a TRUE or FALSE condition test such as A<O or B$=’'LONDON BRIDGE"’. <statement> can be any BASIC statement or a group of BASIC statements. <line> means the line number. <name> can be the name of a program or the name of a function. EXAMPLES This section includes various one line examples and short programs to illustrate how the KEYWORD may be used. A brief explanation of the example will be given. For short example programs the following symbols are used: = Anything inside this i square is an example program which can be RUN on your computer. The result of the example program will be displayed in this box. A supplementary explanation may be given inside this box on the same line. POINTS TO REMEMBER This section gives details of the do's and don'ts of the KEYWORD. A comprehensive list of hints and advice will be given, as well as any supplementary description of the KEYWORD. 256 BUG HUNT This section gives possible causes of errors associated with the KEYWORD and it is intended to help you debug your program easily. ASSOCIATED KEYWORDS AND OTHER REFERENCES This section lists all the associated KEYWORDS which are usually used in conjuction with the KEYWORD. If details of, or an example use of the KEYWORD is mentioned in any other section, such as the Advanced Programming section, then the list of these chapters will be given. 257 ABS ABSsolute value DESCRIPTION This function returns the absolute value, which is the difference between the number in the brackets and O, which basically means that all negative numbers become positive, and positive numbers are left as they are. For example, the absolute value of — 2.6 is 2.6. It is also used to calculate the difference between two numbers. SYNTAX ABS(<num-const>) ABS(<num-var>) ABS(<num-exp>) EXAMPLES poceas 10 PRINT “ABSOLUTE VALUE OF -1000 IS "; ABS(-1000) 20 A=1984 30 B=1960 4O PRINT ABS(B-A) — ABSOLUTE VALUE OF -1000 IS 1000 24 POINTS TO REMEMBER ABS is a function whose argument and result are both numeric. The ABS value of an undefined variable is zero. BUG HUNT Mixing of an ASB function with a string will result in a Type mismatch error. ASSOCIATED KEYWORDS AND REFERENCES SGN Section 1 Chapter 15 : FUNCTIONS 258 AND logical AND operator DESCRIPTION AND is one of the logical operators which performs a multiple condition test using Boolean algebra. Boolean algebra is a very necessary part of computer science. For more details, look up the section on Boolean Algebra. AND is often used within the IF... THEN .. ELSE statements to test more than two conditions before resultant action is taken. For example: IF A=5 AND B$="GODZILLA” THEN PRINT “AAARGH!" ELSE PRINT “OK!” The Booiean algebraic operation of AND can be listed in a truth table as below: AND Xx Y X AND Y ¢) 0) 0) 1 6) 6) 0) 1 6) 1 1 1 Therefore 100 AND 50 can be calculated as follows: X 100 0000000001 100100 AND Y 50 00000000001 10010 = 39 0000000000100000—i“(<i‘i;O;~*;*W SYNTAX IF <condition> AND <condition> ..... THEN ....... ELSE aca: <num-var>=<num-const> AND <num-const> <num-var>=<num-var> AND <num-exp> or other numerical combinations. EXAMPLES Example (1) of IF THEN ELSE condition testing: 10 T=12 : HUNGER$=""VERY" 20 INPUT "DO YOU WANT LUNCH (Y/N)"; AN$ 30 IF T=12 AND HUNGER$="VERY" AND AN$="TY" THEN PRINT “HAVE SOME SANDWICHES" : END 4hO PRINT "FINE. HAVE SOME LATER THEN." | 259 RUN this program twice. See what happens when you enter yes (Y) and when you enter no (N). — RUN DO YOU WANT LUNCH (Y/N)? Y HAVE SOME SANDWICHES Ok RUN DO YOU WANT LUNCH (Y/N)? N FINE. HAVE SOME LATER THEN. = Example (2) for Boolean algebra: 10 A%=185 : BK=85 20 CX#=A% AND BX 30 PRINT A%;" AND "; B%; "="3 C% 40 PRINT “MULTIPLY THAT BY 100 THEN YOU GET"; 100*(A% AND BX) ba tte tan Sr MULTIPLY THAT BY 100 THEN YOU GET 1700 [ 185 AND 85 = 17 | L POINTS TO REMEMBER Logical AND can be used for bit testing a particular number. For example if Y= 2", where nis the test bit between 0 and 7, then X AND Y = Y if the test bit n is true (i.e.1) and X AND Y = O if the test bit nis false (i.e.0). BIT TEST the 5th bit (n=5) for X=117. Y¥=2\5=32 X117 0000000001110101 AND Y 32 0000000000100000 32 0000000001 00000 therefore true. Boolean operands must be in the integer range of ~ 32768 to 32767 Real numbers are truncated to integers. 