Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Hardware Hacking Programming Build Hardware

The 8-Bit Computer That's Been Built By Hand 161

nk497 writes "Forget snapping a few components into a motherboard — programming enthusiast Jack Eisenmann has made his own PC from scratch. His Duo Adept, as he's named it, features 64KB of main memory, 256 bytes of RAM and, in total, 263 lines of code for his homemade OS. Sure, it can't run Crysis, but it does run a game he's written himself."
This discussion has been archived. No new comments can be posted.

The 8-Bit Computer That's Been Built By Hand

Comments Filter:
  • Old school (Score:5, Interesting)

    by Zakabog ( 603757 ) <john.jmaug@com> on Friday June 17, 2011 @06:38PM (#36480766)

    I wasn't around for this sort of stuff but wasn't this the sort of thing Radio Shacks customers were doing 25+ years ago?

    • Yes, we were. My brother and I started with a Motorola MEK6800D1 board and took it from there. It's amazing what you can fit in 4K of RAM if you have to.
      • by Anonymous Coward

        This guy built a home-built *CPU* from TTL chips. So, no, it's considerably more impressive that using any kind of dev board with an onboard microprocessor.

        (You're starting with RAM already existing - he isn't.)

        • by j00r0m4nc3r ( 959816 ) on Friday June 17, 2011 @06:56PM (#36480932)
          TTL chips? Luxury! When I was a lad we had to use coconuts and vine to fashion NAND gates.
          • TTL chips? Luxury! When I was a lad we had to use coconuts and vine to fashion NAND gates.

            Gilligan, don't forget the NOT gate that we fashioned out of the two transistors from the radio or the infinite power source that came out of it. Now why couldn't have that satellite flown overhead a few moments after we completed the digital telecommunicator device?

            --The Professor

        • Yup, this isn't just going and getting some old 6502 or Z80 and building a new computer around it. That's been done plenty over the last thirty-odd years. This guy has actually built an 8-bit CPU. Fuck, I'd be impressed if he'd built a 4-bit processor, but this is pretty damned cool.

        • by JeremyR ( 6924 )

          That's pretty much what we did in a CS/EE class. Designed a CPU from scratch, put the microcode on a FPLA and used a bunch of supporting TTL chips. If I recall correctly, it was 16-bit, but it might have been 8. He has taken it a step further with video output, and that's impressive (at least to me).

          • Same here, in computer design. Used some PLDs with ridiculous gate limits and at the end of the project, we were playing a counting game via serial

          • by tibit ( 1762298 )

            These days, you don't even need an FPGA. Take any fast multicore chip like Parallax Propeller [parallax.com] or fast multithreaded chip like XMOS XC-1 [xmos.com] and you can emulate pretty much any retro 4 or 8 bit CPU at native or faster speed. With video output. All pretty much single chip -- all you need is a clock crystal and some voltage regulators. The propeller has 8 completely independent cores called cogs, each with 4 kbytes of dedicated RAM, and 32 kbytes of shared RAM in so-called hub. XC-1 has hardware multithreading wit

        • Re: (Score:2, Interesting)

          by Anonymous Coward

          Err, I may have confused this with another story - from here [homebrewcpu.com]. If I have, sorry.

          Neverthless, it seems this story is, however, not unique even in the modern day.

        • by qubezz ( 520511 )

          The 1970's called, it said you are low tech. Here's an example of a single TTL board [clemson.edu] from a VAX. There must be about 300 individual TTL chips from the 7400 series on it (where one chip has 4 nand on it, etc). The left 29 boards in this VAX [museum.com]are the cpu.

          It is very noble to build your own CPU architecture with your own instruction set, however building CPUs out of gates in individual chips is just an exercise in wasting money when you can do the same thing on FPGAs, like the guy that built an entire Cray-1 [chrisfenton.com] on

    • Pretty much. I would've been impressed if he had done it from actual transistors not full-blown ICs, but given CompEng students make stuff like this in school (admittedly, with FPGAs, but the concept is the same) it's not as complex as it looks. I'd say the hardest part is probably the output to monitor. The wiring would be tedious, but not unthinkable.

      • Re:Old school (Score:4, Insightful)

        by msobkow ( 48369 ) on Friday June 17, 2011 @07:00PM (#36480976) Homepage Journal

        The hard part is finding the loose wire.

      • by Arlet ( 29997 )

        It's a lot easier with FPGAs, because the synthesis tools will make sure you meet all the timing constraints and setup/hold requirements, or will complain if it can't. They will also complain if you connect two outputs together, or do other silly things. With real hardware you have to do all of that yourself.

    • Yes. I built a simple CPU+memory system in school around 1979 or so. That's 32 years ago.

