Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
Unix Hardware Linux

DIY CPU Demo'd Running Minix 313

DeviceGuru writes "Bill Buzbee offered the first public demonstration of the open-source Minix OS — a cousin of Linux — running on his homebrew minicomputer, the Magic-1, at the Vintage Computer Festival in Mountain View, Calif. The Magic-1 minicomputer is built with 74-series TTL ICs using wire-wrap construction, and implements a homebrew, 8086-like ISA. Rather than using a commercial microprocessor, Buzbee created his own microcoded CPU that runs at 4.09 MHz, and is in the same ballpark as an old 8086 in performance and capabilities. The CPU has a 22-bit physical address bus and an 8-bit data bus."
This discussion has been archived. No new comments can be posted.

DIY CPU Demo'd Running Minix

Comments Filter:
  • by Pedrito ( 94783 ) on Sunday November 04, 2007 @03:53PM (#21234137)
    This is the ultimate nerd project... The only way it could be more of a do-it-yourself project would be building it with all analog parts. I'm very impressed. The guy appears to have been really meticulous. Everything appears to be pretty well documented... I've only gone through about 1/4 of the stuff he has available. It's a lot of material. I definitely wouldn't have the patience to do a project like this...
  • Family Analogy (Score:3, Insightful)

    by vga_init ( 589198 ) on Sunday November 04, 2007 @03:59PM (#21234205) Journal
    As long as we are using the family analogy, wouldn't Minix be more like an uncle to Linux?
  • by tjstork ( 137384 ) <todd.bandrowsky@ ... UGARom minus cat> on Sunday November 04, 2007 @04:01PM (#21234219) Homepage Journal
    The guy went and built his own cpu from scratch, then ported his own o/s to it.

    Really, just don't get more hardcore than that....

    I salute him!
  • Re:Wow. (Score:4, Insightful)

    by RattFink ( 93631 ) on Sunday November 04, 2007 @04:04PM (#21234233) Journal
    Somehow I don't think the goal of this project was to build a processor to compete with commercially available processors. A small hint might be the fact that there isn't likely a huge market for a processor pushing 5lbs.
  • Re:Wow. (Score:5, Insightful)

    by serviscope_minor ( 664417 ) on Sunday November 04, 2007 @04:22PM (#21234409) Journal
    All that to get a fraction of the performance of, say, a $10 embedded CPU that can already run Linux. Nice.

    I guess you don't program computers, since you'll never be as good as, say, Donald Knuth, so you may as well give up. You don't do any sports, since you'll never by Olympic standard. No music for you either, since you're not up to the standard of Nigel Kennedy. I'm sure you have no hobbies, since someone else could do it better too. If fact, you may as well sit in a hole your entire life since whatever you do, someone will probably do it better. Come to think of it, there's probably someone out there better at sitting in a hole than you.

    Now, please hand in your geek card at the door as you leave.
  • Re:yawn (Score:4, Insightful)

    by that this is not und ( 1026860 ) on Sunday November 04, 2007 @04:34PM (#21234495)
    If you're just gonna use an FPGA, why not just design a virtual PC purely in software.

    This thing is cool. Most current 'seniors' would hold a wire-wrap gun wrong and injure themselves.
  • by tepples ( 727027 ) <tepplesNO@SPAMgmail.com> on Sunday November 04, 2007 @04:36PM (#21234515) Homepage Journal

    All that to get a fraction of the performance of, say, a $10 embedded CPU that can already run Linux.
    The $10 embedded CPU won't be all that helpful once the major processor vendors stop selling processors to makers of computers that run anything but programs that the computer maker or the government has whitelisted.
  • Re:yawn (Score:5, Insightful)

    by Detritus ( 11846 ) on Sunday November 04, 2007 @04:39PM (#21234539) Homepage
    It's more educational to do it with MSI TTL and wire-wrap. You learn something about power distribution/filtering, race conditions, fan-in and fan-out, etc. All of the analog things that you need to know in the real world.
  • by Anonymous Coward on Sunday November 04, 2007 @04:43PM (#21234567)
    Sorry, but you're fucking stupid. Do you think it takes so much less to build 74-series gates? Cobble them together from sand between cleaning the horse stable and milking the cows? Somehow, in your deluded, retarded mind, society collapses but there's still electricity and UPS delivery of parts? Not to mention refined silicon and fabbing?
    Retard.
  • by Anonymous Coward on Sunday November 04, 2007 @04:54PM (#21234659)
    People so fucking dim don't deserve my politeness. And the people who modded him insightful? Jesus wept.
  • Re:yawn (Score:4, Insightful)

    by bitrex ( 859228 ) on Sunday November 04, 2007 @04:56PM (#21234673)
    I bet you'd also tell the team who built a replica Wright flyer a few years back that they were wasting their time, and would be better off building a Zodiac sport plane kit.
  • by Tony Hoyle ( 11698 ) <tmh@nodomain.org> on Sunday November 04, 2007 @05:30PM (#21234929) Homepage
    I prefer iphoneization. It annoys the mac zealots and makes for more amusing slashdot threads :p
  • Re:yawn (Score:4, Insightful)

    by Bassman59 ( 519820 ) <andy&latke,net> on Sunday November 04, 2007 @05:44PM (#21235049) Homepage