260 BUG HUNT An overflow error occurs when one of the operands lies outside the integer range. A Type mismatch error occurs when one of the operands is a string. ASSOCIATED KEYWORDS AND REFERENCES IMP Section 1 Chapter 6 : THE USE OF CONDITIONS Section 2 Chapter 6 : BOOLEAN ALGEBRA Section 2 Chapter 7: BOOLEAN II: THE IF THEN ELSE 261 ASC ASCII code DESCRIPTION Every character has a corresponding code number, called ASCII code. (ASCIl stands for American Standard Code for Information Interchange.) The computer converts all strings into numbers because it understands numbers better than strings. The ASCII standard was set up so that all computers using the ASCII code have the same number codes for their alphanumeric characters. This makes communication between computers easier. It is sometimes necessary to find out the ASCII code when programming; ASC converts a particular character or the first character of a string into the corresponding ASCII code. SYNTAX ASC(''<string>"’) gives the ASCII code of the first character of the <string>. The <string> can be more than one character long. ASC(<string-var>) gives the ASCII code of the first character of the string variable. EXAMPLES PRINT ASC('’A’’) will result in the display of ASCII code of '’A’’ which is 65. 10 A$="FRIDAY" 20 B=ASC(A$) 30 PRINT "ASCII VALUE GF THE FIRST LETTER OF "s3A%3" IS"sB The above program will give: ASCII VALUE OF THE FIRST LETTER OF FRIDAY IS 70 POINTS TO REMEMBER All characters, including the control codes such as <carriage return > and the graphics characters, have their own unique ASCII value and these can all be calculated using ASC. 262 Only the first character of a string is significant in ASC. The remaining characters are ignored. The reverse process of generating a character from an ASCII number can be achieved using CHR$, but this cannot give more than one character at a time. BUG HUNT If a string variable has nothing init, i.e. ifitis a null string, this will result in an Illegal Function Call error. . For example, 10 PRINT ASC(‘ ”) will result in an Illegal Function Call at 10. A type mismatch error occurs if the argument is numeric. ASSOCIATED KEYWORDS AND REFERENCES CHR§ SeeAPPENDIX for the ASCII character code table. Section 1 Chapter 20 : THE ASCIi CODES 263 ATN arc-tangent DESCRIPTION This function returns the arc-tangent of the argument in radians between —PI/2 and Pi/2 SYNTAX ATN(<num-const>) ATN(<num-var>) ATN(<num-exp>) ATN will give the arc-tangent in radians in double precision. EXAMPLES PRINT ATN(1.5) .982 79372324731 PRINT ATN(3/6) .46364 760900081 PI=4*ATN(1) .... 4*ATN(1) calculates the value of PI to 14 decimal places. POINTS TO REMEMBER ATN always gives a double precision number. The argument can be any numeric type. BUG HUNT AS itis a trigonometry function, it should obviously not be mixed with strings. If it is, a Type mismatch error will result. ASSOCIATED KEYWORDS AND REFERENCES TAN SIN COS Section 1 Chapter 19 : MATHEMATICAL FUNCTIONS 264 AUTO AUTOmatic line numbering DESCRIPTION BASIC programs require a line number for every program line. When typing a long program, the AUTO command will put you into the automatic line numbering mode which will offer you a new line number every time you press the <RETURN> key. This makes life a lot easier for the programmer. The line numbers will have aconstant step size. SYNTAX AUTO will give line numbers from 10 in steps of 10. AUTO <num-const> will give line numbers from the specified number in steps of 10. AUTO <num-const>,<num-const> will give line numbers from the specified number in steps of the second number. AUTO, <num-const> will give line numbers from 0 in the specified step size. AUTO <num-const>, will give line numbers from the specified line nuinber with the same step size as previously used. EXAMPLES AUTO will give line numbers 10, 20, 30, 40, and so on. AUTO 1000,5 will give line numbers 1000, 1005, 1010, 1015, and so on. POINTS TO REMEMBER Every time the <RETURN-> Is pressed the computer will offer you a new line number. There are two ways to exit the AUTO mode: 1) Press <CTRL> and <STOP> keys at the same time. 2) Press <CTRL> and <C> keys at the same time. The range of tine numbers you can have is from 0 to 65529 and you can select the step size from 1 to 65529. When the computer reaches line 65529, it automatically ceases to give you line numbers and puts you into command mode. 265 if there is an existing line number within the program you are editing, the computer will give you a warning signal in form of a star sign “*” after the line number. If you do not want to ruin that line, press the return key but if you are replacing that line, ignore the warning “*” The function key (F-2) is initialised to “AUTO” when the computer is switched on so you can get into the AUTO mode at the touch of a button. The line numbers must be integers. BUG HUNT If either the line number or the step size specified is not an integer, thena Syntax error will result. It is quite easy to erase part of a program unwittingly with the AUTO mode so watch out for the “*” warning. AUTO can be included within a program but it is rather silly. Don't do it. ASSOCIATED KEYWORDS AND REFERENCES RENUM Section 1 Chapter 7: USEFUL COMMANDS AND HINTS FOR WRITING PROGRAMS Section 2 Chapter 1: ADVANCED PROGRAM EDITING 266 BASE DESCRIPTION This system variable gives the current base address position of the various screen tables in the video RAM. BASE can be evaluated or assigned like an ordinary variable, but it should only be used in advanced graphics programs. BASE is a 16 bit integer. SYNTAX BASE(<numeric>) BASE(0) -—— base address of name table of text mode O. BASE(1) _ unused BASE(2) = base address of pattern generator for text mode O. BASE(3) — unused BASE(4) — unused BASE(5) — base address of name table of text mode 1. BASE(6) — base address of colour table of text mode 1. BASE(7) _ base address of pattern generator of text mode 1. BASE(8) = base address of sprite attribute for text mode 1. BASE(9) = base address of sprite pattern for text mode 1. BASE(10) — base address of name table of high resolution graphics mode 2. BASE(11) = base address of colour table of high resolution graphics mode 2. BASE(12) = base address of pattern generator of high resolution graphics mode 2. BASE(13) —~ base address of sprite attribute table of high resolution graphics mode 2. BASE(14) — base address of sprite pattern table of high resolution graphics mode 2. 267 BASE(15) — base address of name table of multi colour graphics mode 3. B EEP BEEP from the speaker BASE(16) -—- unused. BASE(17) base address of pattern generator DESCRIPTION table of multi colour graphics mode 3 BEEP gives exactly the same beep as PRINT CHR§(7) for 0.04seconds. BASE(18) _ base address of sprite attribute table SYNTAX of multi colour graphics mode 3. BEEP BASE(19) base address of sprite pattern table of multi colour graphics mode 3. EXAMPLES EXAMPLES Beer one BASE(0) POINTS TO REMEMBER BEEP is equal to CHR§$(7) cece cemien ASSOCIAT The argument can only be between 0 to 19 and must be an integer. No ED KEYWORDS AND REFERENCES other numbers are recognised and an Illegal function