    • I wasn't around for this sort of stuff but wasn't this the sort of thing Radio Shacks customers were doing 25+ years ago?

      Indeed. I built a 6502 machine with the help of an electronics magazine, starting with actually etching my own circuit boards. It had an hexadecimal display and keyboard (thanks to manual Dymo of old), only the imagination was the limitation. And yes, I did write a game for it.

      • I had a book, back in the long ago, called "How to Build a Microcomputer and Really Understand It". It walked you through building a 6502 based machine. It had PCB printouts that you could photo-resist onto circuit boards. You would build dozens of "nybble cards": circuit boards with edge connectors and a binary pattern of diodes to encode 4 bits of information. By inserting these cards two at a time into edge connector sockets, you could do some simple programming. I loaned it out and never got it ba
      • by strags ( 209606 )
        To be fair, that's not exactly comparing apples and apples. You didn't actually MAKE the 6502 itself, which is much closer to what he's done here. That said, yes, plenty of others have done this.
      • by qubezz ( 520511 )
        Is that you [oldcomputers.net], Chuck Peddle [wikipedia.org]?
    • Circa 1975-76, this was just a design exercise at the tail-end of our logic design class--we just called it a minicomputer then, using SSI and MSI TTL modules--it would have been a bit tricky to come up with 64K of storage then, since 1Kbit to maybe 4Kbit chips was state of the art. If anything, you do have to commend the guy, not for design, but for getting such a large number of proto-boards and all those aggravating wires hooked up and get all of it working. A wire-wrapped version would be more compact
    • There are two differences. One, he didn't use a prefab microprocessor - he built one from gates, counters, etc. And two, his website hasn't collapsed from the slashdotting.

      • by drolli ( 522659 )

        Well. i thought he soldered it from discrete elements, like transistors. that would be something.

    • Building a functioning 8-bit CPU and writing apps for it was required for graduating from our CS program at UC San Diego... not seeing why this is newsworthy.

      • by jhoegl ( 638955 )
        I think its the fact that there are still "web rings"... WTF?!?!?
    • Not really, no. 25 years ago you could buy a Z80 or 8085 in that kind of store for cheap, so there was no need to build your own CPU from TTL.

      Yes, I was there.

  • by John.P.Jones ( 601028 ) on Friday June 17, 2011 @06:39PM (#36480782)

    Ever since Cosmos I can't take the phrase 'from scratch' seriously.

    Also there is this TED video where a guy tries to build a toaster from raw materials...

    • Obviously, it's easy peasy to create your own universe... I mean, it's as easy as saying "let there be light".... Ooops, damned, I did it again. *sigh*

    • by tylernt ( 581794 ) on Saturday June 18, 2011 @01:14AM (#36483060)

      Also there is this TED video where a guy tries to build a toaster from raw materials...

      I don't think people appreciate the "tech tree" (to use Starcraft parlance) you have to walk down to get to the simplest of modern household items. The toaster is a good example, but now imagine starting from zero -- you can't even start with iron ore, because you don't have any tools to mine it with! So start with banging rocks to get something sharp you can use to cut down a tree, so you can make a handle to make a stone axe. Hopefully this is enough to get some iron ore, but now you also need to make something to smelt your ore in, such as a bloomery. And for that, you need charcoal. And for that...

      Basically, the TED guy making his toaster cheated by used modern tools to get his raw materials. And even with cheating, his toaster never toasted any bread.

      The tech tree for a dollar store pocket calculator is staggering, let alone a Space Shuttle. I don't think many people are conscious of this when they toss that toaster in the garbage and spend $10 on a new one.

  • holy shit, who knew that web rings still existed in the 21st century!

    • holy shit, who knew that web rings still existed in the 21st century!

      Throwback to the 80's computer. Throwback to the 90's webpage.

    • As a web ring, it sucks. I want to open the next site in the ring in a new tab, but it's Javascript controlled. It doesn't allow me to do it.
  • They want their homebrew computer back.

    As an aside: this is obviously someone who uses a Mac to be a hipster...

    • by kwoff ( 516741 )
      The mid 1990s also called, and they want their WebRing back.
      • This is one of the best web rings though. Some really cool systems in there. Snag is many of those pages are served by the actual homebrew CPUs themselves, so they could be a bit slow (oh no, they're being slashdotted!).

  • Reminds me of the Maybe systems that MIT undergraduates build by hand in 6.004 (or used to, when I was involved) that were then programmed to emulate about 3 or 4 different architectures.

    • Yeah, there's a upper-level undergraduate course that does single-board computers with a 8088 MPU and some supporting hardware. It's a mess and I personally believe that the course should be changed to give a "interfacing with reality" bent to it, as a single MCU can be tuned to do the same (external memory bus, etc) and you can go beyond the "look I made a light blink" to "Look I can actually do something useful with this thing".
      • Yeah, there's a upper-level undergraduate course that does single-board computers with a 8088 MPU and some supporting hardware. It's a mess and I personally believe that the course should be changed to give a "interfacing with reality" bent to it, as a single MCU can be tuned to do the same (external memory bus, etc) and you can go beyond the "look I made a light blink" to "Look I can actually do something useful with this thing".

        Upper-level undergraduate course at my university, which is not MIT.

    • 6.004 was awesome, both taking it and helping teach and debug other student's projects as a lab assistant. It's was a great introduction to the basic skills required to be a firmware engineer in today's job market, since you really got to figure out, clock by clock, how a CPU operates.

  • not quite... (Score:1, Insightful)

    by Sebastopol ( 189276 )

    ...He still used a microprocessor in an integrated circuit. In college back in the 1980's some ubernerds built a 4004 with discrete transistors.

    But still, i give this person _HUGE_ props, breadboarding a circuit that complex is very, very, VERY time consuming amount of debug. it would drive most people insane, literally, it would break their brains to try and debug this.

    • He still used a microprocessor in an integrated circuit.

      Really? Which one did he use?

    • by Osgeld ( 1900440 )

      agreed, I have constructed a "computer" around a Z80, and there for a while ... wow. from countless hours tracing to out right fury as I yanked entire sections out with both fists and even in the end it never worked correctly

    • You appear to be wrong. He built a minicomputer from logic chips. Look at the schematic - why would there be an instruction decoder if he used a microprocessor?
      I wonder why he didn't wire-wrap it. This is a crazy way to build a computer - thousands of blue wires, any one of which could fail with a loose connection at any time.
      • my bad, i meant to say that he used "just" integrated circuits not a microprocessor.

        still, that's a big "just"

    • Re:not quite... (Score:4, Insightful)

      by Darinbob ( 1142669 ) on Friday June 17, 2011 @08:06PM (#36481572)

      It looks like he used actual 74xx series TTL chips to make the CPU. From the parts list he isn't doing microcoding, and isn't even using ALU or bit-slice MSI chips. It's the real thing.

    • by CODiNE ( 27417 )

      Oh good, that'll be next on my list of things to try and drive myself insane with.

      1) Hinton's cubes
      2) breadboarding an 8-bit computer

      Please let me know if you think of any other surefire ways to go nuts. Okay, now off to watch Brainstorm again.

  • by Scarletdown ( 886459 ) on Friday June 17, 2011 @07:03PM (#36481004) Journal

    I have to ask (since at the time I am writing this, no one else has done so yet)...

    Does it run Linux?

    And if it does, just imagine a Beowulf cluster of these things.

  • ... the portable version?
  • This computer is of no use if it can't store my bitcoin wallet.
  • by Anonymous Coward

    Here's one that running Minix: http://www.homebrewcpu.com/

  • dont forget to scavenge that printer memory from old dot matrix printers like the old days ;-)

  • I used to do this sort of thing 30-35 years ago. I remember building some of the first altair machines and hacking some cp/m code with Neil Colvin in his basement. I got together with a couple of guys to wire wrap one of the first (if not THE first) S100 bus graphics boards. We used to cobble up single board computers like this all the time, but they cost thousands of dollars to build. Good times, good times. Glad its the 'in' thing now, I feel like I know something ahead of the curve.
  • If this was Minecraft someone would wander by, grab and handful of those pretty blue wires and rip them out.
  • If you wish to make a computer from scratch, you must first invent the universe.
  • by Jim Buzbee ( 517 ) on Friday June 17, 2011 @07:35PM (#36481300) Homepage
    Reminds me of the one my brother built here [homebrewcpu.com] except my brother's computer runs Minux.
    • I saw that once before, and if I remember correctly, I said aloud, "That's fucking cool!" Really, that thing is amazing. I like that it actually runs Minix. Probably the most complete "homebrew computer" I have ever seen.
  • by iamacat ( 583406 ) on Friday June 17, 2011 @07:42PM (#36481362)

    Everyone has right to their own hobbies, but think what can be accomplished with the same amount of labor and modern parts. Instead of making a CPU from hundreds of TTL gates, build a personal supercomputer from hundreds of ARM processors and custom operating system to effectively use that power for virtual reality or physics simulations. Hobbyists who has done this decades ago were futuristic not retro, creating devices that were not widely available, at least to private individuals.

    • It's the right direction for demonstrating that computers are based on discrete logical components, no matter how tiny and embedded in a chip; and the right direction for demonstrating that, given enough time and information, it would be possible to truly understand any digital device.

      Hmm, I wonder how many TTL chips I would need for a nice little PDP-11...

