CD Ware Multimedia 1995 May

home *** CD-ROM | disk | FTP | other *** search

/ CD Ware Multimedia 1995 May / cd Ware (Juegos) Epimundo.iso / DOS / TOOLS / STARTREK.ZIP / CHAPTER.001 < prev next >

Wrap

Text File | 1989-05-15 | 53.8 KB | 1,057 lines

Chapter1 page 1 STARTREK THE COMPUTER PROGRAM By Joe Kasser CHAPTER 1 1.0 Making Friends With Your Computer For some reason, when most people first meet computers they are afraid of them. If they sit down at the keyboard they tend to freeze up. Visions of the computer running amok and causing tremendous amounts of destruction appear in their minds. Well just take a look at one. Does it look fierce? It may look complicated but it certainly doesn't look terrifying. Young children given the chance will spend hours pecking at the keyboard, so why should such a device inspire fear or feelings of inadequacy in adults. There is absolutely no reason to be afraid of it. First of all, it is completely benign. It has no feelings at all. It is an inanimate object. You can't hurt its feelings even if you physically assault it with a hammer, blow torch or other instrument of destruction. You may destroy it, but you definitely won't hurt its feelings. The computer can be likened to an unfinished being. It has been created (built) and is complete. It now requires instructions to perform any task. These instructions are called computer programs or software, and the act of putting together these instructions is called programming. Once a program is put into the computer it "comes to life". It can now do something. That something is almost anything that the human mind can conjure up. The computer can play a game, can control what ever is plugged into it, can teach, can in fact do what ever its instructions tell it to do. The first thing to remember is that the computer always does what it is told to do. That means for example if you are doing the programming, that the computer does what you tell it to do, which is not necessarily the same thing as what you want it to do. If you make a mistake in instructing the computer, how is it supposed to know what you meant to tell it? The best way to learn how to tell it to do things (ea. learning to program it) is to work with it. Thus first read the books that come with the machine that tell you what it does and how to use it. Then try and read a book that someone else has written about the machine that explains how to use it in greater detail than the documentation supplied by the manufacturer. You may then be ready to have a go at it yourself. Once you have decided to buy a computer make friends with it. In fact even if you haven't yet decided to buy one, its not a bad idea to make friends with one. Go to a computer store or the home of a friend who has one, sit down in front of the thing and play a few games. Games are the best method of introducing the beginner to the computer. Copyright (c) Joe Kasser 1989 Chapter1 page 2 STARTREK THE COMPUTER PROGRAM By Joe Kasser First familiarize yourself with using the computer, by playing with it. You can do an awful lot with a computer without knowing the first thing about programming. Applications packages abound. Apart from the myriads of games there are many serious packages such as spread sheet calculators, word processors, custom packages for various professions such as doctors or dentists, and financial and other business programs. Each comes with its own documentation that explains how to use the software. You only need to get involved with programming if you intend to write your own software or customize existing software to your requirements. When getting involved with programming start by typing up a program from a book or a magazine. If you don't understand the programming language, don't try to follow what is happening, just concentrate on not making any mistakes. You will become familiar with the jargon by working with it. The terms will enter your sub-conscious and you will slowly learn what they mean. When you type in a program, save what you have entered every few lines. If you are working with cassettes play back the cassette from time to time to verify that the recorder is adjusted properly. There are few things more frustrating than spending hours working at the machine only to lose the result due to a power failure or bad disk/cassette save operation. Once you have entered the complete program into the machine, save it, then have someone else help you verify the typing by reading the listing as you check the screen copy. It is best to work from the end of the listing back to the beginning if you can, so that you will not tend to follow the "story" and anticipate what is coming next. Edit any mistakes out of the program, save the verified version and then run the program. The first time around several things may happen including 1) the program performs as the author said it would, 2) the program does weird and wonderful things approximating to what the author promised it would, 3) the computer displays error messages, 4) a combination of 2) and 3) occurs. In the event that the computer fails to perform as the author said it would, recheck the program against the original magazine or book listing. If the program is written in BASIC and you get an error message of the type SYNTAX ERROR AT LINE ..... the computer is telling you that you have made a typing error in line .... . If the program has somehow been corrupted, reload it, reverify the listing and try again. If you still can't find anything wrong then in the case of a magazine article you can either hopefully wait for a correction to be published a few months later or try and find the error yourself. For that "User groups" and computer clubs are very handy, because they include people who have the same computer that you do and may know a little more about it than you. In fact it is advisable to try to Copyright (c) Joe Kasser 1989 Chapter1 page 3 STARTREK THE COMPUTER PROGRAM By Joe Kasser talk about the program before you even begin to type it up for you may find someone in the User group or club who has already typed it up and gotten it working. The procedure of getting the program working by taking out the errors (known in the trade as "bugs") is called de-bugging the program. You will then be forewarned as what to do to get the program to work, or better still be supplied with a cassette or disk for your personal use, saving you the bother of typing the thing up yourself. Once the program is working correctly (more or less), you are ready to start modifying the program. Get hold of a book that describes the programming language that you are working with so that you can use it as a reference book and dictionary. Dissect the program to see what the author made the computer do. See how the idea for the game was translated into a computer program. Trace the flow of the program and make small changes. For example, change the color of the display; change the dialog; as you perform these small modifications you will build up confidence in yourself. You are now instructing the machine. You are taking over from the original programmer and are making it perform to your wishes. You are becoming a programmer. When you have played with other people's programs for long enough the time will come when you are ready to write your own program. All computer applications are limited by either hardware or software. Control applications may be limited by the input/output (i/o) capabilities of the machine, games may be limited by the amount of memory in it, graphic capabilities are limited by both the resolution of the screen, the colors available and the speed that the computer can update the screen. The first generation of computer games were limited in this manner. In the early days of personal computers, the only terminal available was a teletypewriter. This was a slow machine which printed capital letters on paper. This type of machine is still used in TELEX applications. The computing power was there but the graphics i/o capability wasn't. The first games were thus mostly some sort of simulation games with a minimal amount of i/o. Video displays and CRT terminals soon followed. The same technology that brought us the video displays also allowed for video games. At first these games were dumb, namely tennis or ping-pong. then as the technology improved, games and personal computers also improved. The following generation of computers were designed mainly as game playing machines and incorporated relatively high resolution color graphics capability. The lowest resolution machines were those designed to work into the antenna of a conventional television receiver. These machines made real time action games available in the home and have proved to be very popular. Higher resolution plug in add on's and video monitors added more and more capacity to the machines and as technology progresses games became and will become more complex and sophisticated, the i/o interface will become better, more "on- line" memory will become available and greater realism will Copyright (c) Joe Kasser 1989 Chapter1 page 4 STARTREK THE COMPUTER PROGRAM By Joe Kasser become possible. The simulation games were superseded by the action games which required some degree of co-ordination yet little in the way of thought. In the current generation of simulation games, the game is programmed about a behavior model. In this book we shall discuss a Simulation, the theory behind it and the techniques that can be used to translate the simulation model into a workable and enjoyable game. The computer is dumb. It does exactly what you tell it to do. Not only that, you have to tell it in detail how to do what you want it to do. There is no point in telling the computer to "rule the world", you have to tell it in detail how to build up armies, cause subversion in other countries, fight battles and then set up your world government. You may ask what is the point of having a computer, if it has to be told everything in detail. If you are going to have to explain each point meticulously, you may as well do the job yourself. Well the computer is best doing repetitive tasks. If the calculation is to be done once, use a pocket calculator, if it is going to be done many times, use a personal computer. The computer has to be instructed by a human being. When there is a mistake in a bill or account statement, it is very unlikely that the computer is at fault no matter what you are told. The fault is most probably human. The mistake is either in the program or more probably the mistake was made at the time of the entry of data into the computer. If the person entering the amount of your payment into the computer makes a mistake by entering the wrong amount and the mistake is not caught, the program will grind on on its way and will send you messages based on the faulty input data, because after all, how was it supposed to know. Amongst computer professionals this