2015
04.09

About a week ago I commented on an issue zaphod was experiencing in his four bit processor build, mentioning something I had realized in a similar project of my own. Zaphod encouraged me to post my own project. I have been reluctant to do this because I do not like to let others see my work before I feel like it is ready. This project largely exists in my mind, but what does not exists as approximately 71 MB worth of files on my NAS. At present I do not recall whether or not any of the originally breadboarded sub-projects are still assembled. I would have to go to where all of that stuff is stored to have a look.

According to the creation dates of some of my files I started this project in May of 2009. My story is very similar to Zaphod’s. At the time I felt like my understanding of digital logic was severely lacking. I decided that I wanted to improve my skills/knowledge/ability in that arena, and to that end I decided to build a computer – from transistors.

I have worked on and with computers for some time, and was most recently employed as a software developer. I am not completely ignorant about how computers work, but I should note that I am not an electrical engineer, or even an EE student. Electronics is a hobby for me, and I am almost entirely self-taught in this domain.

In other words, do not look to this project expecting to see real engineering. The primary goal of this project is, and has always been, to learn about digital logic. In this sense, this project has already been a tremendous success. I am not terribly ashamed to admit that some of my initial ideas turned out to be quite unrealistic, and what this computer will do(assuming it is ever actually completed) has changed considerably since May of 2009.

When I started this effort I set for myself a number of constraints, most of which I do not remember(this is what happens when one does not write things down). The one I do remember has since been violated, but for good reason, which I shall address momentarily. The reason I remember this one constraint is because it gave this project it’s name:
DDC-01

The constraint was this:

An imaginary scenario whereby all of the worlds integrated circuitry has been destroyed in some sort of technological doomsday. In this scenario the only resources I have access to are:

  • pens and paper
  • a large supply of discrete components
  • the ability to create PCBs
  • the necessary mechanical components

I specifically did not include books or manuals on the subject because I find it easy to imagine that there is/are some book(s) from the 60s or 70s out there that provide step-by-step instructions on how to build a computer from discrete components.1

I didn’t want to end up taking the lazy way out and just use a series of instructions. I wanted to force myself to create my own design by reasoning from first principles(which in this case is a basic understanding of electronics).

It turned out that there was way too much that I did not know for this to be a realistic constraint, and that it was, in fact, in opposition to the primary goal of the project. I decided to use the wealth of information available on the subject, but to try to stay true to the original constraint by not just blindly using the work of others.

You shall see, dear reader, the extent to which I have attempted to adhere to this goal in future project log entries. I will note, however, that when I started this project there wasn’t as much information available on DIY CPUs as I would have expected. I found a number of projects involving CPUs created from 74 series logic, but nobody attempting to build one at the transistor level.

These project logs come along quite late in this project, and since I am the one creating them I will take the liberty of leaving out some of my more…ignorant…pursuits. I will say though that the overall design has changed considerably from what I initially imagined. Many of the aspects of the expected implementation have changed considerably as well.

If you want to follow this project, but do not understand how a transistor works, do not fear. It turns out that I didn’t understand how a transistor works either. I will be attempting to document much of what I have learned along the way, and what I hope to learn going forward.

Perhaps some hapless soul in the (hopefully)distant post-technocalypse future will be able to rebuild a diverse technological ecosystem starting with only a prudently generated printout of these project logs and a sufficiently large supply of discrete components.

For those who have read the preceding wall of text and are still wandering, DDC-01 stands for DoomsDay Computer 01.
 


 

 

1Williams, Al. “400 Transistors and 1800 Resistors Form This 1967 Personal Computer.” hackaday.com. hackaday.com, 29 Oct. 2015. Web. 01 Nov. 2015. .

No Comment.

Add Your Comment