      • Years ago, I had some youngsters ask what microprocessor our schools PDP-11/34 used. They had a tough time grasping that it didn't use a microprocessor at all and that the CPU was actually a couple of circuit boards. I think that they thought that a microprocessor was some sort of magical thing that couldn't be implemented in another fashion.

        Hmm, I wonder how many TTL chips I would need for a nice little PDP-11...

        I used to own a PDP-11/10. The CPU was two fairly large circuit boards. I also got a set of schematics when I got the computer. Sadly, they and the computer parted

        • I don't recall very well, but wasn't the 11/34 the first PDP-11 using the Western Digital mpu chips? The ones that DEC tried to run WD out of business for, so they could buy the company cheaper than the chips?

          • (replying to self) - according to Wikipedia [wikipedia.org], the LSI-11 was first used in the 11/03, which was on the Qbus architecture, not the Unibus used by the 11/34. I stand corrected. :)

  • He wired up a bunch of ICs. For "From Scratch", google 'toaster from scratch...

    • by Arlet ( 29997 )

      Or, as the late Carl Sagan said: "if you wish to make an apple pie from scratch, you must first invent the universe".

      You have to start somewhere, though.

  • Another interesting computer [pdx.edu] built from electrical relays.
  • in the old days doing things 'by hand' was the only way it got done. But i guess ill give him credit for the patience that it requires to do this sort of stuff. ( does bring back old memories however.. sore fingers, smell of burnt solder in the air.. )

  • by tftp ( 111690 ) on Friday June 17, 2011 @10:36PM (#36482512) Homepage

    He certainly had some fun building this. The wires are used probably because he wasn't sure that everything works right; it's much easier to rework a wire connection than an inner PCB trace.

    I suspect he is a strong amateur, but not a professional. A professional would design the whole thing in a simulator first, and once that works he'd implement it on a PCB (if not an FPGA.)

    I personally haven't built processors, but I built a few peripherals for PDP11/LSI11, all from discrete logic. And I serviced IBM 360/370 systems [long time ago] - they were built exactly this way, but were a bit more modular.

  • I somewhat dislike the Slashdot commenting system. So for the 3rd time, since I want to stand up for myself, to avoid having a "hidden" comment I am going to restate: I, Jack Eisenmann, built the DUO Adept in highschool, and I have no formal education in electronics. I learned everything by experimenting with breadboards, getting tips from online users, and poking around Google. I don't feel that the blurb does justice for my accomplishments.
    • As someone who had the same thing happen, let it go man. In 5 years you will be wondering how you had your head that far up your ass commenting on this and identifying yourself.
    • Oh, you made an account. See my other reply [slashdot.org]

    • Don't worry, we heard you. :) So, what's the next big thing?

      Someone above posted this idea, what do you think about it?

      Instead of making a CPU from hundreds of TTL gates, build a personal supercomputer from hundreds of ARM processors and custom operating system to effectively use that power for virtual reality or physics simulations.

    • by JeremyR ( 6924 )

      It is an impressive accomplishment, and you should be proud of that. It demonstrates initiative, patience, imagination and maybe a bit of ingenuity.

      There are some responses here that amount to "BFD" and you would do well to ignore those. There are others that put the scope of such a project in context, comparing it to what an undergrad student might do in a CS or EE class (and without much more experience or education than you have). I believe these responses are intended not to diminish what you've achieve

  • Very similar to the "Educ-8" TTL computer from Electronics Australia mag in 1974

    see http://en.wikipedia.org/wiki/File:EDUC-8_Microcomputer.jpg [wikipedia.org]

    and http://www.sworld.com.au/steven/educ-8/ [sworld.com.au]

       

  • Actually, lots of scratches on my back by my wife, who then carried our son to term. Does that count?
  • This story reminds me of a quote from one of Tanenbaum's books.

    In theory, after fully understanding this chapter, the reader should be able to go out and buy a large bag full of transistors and build this subset of the JVM machine. Students who successfully accomplish this task will be given extra credit (and a complete psychiatric examination).

    I'm not sure if the guy from this article qualifies, but it sounds like he is close.

  • In an alternate universe where I missed theoretical CS lectures, I wanted to build a (Universal) Turing Machine from scratch... but soon ran out of tape. Then I wanted to build a Register Machine from scratch, but ran out of memory before realizing that it was equivalent in power to the Turing Machine I couldn't build. Then I got hold of an Oracle (sadly not the company), but this Oracle-augmented machine STILL had limitations and couldn't compute EVERY imaginable function. Every time I wanted to showcase t
  • 263 lines on the display. The article describing it gets it wrong, too, but there's an OS listing at the site.

  • I wish my laptop booted as fast as his computer!

I have hardly ever known a mathematician who was capable of reasoning. -- Plato

Working...