[editor : Carl T. Helmers Jr.] [publisher : Virginia Peschke, Manfred Peschke] [art : Mary Jane Frohlich, Wai Chiu Li, Ellen Shamonski-Bingham] #Magazine
#Abstract
Model railroading can give the computer hobbyist an action packed computer application. Before tackling the job, however, it is important to have a firm understanding of what's involved in the design of a model railroad. In his article, How to Computerize Your Model Railroad, David C Brown explains in detail the problems faced by the model railroader and points to ways in which they can be solved. He then goes on to cover the requirements for microprocessor interfaces to the model railroad and some thoughts on the software of an operating system to give realism to the model.
Sometimes a small amount of hardware can speed up software considerably. A perfect example is provided by Tom Hall in his article which shows how This Circuit Multiplies. This circuit is a hardware multiplier which takes 8 bit operands and replies (ten clock periods later) with a 16 bit product. Here is an example of putting an inner loop into hardware, thereby speeding up an operation.
The model railroad is an ideal way for the personal computing enthusiast to enter the fascinating world of real time control: such a system is realistic but manageable. Authors Jack Hart and Ed Badger show you how they did it in A Train Control Display Using the LSI-11 Microcomputer.
Sometimes a bit of serendipity falls out of an application or project. Ken Barbier describes one such case in the form of his technique of using a television display circuit's ability to generate a raster with various synchronous patterns to debug and verify its operation. Read The TV Oscilloscope.
There are many ways to make a computer talk, but how do you get it to listen? Speech Recognition for a Personal Computer System discusses a topic which has fascinated and frustrated experimenters for years. Author James Boddie of Bell Labs (the pioneer researchers in the field) covers the history of the subject and presents a practical system which can be realized by a personal computing experimenter.
Real world systems may not be perfect, yet programming demands perfection. (If we drop a bit in one machine instruction, it becomes another instruction altogether.) As a background discussion of a large subject, W Douglas Maurer presents some information on How to Pick up a Dropped Bit using some of the techniques of error detection and correction.
What's a sure cure for memory megalomania? Why, Give Your Micro a Megabyte as outlined in Robert Grappel's short tutorial on some large memory design techniques which will prove adaptable to microcomputer systems. There should be virtually no reason to complain if his ideas were put into practice more universally.
An Introduction to Numbers, by Webb Simmons, serves as an introduction to the concepts of fixed, scaled and floating point numbers. Here you'll find some basic forms for each type, how the forms differ from each other, and how each can be used.
If you've ever been frustrated by the drudgery involved in relocating machine language programs with nothing but toggle switches, then Leor Zolman's A Machine Code Relocator for the 8080 is for you ! Just enter six key pieces of information and the program does the rest automatically, even fixing up all your address references.
For beginners first learning about computers, we often get requests for some basic information. In BASICally BASIC, Robert Baker gives an in formal introduction to the nature of the BASIC language and its uses in programming. Finding out what a typical BASIC can do is a good starting point in your personal evaluation of products available in the personal computing marketplace.
Given latitude and longitude of two points on the earth, how do you calculate the distance and bearing? If you use a little BASIC program by Rene Pittet, you can answer the question of How Far - Which Way? using a small processor.
If you have a Southwest Technical Products' TVT II, there is a simple circuit that you can add that will give you manual and computer control over the cursor's movement, erase and bell functions. In his article, Add Cursor Control to Your TVT II , Brother Thomas McGahee describes this simple circuit which can be attached to a TVT II.
Poor KIM . If one puts KIM-1 inside a fancy case, the built-in jewels of keyboard and displays get hidden . But Robert Grater comes to the rescue by Giving KIM Some Fancy Jewels, which consist of a remote set of displays connected to the processor by cable.