    It's more educational to do it with MSI TTL and wire-wrap. You learn something about power distribution/filtering, race conditions, fan-in and fan-out, etc. All of the analog things that you need to know in the real world.
    Of course you need to understand power distribution and filtering for an FPGA board. And FPGA design is NOT software design (as much as people seem to think that "Verilog is like C"), so to do a proper FPGA design, you really DO need to understand things such as race conditions, fan-in, fan-out (yes, loading is important in an FPGA), as well as synchronous logic design.
  • by DAharon ( 937864 ) on Sunday November 04, 2007 @05:48PM (#21235105)
    You might think so, but it seems to me that the sheer amount of energy required just to make those IC's available to him (growing the silicon crystals, shipping them, cutting them, etching them, shipping them again, packaging them, shipping them again, etc) would make it impossible to reproduce this with "limited resources".

    But then again, if you cracked open all the electronics sitting in the garages of your average town you might come across a small mountain of TTL chips.

    Maybe.

  • Re:Wow. (Score:5, Insightful)

    by DarkOx ( 621550 ) on Sunday November 04, 2007 @06:16PM (#21235347) Journal

    All that to get a fraction of the performance of, say, a $10 embedded CPU that can already run Linux. Nice.
    Thank you for your post. I will never understand how even on a site targeted mostly at geeks people can't get that:

    Some times people do/make things they could easily buy because they want to, to learn, to feel connected to those who came before them and did it on thier own, or to just have something they built with their own hands.

    Please if you can't understand that at least don't mock others who do~!
  • Mod parent up (Score:2, Insightful)

    by Anonymous Coward on Sunday November 04, 2007 @07:05PM (#21235751)
    Anyone who's spent time debugging their FPGA designs in a lab, going bleary-eyed staring at timing diagrams, can attest to that.

    However, something has to be said about wirewrap in educational settings: It's a lot easier to make out the connections of a macroscopic object that you wrap yourself, than staring at a colorful diagram of what your usual FPGA route plotter comes up with.
  • by p0tat03 ( 985078 ) on Sunday November 04, 2007 @07:42PM (#21235979)

    If you want to exploit clever loopholes in things, go into science. As a fellow engineer I completely understand why your prof took off marks for your trick - it's bad engineering practice. You were in school to train to be a professional engineer, and with it comes certain responsibilities and mindsets. Sure, this one project was for a college course, and nobody's ever going to die from it, but in your school projects you are expected to show the same due care and diligence that would otherwise be expected of you in the workplace.

    A better course of action would be to document the loophole and suggest in your documentation that, in certain, very controlled circumstances, this can be used to optimize performance (but it's a PLC, seriously, performance?). As engineers we're expected to do things by the book, following accepted standards, and if we deviate from it we are to document it fully with gigantic red underlines or whatever. This is the type of procedure that keeps planes in the sky and cars on the road.

  • by AJWM ( 19027 ) on Sunday November 04, 2007 @08:54PM (#21236473) Homepage
    I've sometimes wondered how far back in history you'd have to go before the technology was incapable of making a reliable relay and a battery. [...] Perhaps two hundred years ago, maybe more.

    Joseph Henry invented the relay in 1835, ten years after William Sturgeon invented the electromagnet (in turn five years after Oersted discovered electromagnetism in 1820). So the relay was invented a couple of years before Babbage started describing his analytical engine (1837 - the simpler difference engine he described in 1822). Had the knowledge of eg Boolean logic been there, a digital computer could certainly have been built before 1850. (In fact it took until 1937, when Claude Shannon proved in his master's thesis that Boolean algebra could be implemented with relays.)

    Assuming one already knew how to do it -- as with a time traveller -- all you'd need is a supply of wire (and some means to insulate it) and iron to make the relays. Chemical batteries are rather easy to make if you've got a couple of dissimilar metals, but if you can make relays you can probably also make generators. A modern day "Connecticut Yankee" could have given Arthur an electromechanical digital computer. Smelting of iron began in the BC era, and use of meteoritic iron goes way back. The ancient Egyptians certain knew how to make wire (for jewellery), so who knows how far back you could go. It's not so much a hard line as a level of increasing difficulty.
  • The difference (Score:2, Insightful)

    by dallaylaen ( 756739 ) on Monday November 05, 2007 @02:06AM (#21238405) Homepage
    "Linus was inspired by Minix and used it as the initial development platform for Linux" == Informative

    "Linus has copied Minix!!!1111one" == Troll

    Consider the difference between serving a cake, and throwing it into someone's face. Sorry for not using a car analogy.

    Mod GP back to troll, please.

What is research but a blind date with knowledge? -- Will Harvey

Working...