phenomenon is known as "garbage - in, garbage - out" (GIGO). Computer languages are the means by which people instruct computers to perform tasks. Any person who wishes to control the operation of a computer rather than just to communicate with it thus has to learn a programming language. Learning a computer language is easier than learning a foreign language; such as French, Spanish, Japanese or Hebrew because in a computer language a word or expression has only one meaning, and the computer does exactly what it is told to do. Learning a language is a chore unless you remember that it is really a means to an end. The end is the task that you want the computer to do, the means is the language that you use to tell the computer how to do the task. If you can define the task first, then use the language to make the computer perform that task, and you are really interested in that task, learning the language will become a by-product of the operation. Talking to the computer in a language that it understands is called programming. Programming is the art of writing programs Copyright (c) Joe Kasser 1989 Chapter1 page 5 STARTREK THE COMPUTER PROGRAM By Joe Kasser or sequences of instructions for the computer to perform. Being able to write programs is not the same as being able to write good programs. There are many people who earn their living writing computer programs. There are other people who earn their living fixing up programs written by other so called "programmers". Computer programming is easy. Any idiot can do it. Good computer programming is difficult but can be learnt. If you are learning your first computer language, you are also learning programming, which is not the same thing. There are many programmers who know FORTRAN or BASIC yet do not know a word of FORTH. There are people who know many languages yet are terrible programmers, so if this is your introduction to BASIC, you are also learning programming as well as BASIC and it is as well to differentiate between the two. The habits you form now will govern your programming ability from now on, so be sure that they are the correct ones. Computer programs operate on three categories of data. They are constants, variables and parameters. The data elements may be simple or complex and stored as single items or as rows or arrays of elements. A constant is a fixed value that never changes, such as 1, 2, 88, 1023 etc. A variable is a data element that changes its value while the program runs. It is usually referenced by a name such as APPLE or A0. The address of the variable is the location that it is stored in memory, the contents of the location are the value assigned to the variable. A parameter has a value that is fixed for a particular run of the program. For example, in space war games, the number of enemy vessels at the start of each game may be different but constant for the particular game. That number can thus be considered as a parameter. Since most BASIC interpreters process constants differently to the way they operate on variables, it is usually advisable to store constants as parameters. Thus for example, instead of using 0.1 as a constant if it will be used many times, use P1 (point one) as a parameter equal to 0.1 Values are assigned to variables and parameters using the LET statement in BASIC such as in the statement LET P1 = 0.1 Alternatively, you will also meet the LET statement in the format LET A0 = A0 + 5 Here, the statement means, take the current value stored in the variable A0, add 5 to it and save the result (in the location addressed as) A0. If you are at all familiar with algebra, you will have noticed right away that there was something wrong with the equation. You are correct, the statement is not an equation, it is an expression. It could have been expressed as LET A0 + 5 -> A0. Copyright (c) Joe Kasser 1989 Chapter1 page 6 STARTREK THE COMPUTER PROGRAM By Joe Kasser Data comes in a number of formats. If the format is numeric (ea. it is numbers), the data can be integer or floating point. Integer numbers do not have a decimal part. They are whole numbers, floating point numbers have decimal parts. Example of floating point numbers are 1.652, 12.13, 0.003, and 567.8. Note the different position of the decimal point in each of the numbers, namely it floats around. An other form of data is text material in ASCII format. These are words that make up headings and titles, such as "SHORT RANGE SENSORS" or "PHASERS". In BASIC such material are called "strings" because they consist of strings of letters joined together. Variables, constants or parameters containing strings of text are called string variables, string constants or string parameters. They are identified by a dollar sign termination (pronounced "string") such as A$ ("A string") or D$ ("D string"), and can be treated just like regular numeric variables, thus LET A$ = "SHORT RANGE SENSORS" sets up the contents of the string variable A$ as SORT RANGE SENSORS. Quotation marks have to be used in BASIC to define the start and the end of the string of characters. Variables can be stored individually or in arrays. Individual variables can be named, such as APPLE, PEAR, ORANGE or BANANA. An array can be considered as a group of variables that can be referenced with respect to some kind of common denominator. For example, if we had an array named FRUITBOWL then each item in the array can be a separate fruit, and can be referenced with respect to the FRUITBOWL as FRUITBOWL(1) for the first element, FRUITBOWL(9) for the ninth, etc. An ideal task for the beginner to learn a language on is a simulation game which is written around the computer that the beginner has available. For in that case, there is complete control of all inputs and outputs, This kind of game in which the player makes decisions based on the information available to him or her available at the time), can be made sufficiently sophisticated and complex so as to make writing it an adequate challenge for anybody. The techniques used in writing a good game are the same that programmers use in professional activities. Writing a good game poses a challenge that allows you to develop good habits and techniques for programming and also allows you to learn a language in an interesting manner. By taking an orderly approach to the game design, complex operations may be clearly understood and converted to computer code with the aid of a language reference manual, irrespective of the language being used. 1.1 Defining the Game or Task Pick a task that interests you. you can invent a new game Copyright (c) Joe Kasser 1989 Chapter1 page 7 STARTREK THE COMPUTER PROGRAM By Joe Kasser or rewrite an existing one. If you are familiar with a game written in one language such as BASIC, and are learning another such as FORTH, examining the different approaches used by programs written in two languages to achieve the same end results will be educational in itself. If you plan to interest others in playing your game once you have finished it, there are a number of factors to be considered. The first one is the ease of learning how to play the game. There should be no need for someone to sit down and study a 176 page manual (no matter how well written) in order to learn the rules. If you can model real life, a well known television series, a best selling book or some other widely known environment, your game will have a wider appeal than if you had invented the game from scratch. The disadvantage of basing the game on an existing work of fiction is that it has been copyrighted and depending on what you plan to do with your finished product, you may have to enter into negotiations with the owner of the copyright. If you invent your own game, or your own situations, try and use characters or situations that are familiar to the players before they sit down to play the game. If you cannot do this, then make the options available to the player self evident. The second factor involved is the degree of complexity or sophistication of the game. The more complex it is, the greater the enjoyment in playing it, however that also means that it will be more difficult to learn the rules. There is a tradeoff involved between "getting them interested" and "keeping them interested". Another factor is realism. the player has to believe that the game takes place in a logical and believable environment. Don't create an absurd or unbelievable situation and expect anyone to play your game more than once. Books, films and plays are entertainment. A computerized game is interactive entertainment. You, the game writer create the background and set up the rules for the consequences of any action. you then let the player wander through your creation, and each playing of the game will be different. It is this difference that poses the challenge and makes them come back again and again. Embellishments or "bells and whistles" can leave a very positive impression on your game players. How many times have you heard said about a movie that "the plot was terrible, but the special effects were great". Put as many special effects as you can within reason into your game. The first step in writing the program is coming up with the idea. What should the program do? What do you want the program to do? A game is a good first time program because you have total control over all aspects of the task. You can decide the rules. You can decide on the i/o. You don't have to build or buy a special interface for the computer to talk to the outside world. Your program can be self contained. Copyright (c) Joe Kasser 1989 Chapter1 page 8 STARTREK THE COMPUTER PROGRAM By Joe Kasser Once you have developed the idea write it down. This becomes the specification for the program. It should describe in words what the program is to do. It does not necessarily describe how it will do the job. The specification is thus independent of the programming language. The next stage is to decide on the language. The decision may be made for you, for example if the personal computer comes with a built in language when you purchase it, and you don't have any others, that is the one you will use. Most computers at this time come with a dialect of BASIC built into them. The vast majority of computer programs can be written in almost any computer language. Some programs however are optimally written in languages designed for particular applications. FORTRAN for example optimizes mathematical computations, COBOL was designed for business applications, BASIC for teaching programming, FORTH for process control, and so on. To repeat, a game is an ideal task for learning a computer language. In defining the game you set up all the conditions and constraints, you are bound only by your imagination and the physical limitations of the computer. (There is no point in writing a graphic shooting gallery game if the only operating interface you have is a teletypewriter). Once the task is defined, you know what the computer is going to do. The next stage is to decide how it is going to do it. The complete task must be broken down into a number of sequential steps. How detailed these steps have to be is a function of what level of language is being used to write the program. For example, adding two variables together could be written in BASIC as LET A0 = A1 + A2 which instructs the computer to add together the values assigned to the variables A1 and A2 then to store the result in the variable A0. A similar sequence written in 8080 assembly language could look something like LDA A1 ;LOAD THE CONTENTS OF A1 INTO THE ACCUMULATOR LXI H,A2 ;POINT THE H AND L REGISTERS AT A2 ADD M ;ADD THEM STA A0 ;AND SAVE THE RESULT IN MEMORY AT LOCATION A2 In the low level assembly language, you have to assume that the computer knows nothing. In order to have it add those two values, not only do you have to tell it to add the numbers, but you first have to tell it where they are stored in memory, then you have to tell it to fetch them from memory, add them together and where to store the result. In the higher level language, you don't care where the data are stored in memory, you let the computer assign memory space to the variables, and all you do is tell it to add them. Earlier we saw that the computer is ignorant until Copyright (c) Joe Kasser 1989 Chapter1 page 9 STARTREK THE COMPUTER PROGRAM By Joe Kasser instructed otherwise. How then can the computer know by itself where to store those numbers? A statement such as LET A0 = A1 + A2 only tells the computer what to do with A0, A1 and A2, it says nothing about where they are located in memory. In order to understand this we have to take a brief look at the internal hardware of the computer as shown in figure 1.1. The majority of the hardware consists of memory which is used to store both instructions (programs) and data. The instructions are carried out in the Arithmetic - Logic Unit (ALU) of the Central Processing Unit (CPU). The interface to the user is located in the i/o module. When the computer is first turned on, the internal circuitry of the hardware dictates the memory location that will be read as the first instruction. Each time that the microcomputer is "powered up", that location will be assumed to contain the fist instruction. In the case of the 8080 or Z80, that location is the first location of memory at an address of 0. The starting location for other microprocessors differs. Most microcomputers contain hardware read only memory (ROM) at the starting address location or some other circuitry to fool the CPU into thinking it is looking at the starting address when it is really looking elsewhere. The actual task performed by the program in the ROM at the starting location differs in each machine. In some, it may initialize a monitor program such as AMS-80, in others it may initialize BASIC, or boot up a Floppy Disc Operating System (DOS). Now consider what actually happens when when an instruction is carried out. At the micro - micro level, just after initialization, the computer is going to carry out its first instruction. The first thing it does is fetch the first part of the instruction from memory into the Central Processing Unit (CPU). The hardware in the CPU determines the type of the instruction and then performs the remainder of the instruction. Each instruction is divided up into a number of parts. There is the fetch cycle in which the instruction is moved from memory into the CPU and then there is the execution cycle in which the instruction is carried out. The mechanism to perform these functions are built into the hardware of the machine and are known as the micro code. Different microprocessors have different micro codes and thus also have different sets of instructions. There are some special purpose microprocessors (usually of bit slice architecture) that allow you to define your own micro code and thus your own basic instruction set. The micro code operates on a set of bits called a byte. If the machine is an 8 bit machine it has an accumulator that is 8 bits wide, if it is a 16 bit machine it has an accumulator that is 16 bits wide. The byte in the accumulator is represented in binary (Base 2) math as a number. For our convenience as programmers, these binary numbers are grouped together and represented by Octal (Base 8) or Hexadecimal (Base 16) values. An example of such a number is 11000011 in Binary, 303 in Octal Copyright (c) Joe Kasser 1989 Chapter1 page 10 STARTREK THE COMPUTER PROGRAM By Joe Kasser or C3 in hexadecimal notation. The CPU works with digital signals that only have "on" or "off" states, represented by a 0 for "off" and a 1 for "on". It is much more convenient to refer to C3 00 F0 as an instruction than 11000011 00000000 11110000. If memory is filled with binary numbers, how then does the computer know which are instructions and which are data? The microcode knows how many bytes an instruction requires. That knowledge is built into the hardware as part of the architecture of the machine. So once it has read and decoded the first byte it knows what is to follow. The first byte is known as the operation code (op code) because it defines what operation is to be performed. Thus for example in 8080 code, a branch or jump instruction that causes the program flow to continue from some location other than the next one in the sequence, is always followed by two bytes containing the address of that location. The first few instructions