26
Bootstrapping Your Own CPU (programming.dev)
submitted 1 year ago by to c/technology@beehaw.org
17 comments fedilink hide all child comments
 

Dear community,

Let's just say I'm a country that wants to create my own CPU only using knowledge/tech/techniques that are in the open and nothing proprietary. When I said CPU, let's just say something that can run a C program, and eventually the linux kernel.

Is creating one out of publicly accessed knowledge and resources even possible, and how minuscule the tolerance need to be? Is there even a successful open CPU project out there?

I'm asking this because of an anxiety that I have when knowing only several companies in the world know how to create a CPU.

you are viewing a single comment's thread
view the rest of the comments
[–] 7 points 1 year ago* (6 children)

It may be possible to design but near impossible to fab without having access to all the technologies in the fab pipeline. These are things you aren't likely to be able to quickly recreate. Hence why Russia and China, despite working on CPU's for a long time are still behind Taiwan.

  • permalink
  • fedilink
  • source
  • hideshow 6 child comments
    •
  • [–] [S] 4 points 1 year ago (5 children)

    I'm not trying to create the new Core i7 here, just something that can run as well as a raspberry pi. My imagination that a nation/company need to start small and aim small, before exploring other more complex architecture.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 5 child comments
    •
  • [–] 7 points 1 year ago* (2 children)

    Well that's about the level that Russia is at after working at it (okay mostly stealing others' technology) for 50 years.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] [S] 2 points 1 year ago (1 child)

    holy shit

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 2 points 1 year ago* (last edited 1 year ago)

    As a matter of fact when looking at their own in country manufactured lithography, they are hoping to make a fab that can hit 350nm by 2030. That's the equivalent of a pentium II from 30 years ago. The raspberry pi 5 is a 16nm chip.

    They had developed much faster chips but those relied on machines that are now under sanctions so they are trying to build their own sad fab.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 3 points 1 year ago (1 child)

    If you are truly starting from scratch, shooting for Raspberry Pi performance isn't starting small, thats a huge goal. It's a complex chip built on a fairly modern process node (28 nm for the 4B) using the second-best-established architecture.

    The reasonable goal to shoot for would be an 8086-like chip, then perhaps something like a K3-II or early Pentium, then slowly work your way up from there.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •