I must have cut thousands of these "pits"...

I thought about waking my wife...

The Mark-8 Minicomputer

Part 2: The Design
Last revision: March 21, 1999, Copyright © 1999 Jon Titus


My prototype of the entire computer was wired point to point using Veroboard. Experimenters from the 70s may remember Veroboard because many people favored it for experiments and projects. The board provided a grid of small holes on 1/10th-inch centers, which were just right for integrated circuits and other components that used, or could accommodate a 0.10-in. spacing between leads. One side of the Vero board contained narrow strips of copper that ran down the board from hole to hole in one long strip. If you wanted to break a run of the copper, you used a small hand tool which looked like a small countersink bit. The tool cut through the copper to isolate a specific run. The tool cut through only the copper and a bit of the underlying phenolic board material. The idea was to cut through the conductor, not all the way through the board.

I must have cut thousands of these "pits" in two large pieces of Veroboard. Luckily I caught on right away and used a drill bit in my Unimat (set up as a drill press) to cut the pits. The entire surface of both Veroboards were set up for 14-pin and 16-pin IC sockets. The exception was the 8008 chip, which required an 18-pin socket. Because IC sockets were expensive, I used Molex strips to create the sockets. The Molex strips provided individual, uninsulated IC socket pins in long strips on a roll. I cut the strips into 7-pin and 8-pin segments and soldered them in place to form inexpensive IC sockets. I figured that if the computer circuit didn't work, at least I could salvage the chips.

After making up the sockets, I plugged in all the chips except for the 8008. Having the chips in place helped me make the proper wiring connections. Without quick reference to what chips were where, I would have often lost my place.

The prototype allowed two contacts in the Veroboard per IC pin, so I had to carefully plan my wiring. With only two contact points, all the signals had to run in a daisy-chain fashion, from one IC to the next.

After I completed wiring the prototype, I tested the various sections of the computer circuitry to be sure that these subsections worked properly. Without exception, they operated as designed. After thoroughly testing these circuits I carefully plugged in one of the three 8008 chips that Intel had sent me and kept my fingers crossed. I plugged in the power supply and the computer worked the first time. I couldn't believe it. I had built a lot of hobbyist circuits and kits, and almost nothing ever worked the first time. I was amazed that such a complex circuit worked. My first "programs" were simple ones that put the computer into some simple loops--jumping from one location to another, and back. I used the single-step function to make sure the operations were really running.

I did have one bent pin on an IC that causes a problem, but I quickly found the pin, which was rubbing against an adjacent pin and causing a short. The prototype was up and running about 2:30 in the morning and it was all I could do to contain my enthusiasm. I didn't sleep a wink the rest of the night. I thought about waking my wife, but decided against it. She had no interest in computers and would have had no interest in the fact that I had built a working computer.

Finally I had my own computer. The point-to-point wiring was neat, but the integrated circuits were so close together that it was difficult to use the computer for more than testing the basic design. That prototype contained three banks of 256-byte memory, so it let me experiment with less than a kbyte of memory.

Radio Electronics

After thoroughly testing the prototype, I talked with Larry Steckler at Radio-Electronics magazine about publishing information about the Mark-8 (it now had a name) as a construction project. I also approached Popular-Electronics magazine, but they showed no interest. Larry was a bit skeptical, so I designed circuit boards, got prototype boards, made a few modifications, and put the boards in a ready-made metal box to give it a professional look. Larry visited Blacksburg one day in late winter or early spring in 1974 to see that the computer actually worked. We had it set up with Don Lancaster's TV Typewriter, a keyboard, a digital-to-analog converter (DAC), and an oscilloscope. The DAC operated based on simple commands from the keyboard and it would change the shape of the waveform on the oscilloscope accordingly. I think my brother Chris and I had set up the DAC to produce a triangle wave, sawtooth waves, and a square wave. Generating a sine wave was a bit beyond our capabilities. We didn't want to load in by hand a sine look-up table, nor did we want to write a sine-calculator algorithm in assembly language for the 8008.

Based on his experience with us in Blacksburg, Larry agreed to publish the Mark-8 article, so I set to work writing it and I also wrote some extra experiments that R-E published in a separate booklet. The experiments covered how to use the computer with various I/O devices so that the builders--if there ever were any--could see how they could use the computer to do useful things.

Radio Electronics

The cover of the July 1974 Radio Electronics, in which the Mark-8 made its first appearance. Click for a larger image.

I had to scavenge the prototype to obtain the switches and integrated circuits for the final version, so the hand-wired boards eventually went into the trash. People have asked why I didn't save them. Remember, I was a graduate student who was earning very little money at the time.

In the spring of 1974 I lugged the computer, including the external power supply to New York City so that R-E could have a photographer take several shots for the cover. I'm not sure of the exact dates, but it was just before the weekend in '74 when Virginia Tech's basketball team won the National Invitational Tournament.

backThe Beginning
Copyright ©Andrew Davie (site design) and Copyright ©1999 Jon Titus.
Contents may not be reproduced in any form without permission.
The After Effects