in the start up ROM are in binary code. Since binary code even in its Octal or Hexadecimal representation is not easy to use for most programmers, assembly language and assembler programs were invented. Assembly language allows us to refer to an instruction in a pseudo english manner somewhat symbolic of what the instruction really does, for example, again in 8080 assembly language JMP means jump (to an other instruction located somewhere else in memory), MOV means move something from one register to an other, MVI means move something (like a constant) immediately into a register. The assembler also allows us to refer to memory locations by name instead of by the address of the absolute location. Figure 1.2 is an example of a program listing made using an assembler. An editor program is first loaded into the computer. This is a (rudimentary) word processor that allows the assembly language instructions to be stored in machine readable form. The assembler is then invoked to convert the Assembly Language Source code to a binary object code. This object code is then loaded into memory and is then executed by the machine in turn. Note the use of comments after the instruction. The format of the listing is address location, hexadecimal representation of the object code, assembly language (split into op code and numeric parts) and comments. In 8080 assembly language, the comments have to be preceded by a semi colon. In BASIC comments are preceded by a REM statement. Comments are very important and are used to explain why something was done in a particular manner. They serve two purposes: the first being to explain to others what you did and how you did it, the second reason is to remind you of what you did when you come to make some modifications to the program after some time has elapsed since you wrote it. Writing in assembly language has many disadvantages including the one that the programmer has to give very detailed instructions and has to keep track of the locations of all the data. There must be a better way, and for many applications there is, and it is known as a high level language. Copyright (c) Joe Kasser 1989 Chapter1 page 11 STARTREK THE COMPUTER PROGRAM By Joe Kasser 1.2 The high level language The high level language of which BASIC is just one example is itself a program that is a tool which allows the programmer to concentrate on telling the computer what to do. In assembly language, not only do you have to tell the CPU what to do and how to do it, but you also have to tell it where everything is. The assembler facilitates the job but the onus is still on the programmer. The high level language on the other hand has built into it a lot of "where" information. From now on, when using the high level language, if you define a variable such as APPLE, the language will assign a memory location to APPLE and some bytes (space) to store data in that location. Most versions of BASIC do not even require you to tell the computer that you are going to use APPLE as a variable until the first time that you actually use it. If we go back to the A0 = A1 + A2 example that was mentioned above, the full assembly language listing also has to include A0: DS 1 ;DEFINE SPACE FOR 1 MEMORY LOCATION A1: DS 1 ;DEFINE SPACE FOR 1 MEMORY LOCATION A2: DS 1 ;DEFINE SPACE FOR 1 MEMORY LOCATION which tells the assembler to define the locations of the variables and reserve one byte of memory to be associated with each of them. If these pseudo instructions were not included in the source code listing, the assembler would flag every reference to A0, A1 or A2 as an error (referencing an undefined memory location). The use of the term pseudo instruction, defines a category of directives to the assembler, rather than instructions that are executed by the CPU. Consider now what happens when BASIC meets and processes the instruction LET A0 = A1 + A2 which instructs the computer to add together the values assigned to the variables A1 and A2 then to store the result in the variable A0. BASIC first locates the variable A1 and A2 and adds their contents together. It then searches for and stores the result in the variable A0. If A0 has not yet been defined, BASIC allocates space for it at this time. The programmer does not know where the variables are stored in memory and in most instances does not care. BASIC is available as an interpreter or as a compiler. The interpreter resides in memory together with the program. It sequentially interprets each line of program in turn, converting it on the fly to binary object code and executing it. Since each line of code is interpreted each time it is executed the running speed of the program is relatively slow. A compiler for BASIC on the other hand performs an operation on the program written in BASIC analogous to the operation of an assembler on assembly language to convert it to an object code module. Since the computer then executes a compiled binary object module the execution time is speeded up. Copyright (c) Joe Kasser 1989 Chapter1 page 12 STARTREK THE COMPUTER PROGRAM By Joe Kasser Programs being executed in the interpretive mode can be interrupted and changed at will in an interactive mode, ea. debugged "on line". Programs that are compiled and then executed have to go through an editing (of the program), compilation and execution sequence and cannot be altered during the testing session. The interpreter is thus used when developing programs for ease of debugging. The