call will result. ee ASSOCIATED KEYWORDS AND REFERENCES Section 1 Chapter 28: THE MUSIC MACRO LANGUAGE Section 2 Chapter 16: THE ADVANCED GRAPHICS VI: THE VIDEO RAM 268 269 BINS BINary string DESCRIPTION BIN$ returns the binary equivalent of a decimal argument in a string form. It has the range of — 32768 to 65535 and the argument must be an integer, or anumeric expression which will result in an integer. SYNTAX BIN$(<num-const>) BIN$(<num-var>) BIN$(<num-exp>) EXAMPLES 10 PRINT BINS¢€255) 2O PRINT BINt( 9999) RUN | 11111111 10011100001111 L POINTS TO REMEMBER If the argument is negative, then two's complement form is used, i.e. BIN$(—n)=BIN$(65536—n). To convert a binary number to decimal, use the &B prefix to represent a binary number and equate it to a decimal variable such as below: A%=&B00001 111. BUG HUNT The argument must not be a string or real number. If itis a string or a real number, a type mismatch error will result. Overflow error occurs when integer is outside the range of ~ 32678 to 65535. ASSOCIATED KEYWORDS AND REFERENCES Section 2 Chapter 5: SURVEY OF NUMBER SYSTEM USED IN MSX 270 BLOAD LOAD Bytes DESCRIPTION BLOAD is used to load machine code programs and data from a specified device such as a cassette. It has the option of automatically executing the loaded machine code program, and also of loading the program to a different address from which it was saved. BLOAD is used for autorunning machine code software. Ai the moment the device option is restricted to a cassette but it will Support discs in the future. SYNTAX BLOAD ‘<device-name>:” for loading file with unknown name. BLOAD “<device-name>:<name>” for loading file with a known name. BLOAD ‘<device-name>:<name>".R. This automatically executes the machine code program just loaded. R stands for RUN. BLOAD “<device-name>:<name>”,<num-const>. When the offset <num-const> is specified, all addresses specified in the BSAVE are offset by that value. BLOAD “<device-name>:<name>",R,<num-const> loads with offset and then executes. EXAMPLES BLOAD “CAS:GAME".R BLOAD “CAS:ADVENT”,&HOOFF POINTS TO REMEMBER lf the execution address is omitted when the code is BSAVEd, then the computer assumes the execution address is the start address when the program is BLOADed with the R option to auto execute the program. Most machine code games which are recorded on cassette are likely to be auto executed using this command. The file name must be less than or equal to 6 characters. BUG HUNT Device |/O Error will occur if the file is corrupted on tape. If you get the file name wrong, obviously you are not going to load it into the computer. This is a common mistake. If you aren't sure of the name then omit the file name, as the computer will then load the first machine code bytes found. 271 ASSOCIATED KEYWORDS AND REFERENCES BSAVE Section 2 Chapter 11: CASSETTE SAVING AND LOADING 272 BSAVE SAVE Bytes DESCRIPTION This saves a chunk of memory at a specified memory on to cassette or disc. You must specify the start address and the end address. This command is used for saving machine code programs and data in the form of bytes. You can also specify the execution address which is used when automatically executing the machine code program when loading with BLOAD. SYNTAX BSAVE “<device-name>:<name>”",<start address:>,<end address> BSAVE ‘‘<device-name=>:<name>",<start address >,<end address >,<execution address > EXAMPLES BSAVE “CAS:GAME”,&HA100,&HA200,&HA1 FF BSAVE “CAS:ADVENT",&HCO00, &HCFFF POINTS TO REMEMBER The file name must be less than or equal to 6 characters long. If the execution address is omitted, then the computer assumes the start address is the execution address when the program is BLOADed with the R option, to auto execute the machine code. BUG HUNT Good SAVEing of memory can only be achieved if