Adapteva looks to Kickstarter to fund $99 supercomputing kit

By Shawn Knight ยท 4 replies
Sep 28, 2012
Post New Reply
  1. A new Kickstarter project aims to make parallel computing accessible to everyone by providing an affordable, open and easy to use platform. It's called Parallella and for just $99, backers will receive a full developer kit that includes an Epiphany-III...

    Read more
  2. EEatGDL

    EEatGDL TS Evangelist Posts: 568   +241

    I would support this campaign if I had a way and my own income.
  3. wujj123456

    wujj123456 TS Enthusiast Posts: 33   +8

    They should just put out a tier that will give out a 64-core version. It could be more expensive than $199. But the 64-core version is definitely much more interesting than a 16-core one to me.
  4. I have a feeling that I have grossly misunderstood the general idea/mechanics behind this proposal.

    open source software- ok
    less than 5 watts - fine
    open source hardware - what does this even mean, open rights, full access to specs sheets and everyone can make changes? Who is going to make and assemble physical parts then and make up standards to use for aforementioned open source programming?
    up to 45GHz of processing power - impressive, though I guess doable with multiple cpus/cores, wait did they say less than 5 watts?
    supercomputer - what? we must have different definition of supercomputer
    $99 - again, what? never mind using words supercomputer and $99 together, they even seem to imply that you can get the above specs for this much

    And this was all somehow used in one sentence?

    Did adapteva just invent some new technological breakthrough that is on the same scale as transitioning from lamp based computers to transistors, or is this another fine example of weather based computers again?
    TJGeezer likes this.
  5. bugejakurt

    bugejakurt TS Booster Posts: 158   +15

    What I think is meant with this hardware is that you would buy it for $99 dollars and with it you will have an API with which complex computation can be off-loaded manually from a single CPU to multiple CPUs. You would have the power to distribute a processing instruction, distribute fragments of processing and if for example you have a 100 iterative loop you can off load the 100 loops to a set of 25s on 4 processors. All this can be achieved manually through C++/C programming. We don't have information as to the amount of processors which are available on the platform, etc.
    TJGeezer likes this.

Similar Topics

Add your comment to this article

You need to be a member to leave a comment. Join thousands of tech enthusiasts and participate.
TechSpot Account You may also...