[author : Carl Helmers] #Edito
Extract : « In a press release dated March 9 1977, Texas Instruments Inc made it official: a commercial bubble memory product is now available for general use, initially in sample quantities priced at $200 per chip. The device is the TBM0103, a single chip 92,304 bit storage element composed functionally of 144 shift registers with 641 bits per register. (Internally, according to TI, there are 157 registers, but the large size of the chip carries with it the attendant probabilities of defects, so discretionary manufacturing steps are used to connect 144 good registers per package.) The device has a complicated mechanical structure which includes two permanent magnet bias field elements, two orthogonal coils which modulate the bias field to produce a net rotating magnetic field in the garnet film bubble chip, and, of course, a magnetic shield to isolate the chip from external stray magnetic fields. The dynamic data characteristics of the chip are a bit transfer rate of 50,000 bits per second and an access time of 4 ms setup prior to transferring a 144 bit word of data into or out of the device. Conceptually, figure 1 illustrates what one of these memories looks like to a designer or programmer. This figure was constructed based on the limited information about overall characteristics in the press release. The procedure for writing a 144 bit word consists of serially loading the 144 bits into the major loop, then transferring all 144 bits into the minor loops; reading is done in reverse order by copying the currently positioned bits of all 144 minor loops into the major loop then serially reading the major loop. The result is a nonvolatile file storage device which has 641 blocks of 18 bytes per block, a total of 11,538 bytes in one 14 pin dual in line package measuring 1.0 by 1.1 by 0.4 inches (2.5 by 2.8 by 1.0 cm). Using a volume parameter for memory density, this elephant of a memory stores 200,000 bits per cubic inch (13,000 bits per cubic centimeter) in a nonvolatile magnetic medium.
With this short introduction to the existence of a new low cost, high density memory system, what are the ramifications of such a part for the personal computing user? [...] »
TABLE OF CONTENTS
[theme : Applications] [author : Brown] #Electronic #Interface #Entertainment
Extract : « One of the many uses of computers, and especially personal computers, is in the area of real time process control operations. An enjoyable way of experimenting with this area would be to have your small computer control a model train layout. Those of us who attended the Personal Computing 76 convention held in Atlantic City NJ in August undoubtedly saw the Digital Equipment Corporation's display of a train layout controlled by an LSI-11 computer. There is no reason why you could not do something similar with your own system. [...] »
[theme : Arithmetic Hardware] [author : Hall] #Electronic #Mathematics
Extract : « We all know that algorithms for multiplication can be implemented perfectly well in software. But such programmed approaches to arithmetic operations can often prove quite slow. There are a number of possible approaches to gaining speed in operations involving multiplication. These may vary all the way from just tightening up the code to putting a full hardware floating point arithmetic unit into the processor. The latter is often expensive and difficult to build. One approach which is midway between these extremes is to implement the "inner loop" of a software algorithm as some special purpose hardware. An example of this technique is this simple binary multiplier which takes two bytes of information and supplies a 2 byte product value, leaving all the other functions of a full arithmetic package to software. By speeding up the multiplication operation, the entire arithmetic software will run much faster whenever it involves multiplication. [...] »
[theme : Hardware] [author : Barbier] #Electronic #Display
Extract : « Although conceived and designed independently, the display circuitry in my CRT terminal is virtually identical to that of CW Gantt in his article "Build a Television Display" on page 16 of June 1976 BYTE. The major difference is that I did not use the MM5320 sync generator, and my display is therefore not interlaced and its horizontal sweep is not exactly at the usual TV horizontal rate. My display, shown in operation in photo 1, consists of 24 lines of 64 characters, with a frame consisting of 270 raster lines repeated 60 times per second. With this formulation I have a time allotment for 27 character lines with ten raster scan lines per character. The extra three character line positions represent overhead for vertical sync. [...] »
[theme : Peripherals] [author : Boddie] #Audio #ArtificialIntelligence #Book
Extract : « The August 1976 issue of BYTE presented to the personal computer user an introduction to half of the problem of man-machine communication by speech: machine generation of speech. Speech synthesis has been investigated since at least the 1800s. High quality speech can be produced by a computer if it decodes utterances that have been efficiently coded and stored in memory. Poorer quality, but sometimes acceptable speech can be synthesized by electronically simulating the human speech producing mechanism.
Speech communication in the other direction, machine recognition of speech, is usually considered to be a harder task. The purpose of this article is to present some of the problems and to introduce some solutions that may be realized by the personal computer enthusiast.
The ideal speech recognizer would be a machine that could accept casually spoken sentences from many speakers with different dialects and transcribe them in real time into written text or translate them into commands for controlling some function. Unfortunately, there are at least five major problem areas in speech recognition that are still incompletely solved. [...]
[...] BIBLIOGRAPHY
1. Davis, Biddulph, and Balashek, "Automatic recognition of spoken digits," Journal of the Acoustical Society of America, volume 24, pages 637-642, November 1952.
2. Ross, "A limited-vocabulary adaptive speech recognition system," Journal of the Audio Engineering Society, volume 15, pages 414-416, October 1967.
3. Gilli and Meo, "Sequential system for recognizing spoken digits in real time," Acustica, volume 19, pages 38-44, 1967.
4. Teacher, Kellett, and Focht, "Experimental, limited vocabulary speech recognizer," IEEE Transactions on Audio and Electroacoustics, volume AU-15, pages 127-130, September 1967.
5. Purton, "Speech recognition using autocorrelation analysis," IEEE Transactions on Audio and Electroacoustics, volume AU-16, pages 235-239, June 1968.
6. Bezdel and Bridle, "Speech recognition using zero crossing measurements and sequence information," Proceedings of the IEEE, volume 116, pages 617-623, 1969. »
[theme : Software] [author : Zolman] #Listing #Assembly #Programming
Extract : « Many microcomputer hobbyists pass through an early stage in the development of their systems where machine language is the only mode available for programming. My system configuration at that point included an IMSAI mainframe and 4 K memory, a VDM-1 video driver and monitor, and a Tarbell cassette interface. The big problem when programming something like LIFE for the VDM in machine language occurs when it becomes necessary to insert a routine (call it routine y) between locations x and x+1 in memory. [...] »
[theme : Applications] [author : Pittet] #Listing #BASIC #Transport
Extract : « Did you know that your computer can answer the questions of how far and which way for you in microseconds for macrodistances? At least my SWTPC 6800 can with its 8 K BASIC high level compiler language.
The usefulness of this program will be apparent to anyone who sails a boat or flies an airplane. By inputting longitude (degrees and minutes) and latitude of the starting point first, and the destination second, the program will print out the distance in statute miles, or nautical miles with a minor change, and the true bearing. Look up the variation of magnetic north in your area and add that correction for a magnetic compass heading; or better yet, add it into the program. [...] »
[theme : Hardware] [author : McGahee] #Electronic #Display
Extract : « If you own a TVT II and would like a simple circuit to provide for operation of the cursor and erase functions, then consider the circuit shown in figure 1. This simple circuit will accept ASCII control characters and decode them in a manner suitable for controlling all of the cursor control and erase functions, and it even has a few outputs that can be used to provide user defined functions, such as enabling external devices or turning on bells and whistles. [...] »
[theme : Hardware] [author : Grater] #Interface #Display
Extract : « When my KIM-1 arrived I was excited, but a little dubious; after all, I had read all those microprocessor books and still didn't know beans about programming. Well, MOS Technology's excellent manuals took care of that and I'm on my way to knowing what I'm doing. When I say excellent, I mean if I learned it, anyone can!
But KIM has a fundamental problem! I like KIM, but I just can't see myself sitting there with a naked board gathering dust and me dropping ashes all over it while I sweat out a program. Also with MOS Technology's welcomed announcement of the KIM-2 and KIM-3 programmable memory boards and the coming KIM-4 mother board, it's really time to start thinking about putting KIM in a nice enclosure. But in order to bury KIM in an enclosure, I had to somehow bring the displays and keyboard lines out. [...] »
[theme : Reference Materials] [author : Borrmann] #Assembly #Programming
Extract : « Here is a quick reference chart that describes the opcodes for the Motorola 6800 microprocessor. This chart gives all of the hexadecimal codes, including the ones that are unimplemented. The codes are given in a four line format as shown in figure 1.
The first line has three values: a, b and NN. NN is the hexadecimal op code; a is the length, one, two or three bytes, of the instruction; b is the length of execution in machine cycles for each instruction.