the recording device is reliable. Common sense really. If you can't SAVE properly, check your cassette recorder. A file name with more than 6 characters will cause a Syntax error. ASSOCIATED KEYWORDS AND REFERENCES BLOAD Section 2 Chapter 11: CASSETTE SAVING AND LOADING CALL CALL expanded statement DESCRIPTION This statement calls an expanded statement held within an expansion ROM cartridge. The abbreviation of call is '’_'’, the underline mark. CALL is ROM cartridge dependent and therefore when used in a BASIC program the program will lose MSX compatibility. SYNTAX CALL <name>(<list of arguments >) <name>(<list of arguments >) The list of arguments can be anything. It really depends on what the expanded statement does. EXAMPLES CALL SPEECH(loudness%,voice%, '’HELLO"’) POINTS TO REMEMBER Some computers use CALL to execute user’s machine code routine but this is not the case in MSX. To do that you'll need to use USR. The maximum length of the expanded statement is 15 characters. Unless you have an expansion ROM in one of the cartridge slots, this function is useless. Also what CALL does is totally ROM dependent. BUG HUNT Do not use this for commercial software or the program will lose MSX compatibility. A non-existent call name will cause a Syntax error. ASSOCIATED KEYWORDS AND REFERENCES Section 2 Chapter 22 : MSX MEMORY MANAGEMENT AND CARTRIDGE SLOT MECHANISM 274 CDBL Convert to DouBLe precision number DESCRIPTION The CDBL function converts integer and single precision numbers, variables and expressions, into double precision numbers. SYNTAX CDBL(<num-const>) CDBL(<num-var>) CDBL(<num-exp>>) EXAMPLES B#=CDBL(A!) PRINT CDBL(NO%) POINTS TO REMEMBER Double precision numbers are 8 bytes long and are accurate to 14 decimal places. BUG HUNT A Type Mismatch Error will result if a string argument is used or if CDBL is equated to a string variable. ASSOCIATED KEYWORDS AND REFERENCES CINT CSNG Section 2 Chapter 3: TYPE CONVERSION 275 CHR$ CHaRacter string DESCRIPTION CHR$ generates a character from the number given according to the ASCII code. This is complimentary to the ASC function. Since some of the ASCII codes correspond to control characters, such as clear screen, you can activate these using PRINT CHR§. SYNTAX CHR$(<num-const>) CHR$(<num-var>) CHR$(<num-exp >) The above will all return a character string. EXAMPLES PRINT CHR$(66) will print the character '’B’’ on the screen. 10 C1=65 =: REM ASCII code of A is 65 20 C2=72 : REM ASCII code of H is 72 30 C3=84 2: REM ASCII code of T is &d 40 PRINT CHR$(C2) ;CHRS(C1) 3; CHRS(C3) SO AS=CHE$ (C2+32)+CHE$ (C1432 )+CHR$(C3+32) 60 PRINT At If you run the above program, you will find that adding 32 to the ASCII code of capital letters turns them into lower case characters. RLN HAT hat POINTS TO REMEMBER When the number inside the brackets is between 0 and 31, CHR$ will act according to the control code as defined by the the ASCII standard. For example, PRINT CHR§$(7) will give a beep from the speaker. Look up the control code table. You will find that you can embed various control codes within a string, When the number inside the brackets is between 32 and 255, CHR$ will give the corresponding ASCII character. 276 BUG HUNT lf the number inside the brackets is either less than O, or more than 255, then an Illegal Function Error will result. Always remember that CHR§ is a string, and therefore cannot be equated to anumeric. Otherwise a Type Mismatch Error will result. ASSOCIATED KEYWORDS AND REFERENCES ASC See APPENDIX for ASCII character code table Section 2 Chapter 20 : THE ASCII CODES 277 CINT Convert to INTeger DESCRIPTION This function converts single and double precision numbers, variables and expressions, to integer numbers. It has a range between — 32768 to 32767. Anything outside this range will cause