compiler can then be used to provide somewhat greater execution speed and to make the code more difficult to disassemble by others if security of the source code is important. Note that even when compiled, programs written in high level languages always run slower that the equivalent program written in assembly language. Computing as you will find, more and more, is a balance of trade-offs. The trade-off between assembly language and the high level language is usually that of speed (execution time) against ease of telling the computer what to do. In most cases the speed of execution of a program written in a high level language is still greater than the user needs. There are however instances when not only must assembly language be used to meet execution speed requirements, but the actual instructions used must themselves be chosen to comply with the same speed requirements, because different assembly language instructions have different (numbers of machine cycles) execution times. 1.3 Structuring Programs Good programs are well thought out, logical and structured. The are put together in a modular manner and structured so that they have a beginning, a middle and an ending. This means that much thought has to be applied before the program is even written. BASIC is often thought of as an unstructured language. That is not true. BASIC is usually met for the first time in its interpretive incarnation. That means that it is easy to sit down at the computer, type in a few lines of a routine and have the computer do something. A few more lines are then added to do something else. Later some more instructions are added, linking the first two sets. More and more instructions are added and linked together until a program is finished. This finished program is usually very unstructured because it contains numerous instructions to branch out of the sequence to different lines elsewhere in the listing. There will be forward branches to lines still to come, and backward branches to lines that have already appeared. As a result of seeing numerous programs of this kind, the word has gotten out that BASIC does not allow well structured programs to be written. Balderdash! Programs written in BASIC can be just as structured as those written in any other language if you take care to do so. That means don't just sit down at the machine and start typing. Sit down with your word processor, first define what the program is to do, then chart out the flow of the program, lastly sit down and enter it into the computer. You then get small sections of it working and link them to other sections as you add them. However, since you have flow charted Copyright (c) Joe Kasser 1989 Chapter1 page 13 STARTREK THE COMPUTER PROGRAM By Joe Kasser the program beforehand, you enter the various sections into the computer in a logical manner and a structured program results. 1.4 Flow Charts A flow chart is a tool for keeping track of the flow of a program. It shows in outline form what is being done and how decisions affect the sequence of instructions. The operations performed by a computer program fall into distinct categories. There are input/output operations (i/o), processing modules, decisions, etc. Each can be represented by a different shape on the flow chart as shown in figure 1.3. Flow charts are graphic tools, since the computer is a digital device most suited for word processing and other non graphical applications here is a way of drawing flow charts using a word processor. It is also designed to optimize top down program design and bottom up implementation. Figure 1.4a shows a typical extract from a flow chart. At the top there is a test or choice. Branching out from there are two blocks that contains tasks or procedures that are self contained sequences of some kind. If the condition being tested by the test block of the flowchart is met, the program flows along the YES branch and procedure B is performed. Should the test fail, procedure A is carried out. Both branches THEN come together and continue the program flow. This flowchart can be represented as follows; CHOICE =? YES DO.PROCEDURE."B" NO DO.PROCEDURE."A" THEN Note that it is also descriptive of what is being done. Figure 1.4b shows the same flow chart but the NO side is missing. In this example, there is no operation performed if the test fails. This is written as CHOICE =? YES DO.PROCEDURE."B" THEN If the reverse is true, namely nothing is done if the test passes, but if it fails, a procedure is carried out as shown in figure 1.4c, it can be written as CHOICE =? NO DO.PROCEDURE."A" THEN the indentation makes the written chart easier to read. A more complex extract of a flow chart is shown in figure 1.5. Here there are four procedures (A,B,C and D) and two choices (1 and Copyright (c) Joe Kasser 1989 Chapter1 page 14 STARTREK THE COMPUTER PROGRAM By Joe Kasser 2). This flow chart can be written as follows CHOICE.1 =? YES (1) DO.PROCEDURE."B" CHOICE.2 =? YES (2) DO.PROCEDURE."D" NO (2) DO.PROCEDURE."C" THEN (2) NO (1) DO.PROCEDURE."A" THEN (1). Another section of a flow chart is shown in figure 1.6. This is an example of a loop in which something is done (Procedure "A") as long as the value of the variable I has not counted down to zero. This can be written as SET.VALUE.OF."I" LOOP (on I) + DO.PROCEDURE."A" + DO.PROCEDURE."B" + DO.PROCEDURE."C" LOOPEND. The "+" is used on the first line of the chart statements referencing operations or procedures within a loop to signify