Slashdot Mirror

← Back to Stories (view on slashdot.org)

Nvidia Wins $20M In DARPA Money To Work On Hyper-Efficient Chips

Posted by timothy on from the just-tickle-the-electrons dept.

coondoggie writes "Nvidia said this week it got a contract worth up to $20 million from the Defense Advanced Research Projects Agency to develop chips for sensor systems that could boost power output from today's 1 GFLOPS/watt to 75 GFLOPS/watt."

12 of 72 comments (clear)

  1. Result will work great by kthreadd · · Score: 3, Funny

    But you will need their proprietary driver.

  2. Isn't this economic without DARPA funding? by iceco2 · · Score: 5, Insightful

    It seems to me a x75 increase in power efficiency should be worth to nVidia (or any competitor) much more than $20M, why does DARPA need to fund this, this seems exactly like the kind of work which doesn't need DARPA money. DARAPA should spend money where it is not clearly economic for others to do so.

    1. Re:Isn't this economic without DARPA funding? by Anonymous Coward · · Score: 2, Interesting

      Maybe DARPA also wants the thing to withstand radiation. Maybe they want it to have so little computational power that it would not sell in the market. Maybe they want every component to have been made in the US which the market won't care about. Maybe they have other restrictions. You don't know.

    2. Re:Isn't this economic without DARPA funding? by drinkypoo · · Score: 2

      DARAPA should spend money where it is not clearly economic for others to do so.

      Well, that's good advice, and I'm sure they'll take it over at DARAPA. On the other hand, DARPA has certain goals, whatever they might be, and if this is the most economical way to achieve them, then it's money well-spent from their perspective. If you're going to have a thing like DARPA, then you need to permit it to do things like this if you want it to be efficient. On the other hand, if you're going to do things like this, you need substantial oversight in place to prevent abuse. And on the gripping hand, military might is an extremely inefficient way to allocate R&D resources, and a better way to cause technology to move forward would be to put away childish things, and give the people the say in what happens; in a capitalist system like ours, we need to find them jobs so they can afford to buy products so they can vote with their pocketbooks. Of course, then we get into more checks and balances; sane advertising law is to capitalism as oversight is to government funding.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
  3. Units by ringman8567 · · Score: 2

    Why GFLOPS/watt? that is (operations/second)/(Joules/second). why not just operations/joule?

    1. Re:Units by SuricouRaven · · Score: 2

      Practicality. When talking about energy consumption, it's usually given in watts because the practical implications are time-dependant. You've got to account for the time it takes to run the calculations (which may be time-critical - you don't want your amalgamated radar data on a five-minute delay) and need to know the wattage to calculate cooling requirements. While operations/joule and flops/watt are equivilent, it's easier to think in terms of the former.

  4. I sure welcome this by Gaygirlie · · Score: 2

    As things usually do, the results of this research will eventually trickle down to desktops, laptops and mobile devices, and will result in either lesser power consumption or the same power consumption but in higher performance -- either way it's a plus. I just wish the contract could've been given to someone other than NVIDIA as it would be nice if the results of the research were released completely for free to the public instead of being patented up the wazoo, but alas, NVIDIA has so much experience in these things that it just makes sense to slap them with it if you expect results.

  5. NVIDIA is worth $7.87 billion by MSTCrow5429 · · Score: 3, Interesting

    So of course the Federal government needs to blow $20 million of taxpayer money, irregardless of its fiscal condition.

    --
    Slashdot: Playing Favorites Since 1997
  6. Re:did i misread something ? by Rockoon · · Score: 4, Interesting

    We passed 1e+07 operations per kWh in 1965.
    We passed 1e+08 operations per kWh in 1971.
    We passed 1e+09 operations per kWh in 1976.
    We passed 1e+10 operations per kWh in 1981.
    We passed 1e+11 operations per kWh in 1987.
    We passed 1e+12 operations per kWh in 1992.
    We passed 1e+13 operations per kWh in 1997.
    We passed 1e+14 operations per kWh in 2001.
    We passed 1e+15 operations per kWh in 2008.

    citation and graph

    Energy efficiency consistently doubles approximately every 1.6 years, so if we are at ~16 glops/watt right now, then we will blow past DARPA's target early in 2016... just a little over 3 years from now.

    --
    "His name was James Damore."
  7. Money matters, but ... by DavidClarkeHR · · Score: 4, Insightful

    So of course the Federal government needs to blow $20 million of taxpayer money, irregardless of its fiscal condition.

    I prefer it was spent on computing, rather than explosions.

    --
    - Nec Impar Pluribus, or so I'm told.
  8. Re:did i misread something ? by Entropius · · Score: 3, Insightful

    Out of curiosity (and I ask because I genuinely don't know), how many flops/watt do modern smartphones do? What about the GPU coprocessors in them?

    Modern GPU's are great, but they're not even optimized that strongly for power consumption.

  9. Article Summary Is Incorrect by DanielRavenNest · · Score: 3, Informative

    Current NVIDIA K20X compute card produces 5.575 Gflops double precision/Watt:

    http://www.anandtech.com/show/6446/nvidia-launches-tesla-k20-k20x-gk110-arrives-at-last

    Note that these cards are slightly different than consumer graphics cards. They have more double-precision pipelines because scientific computing cares more about that kind of math. They are also much more expensive than consumer cards. The underlying chip design is similar to the 600-series graphics cards. You can think of it as a modified version optimized for math, since the 600 series came out first, and is being produced in higher volume.