probiems. Fractional parts of a real number will be truncated. SYNTAX CINT(<num-const>) CINT(<num-var>) CINT(<num-exp>) EXAMPLES A%=CINT(BIC#) PRINT CINT(1234.56789) will give 1234. POINTS TO REMEMBER Integer numbers are 2 bytes long. BUG HUNT The argument must strictly be within the range of — 32768 to 32767. If not, then an Overflow error will result. A Type Mismatch Error will result if a string argument is used. ASSOCIATED KEYWORDS AND REFERENCES CDBL CSNG INT FIX Section 2 Chapter 3: TYPE CONVERSION 278 CIRCLE draw a CIRCLE DESCRIPTION CIRCLE draws the whole or a part of either a circle or an ellipse in the graphics mode. You can specify the position of the centre of the circle, colour, and even how much of the circle you want the computer to draw. SYNTAX CIRCLE <coordinate-specifier>,<radius>,<colour>,<start angle>. <end angle>,<aspect ratio> All can be <num-const>, <num-var> or <num-exp> but within their allowed range. (See below.) Explanation of <.coordinate-specifier>. There are two ways of specifying the coordinate of the céntre, one relative and the other absolute. The <coordinate-specifier> can be either of the following: 1) (<x-coordinate >,<y-coordinate>) These coordinates specify the absolute position of the centre of the circle on the screen. 2) STEP (<x-offset>,<y-offset>) These coordinates, <x-offset>, <y-offset>, are relative to the last point referred to by the computer. If the resultant offset is outside the screen, then the computer will draw up to the edges of the screen. <colour>, <start angle>, <end angle>, and <aspect ratio> are optional. Range: <x-coordinate > —32768 to 32767 : To place the centre on the screen use the range 0 to 255 <y-coordinate > —32768 to 32767 : To place the centre on the screen use the range O to 191 <radius > O to 32767 but the circle won't fit in the screen if the <radius> is too large. <colour> 0 to 15 See colour code section in COLOR. <start angle> 0 to 2*Pl in radians <end angle> 0 to 2*Pl in radians 279 A “~" minus prefix for the <start angle> and the <end angle> specifies a line to be drawn from the centre to each end. <aspectratio> Oto 32767 <aspect ratio> = (vertical radius)/(horizontal radius) If the aspect ratio is more than 1, then the ellipse will be elongated vertically. if the aspect ratio is less than 1, then the ellipse will be elongated horizontally. When the aspect ratio is unspecified the computer assumes an aspect ratio of 1 thus drawing a circle. If <aspect ratio= is large ail you will get is a vertical line. If O then you will get a horizontal line. EXAMPLES [ 10 SCREEN 2 | 20 CIRCLE (100, 100), 50, 8 30 FOR I=1 TO 5 40 CIFCLE STEP (I*¥5, 9),50,8 SO NEXT 60 GOTO 60 = REM THIS LOCKS THE SCREEN MODE POINTS TO REMEMBER This command only works in graphics modes, so be sure to select SCREEN 2 or SCREEN 3. If the colour is not specified the computer will draw in the current foreground colour. The start and end angles are in radians. BUG HUNT An Overflow error occurs when the value of the <coordinate specifier > or <radius> is out of range. An {Illegal function call error occurs if you get the colour code number wrong. An Illegal function call when <start angle> or <end angle ts out of range. ASSOCIATED KEYWORDS AND REFERENCES COLOR SCREEN Section 1 Chapter 25 : DRAWING CIRCLES AND ELLIPSES 280 CLEAR CLEAR memory area DESCRIPTION CLEAR tells the computer to forget all variables currently resident in memory. This command clears the variable memory space, nullifying all string and numerical variables. You can also reserve memory space with this command for string variables and machine code in the system work area. You can reserve memory space by changing the highest memory location available for use by BASIC. The highest memory location on default is &HF380. You can lower this using CLEAR ,<high-mem>. The memory space created can be used to store machine-code programs and data. SYNTAX CLEAR <string space> Sets the size of the string space. CLEAR ,<high-mem> Sets the highest memory location available for BASIC. CLEAR <string space, Sets the size of the string space as well as <high-mem-> the highest memory location available for BASIC. EXAMPLES CLEAR CLEAR 255 expands the string work area to 255 bytes. CLEAR ,&HF300 creates space for user's machine code program between &HF300 and &HF380 POINTS TO REMEMBER The tnitial string space is 200 bytes. This sets the maximum length of a string variable to 200 characters. If you want the computer to handle string variables longer than 200 characters, you must first execute a CLEAR 255 statement. The memory area between &HFFFF and &HF380 is used by the BASIC ROM as the system work space. ASSOCIATED KEYWORDS AND REFERENCES Section 1 Chapter 7: USEFUL COMMANDS AND HINTS FOR WRITING PROGRAMS Section 2 Chapter 2: CONSTANTS AND VARIABLES Section 2 Chapter 20 : MEMORY MAP 281 CLOAD/CLOAD? LOAD from Cassette DESCRIPTION Use CLOAD to load in a BASIC program from a cassette tape. The computer will tell you if it has found a program and whether it is skipping over it or loading the program. You don’t need to know the name of the program for, if this is unspecified, CLOAD will load in the first BASIC program found. Whether the computer successfully loads in a program or not is really dependent on the quality of cassette tape recorder. CLOAD? (cload question mark) is to verify the BASIC program just loaded from cassette with the one in the computer's memory. The computer will either tell you OK, or give a verify error if the current program is different from the one on tape. SYNTAX CLOAD CLOAD ’'<name>"' CLOAD? CLOAD? '’<name>"’ EXAMPLES Let us say there are two programs called ''SNAKE”’, and ''"LADDERS”’ on the cassette. Then, if you type: CLOAD<return> CLOAD Found = SNAKE aK Try CLOAD ’’LADDERS"’<return> CLOAD "LADDERS" Skip = SNAKE Found = LADDERS OK 282 POINTS TO REMEMBER The program name must not be more than 6 characters long When a program is loaded with CLOAD, any program currently held in memory will be deleted. The function key F7 is CLOAD”’. The recorded BAUD rate can be either 1200 BAUD or 2400 BAUD but the CLOAD will automatically sense which rate the recording is in and loads in accordingly. lf there is an open file (using function OPEN), then CLOAD will automatically close that file and proceed. BUG HUNT Device !/O Error occurs when you try to load in a badly recorded program. Alway use a reliable cassette tape recorder. lf you get the file name wrong, then obviously you won't get anything loaded. Be careful with your spelling. ASSOCIATED KEYWORDS AND REFERENCES CSAVE Section 1 Chapter 11 : SAVING YOUR PROGRAM ON TAPE 283 CLOSE CLOSE channel DESCRIPTION This command is used to close an opened channel and is the opposite to the OPEN command. The buffer associated with the specified channel is also released. SYNTAX CLOSE ........ close all CLOSE #<file number>, <file number> EXAMPLES CLOSE #2 BUG HUNT A Bad file Number error results when you enter a non-existent file number in the argument. ASSOCIATED KEYWORDS AND REFERENCES OPEN Section 2 Chapter 19 : HOW TO USE FILES 284 CLS CLear the Screen DESCRIPTION CLS clears the screen in any mode. In a graphics mode, the screen is cleared to the current background colour. SYNTAX CLS EXAMPLES The following clears the screen to bright red: - 10 SCREEN 2 20 COLQUF ,9 30 CLS 40 GOTO 40 LINE 10 CHOOSES HI-RES GRAPHICS MODE LINE 20 SETS BACKGROUND COLOUR TO 9 (BRIGHT RED) LINE 30 CLEARS THE SCREEN LINE 40 LOCKS ONTO THIS SCREEN MODE >Result : Bright red screen. POINTS TO REMEMBER The text cursor will be repositioned to the top left hand corner. When the function keys