The third line contains the condition flags that are affected by the execution of the instruction. [...] »
[theme : Applications] [author : Hart-Badger] #Electronic #Interface #Display #Entertainment
Extract : « In order to demonstrate the application of a microcomputer to real world situations, a model train system was developed which is controlled by a Digital Equipment Corporation LSI-11 microcomputer. Two model trains were used for the demonstration — one adhered to a fixed schedule while the other was directed along a path to make it avoid the first train. [...] »
[theme : Error Theory] [author : Maurer] #Storage #Encoding #Algorithm
Extract : « The phenomenon of the dropped bit causes difficulties in two distinct areas of computer technology: in the recording of data on tape (or disk, or the like), and in the transmission of data from one place to another. Suppose, for example, that we are recording one hundred 32 bit words on a tape. Out of the 3200 bits that are to be written on the tape, there is a nonzero chance that at least one of them will be wrong. Either it will be recorded as a zero, when it should have been a one (a dropped bit) or it will be recorded as a one, when it should have been a zero (an added bit). Even if all 3200 bits are recorded correctly, there is still a nonzero chance that the next time we read this tape, we will read at least one of the one bits from the tape as if it were a zero, or one of the zero bits as if it were a one. In such a case we again speak of dropping a bit, or adding a bit. Often both dropped and added bits are referred to, generically, as dropped bits, and we shall continue to do so in this paper. [...] »
[theme : System Design] [author : Grappel] #Memory #Electronic #Algorithm
Extract : « What is it that makes a microcomputer "micro?" There are many factors involved, but one of the most crucial is the small memory available in most hobby computers. I had scarcely gotten my Altaic up and running with 8 K of memory before I wanted to write programs that exceeded its storage capabilities. Paging programs in and out from dual cassettes became a major nuisance, so I gathered my pennies and bought another 4 K memory board. In less than a month I needed even more storage. It is a fact of programming life that one must trade storage space for execution speed in programs. I tried to write a program in BASIC to play Mastermind on my Altair. Mastermind has 1296 possible moves, each move composed of four numbers. Storing this optimally would require about 5 K bytes; the BASIC code took nearly 6 K bytes, leaving me almost no space for my program. [...] »
[theme : Software] [author : Simmons] #Encoding
Extract : « The concept of fixed point numbers, scaled numbers and floating point numbers originated in the scientific computer environment at a time when a computer was generally considered to be either a scientific computer or a business computer rather than the general purpose computer of today. Business computers used fixed point numbers but designers felt no need to describe them as such because that was the only kind of number used. Some business computers used decimal arithmetic rather than binary arithmetic and allowed a variable amount of decimal digits for various variables and values of programs. Some business computers referred to their decimal digits as characters and regarded every character whether or not it was a decimal digit as having a decimal digit value. But here we're concerned with the wider concepts of number representation in a general purpose computer. [...] »
[theme : Software] [author : Baker] #BASIC
Extract : « The BASIC programming language was developed in the early 1960s at Dartmouth College as a conversational, problem solving language. It has wide applications in scientific, business, and educational environments since it can be used to solve both simple and complex mathematical problems from the user's terminal. Due to the small number of clearly understandable commands and programming statements required, BASIC is one of the simplest languages to learn and use. [...] »
#Book
Extract : « Scelbi's "6800" Software Gourmet Guide & Cook Book, by Robert Findley, Scelbi Computer Consulting Inc, Milford CT, 1976, 5½ by 8½, 226 pages. Softbound, $9,95 postpaid. [...]
Dr Dobb's Journal of Computer Calisthenics and Orthodontia. Published by People's Computer Company, Box 310, Menlo Park CA 94025. Single issue, $1.50; one year (10 issues), $10. [...]
Fundamentals and Applications of Digital Logic Circuits by Sol Libes, Hayden Book Co, Rochelle Park NJ, 1975, 184 pages. Paperback, $6.95. [...] »
Extract : « Last month, we published Kenneth Welles' article on the driver software for his Economy Floppy Disk. This issue, we continue that article with the machine readable bar code representation of the object code for the programs. These programs were presented as complete symbolic assemblies of the programs on pages 92-96 as "listing 1" of Dr Welles' article in June 1977 BYTE. [...] »
#Association