that a loop is in progress. There may or may not be the need for the programmer to test the value of the loop counter (I), depending on how the loop is implemented. These conventions should cover all aspects of structured program design. If your program cannot be defined in these terms, it probably is not structured correctly. Figure 1.7 shows a typical flowchart commonly met when dissecting programs written in BASIC. As you can see it contains two joined loops that cannot easily be expressed using the convention described above. It is an unstructured example of the unrestricted use of the GOTO statement available in the BASIC language. This type of program is usually a result of programming at the keyboard without prior thought. When using this convention for writing flowcharts, it is desirable to have some method of indicating the start and finish of a procedure. Let us use the ":" (colon) to signify the beginning of a procedure, and the ";" (semi-colon) to signify the end of the procedure. This convention will make learning other computer languages easier later on in your programming career. It is also advisable to split the flow chart for a program into a number of smaller flow charts each explaining a different section of the program. Consider the flow charts shown in figure 1.8 for the operation of reading a book. The whole program can be explained by figure 1.8a as : READING.A.BOOK READ.THE.BOOK ; Copyright (c) Joe Kasser 1989 Chapter1 page 15 STARTREK THE COMPUTER PROGRAM By Joe Kasser Obviously, this means very little so we now have to explain how the procedure is carried out. This is shown in figure 1.8b as follows : READ.THE.BOOK GET.THE.BOOK OPEN.THE.BOOK.AT.THE.DESIRED.PAGE READ.THE.PAGES PUT.IT.AWAY.AGAIN ; not bad, now let's take a look at how the pages are read as shown in figure 1.8c : READ.THE.PAGES LOOP (Till all pages are read) + READ.SOMETHING.FROM.THE.PAGE + WHOLE.PAGE.READ =? + YES (1) END.OF.DESIRED.SECTION.OF.BOOK =? + YES (2) SET.END.OF.LOOP.FLAG + NO (2) TIME.TO.TURN.PAGE =? + YES (3) TURN.THE.PAGE + THEN (3) + THEN (2) START.AT.THE.TOP.OF.THE.NEXT.PAGE + THEN (1) LOOPEND ; Each step is described in words. The "." (period) between words in the description of the procedure signify that a single phrase is being used in the description. If spaces were used confusion could arise from time to time. Comments can be enclosed in paretheses "(" and ")". When such a chart is converted to a program, the labels or line numbers in the program associated with the statements in the flowchart can be added to the flowchart to make cross referencing easy. There has been no discussion as to exactly how the loop is implemented. That will depend on what the loop is doing. Different methods of implementing loops will be discussed later during the course of the book. Programming Programming a computer is done in three stages. In the first stage the action of the program is defined. The specification is written, the flow charts are drawn and the structure of the program is laid out. The flow charts are then converted to programming language statements in the second stage. Once the program is written the third stage involving the test and debugging occurs. Good programmers spend a lot of time performing stage 1. They define everything that has to be done by the program, and how it is to do the job. The coding process does not take too long. The actual time that it takes is usually a function of your typing speed. The test and debugging phase should not take Copyright (c) Joe Kasser 1989 Chapter1 page 16 STARTREK THE COMPUTER PROGRAM By Joe Kasser too long if the program is well defined and the whole task to be performed is well understood. Programmers that spend hours working on their own code at the terminal correcting one bug after an other are not "hot shot" programmers, but rather are the reverse. 1.5 Bugs Errors in programs are called bugs. It is inevitable that errors will occur during the process of specifying, flowcharting and writing the program. These bugs fall mostly into three categories as follows. 1.5.1 Design Errors Design errors are errors made when turning the specifications for the program into the flowchart. This is an example of the computer doing what you tell it to do not what you wanted it to do, because a mistake was made even before the program was written. 1.5.2 Implementation Errors Implementation errors are made when the flow charts are converted to program statements. These errors are also examples of the computer doing what it is told to do. Implementation errors also include forgetting to tell the computer something, such as not defining a memory location in an assembly language program, or incorrectly writing a PASCAL procedure. 1.5.3 Typographical Errors Typographical errors are made when entering the program into the computer. They are usually the easiest to find because the computer can detect most of them. For example, if you are programming in BASIC and you get SYNTAX ERROR IN LINE 200 the BASIC interpreter telling you that you have made some sort of a typographical error in line 200. If however, you type the wrong symbol for a variable, such as A0 instead of A1, what you have done is typed in A0 when you meant to enter A1. This is perfectly valid as far as the computer is concerned. It didn't know what you wanted, it only knows what you tell it. This kind of typographical error falls into the category of implementation errors. Copyright (c) Joe Kasser 1989