are in KEY ON mode, then the text screen will still display the function key list at the bottom of the screen after the screen has been cleared with CLS. To get rid of the key list you'll need to execute KEY OFF. Control L : <CTRL> <L> has the same function as CLS PRINT CHR§$(12) has the same function as CLS, i.e. you can embed a CLS statement within a string. ASSOCIATED KEYWORDS AND REFERENCES COLOR Section 1 Chapter 3 : WRITING A PROGRAM 285 COLOR colour ASSOCIATED KEYWORDS AND REFERENCES , DRAW DESCRIPTION PRESET This statement sets the colour of background, foreground and border. PSET There are 15 colours and one transparent. LINE CIRCLE SYNTAX CLS Section 1 Chapter 22 : COLOUR Section 2 Chapter 13 : ADVANCED GRAPHICS II: COLOUR IN HIGH RESOLUTION GRAPHICS MODE 2 COLOR <foreground colour> , <background colour> , <border colour> All can be <num-exp> but must evaluate to between 0 and 15. Also they are all optional so some of the arguments can be missed out. COLOUR CODE O Transparent 8 Medium red 1 Black 9 Light red 2 Medium green 10. Dark yellow 3 Light green 11. Light yellow 4 Dark blue 12 Dark green 5 Light blue 13 Magenta 6 Dark red 14 Grey 7 Cyan 15 White EXAMPLES COLOR 11 Sets foreground colour to light yellow. COLOR 15,,15 Sets border and foreground colour to white. COLOR 1,5,1 Sets background to light blue and border and foreground to black. POINTS TO REMEMBER When the machine is switched on the COLOR is set at COLOR 15,4,7. i.e. a white foreground, light blue background and cyan border. Function key 1 is predefined as the ’"COLOR”’ statement when the machine is switched on. Also F6 is ‘COLOR 15,4,7"’ which is the default colour. The background colour will not change unless the CLS command is executed. lf the colour is not specified in either PSET, or LINE, or CIRCLE, then these graphics commands will draw according to COLOR. BUG HUNT The arguments must be between 0 and 15; anything else will cause an Illegal function call error. 286 287 CONT CONTinue DESCRIPTION This command tells the computer to continue the execution of a program from where it left off by either a STOP statement or <CTRL> <STOP>. The computer remembers the last line executed before the <CTRL> <STOP> and carries on from the next statement. If the break was in the middle of an INPUT statement, then the computer carries on from the INPUT. SYNTAX CONT EXAMPLES BIRTHDAY" 20 INPUT YEAR,MONTH, DAY 30 PRINT "THANK YOU" RUN GIVE ME YOUR BIRTHDAY 71970 7713 PRESS <CTRL>><STOP =, HERE BREAK IN 20 OF. CONT 10 PRINT "GIVE ME YOUR | 71970 7712 771 THANE YOU OK POINTS TO REMEMBER CONT will work in the following cases: 1) After the program is stopped by a STOP statement. The program restarts from the next line. 288 2) After the program is stopped by an END statement.The program restarts from the next line. 3) After the program is stopped by <CTRL> <STOP>. The program restarts from the next line. BUG HUNT CONT will not work in the following cases: 1) After the program has been edited. 2) After stopping the printer. 3) After halting cassette input/output commands such as INPUT#. The resultant error message from the above will be Cant Continue. ASSOCIATED KEYWORDS AND REFERENCES Section 1 Chapter 7; USEFUL COMMANDS AND HINTS FOR WRITING PROGRAMS 289 COs COSine DESCRIPTION COS returns the cosine of an angle. COS is calculated to double precision. The argument, i.e. the angle, must be in radians. MSX does not recognise degrees. SYNTAX COS(<num-const>>) COS(<num-var>) COS(<num-exp >) EXAMPLES PRINT COS(1.4445432) 12591 79846524 POINTS TO REMEMBER COS returns in double precision. The argument must be in radians. BUG HUNT This is a trigonometric function, and therefore does not mix with strings. A Type Mismatch error will result if you use, for example, COS(A$). ASSOCIATED KEYWORDS AND REFERENCES