Odds & Ends: ECC & NVIDIA Surround Missing

One of the things we have been discussing with NVIDIA for this launch is ECC. As we just went over in our GF100 Recap, Fermi offers ECC support for its register file, L1 cache, L2 cache, and RAM. The latter is the most interesting, as under normal circumstances implementing ECC requires a wider bus and additional memory chips. The GTX 400 series will not be using ECC, but we went ahead and asked NVIDIA how ECC will work on Fermi products anyhow.

To put things in perspective, for PC DIMMs an ECC DIMM will be 9 chips per channel (9 bits per byte) hooked up to a 72bit bus instead of 8 chips on a 64bit bus. However NVIDIA doesn’t have the ability or the desire to add even more RAM channels to their products, not to mention 8 doesn’t divide cleanly in to 10/12 memory channels. So how do they implement ECC?

The short answer is that when NVIDIA wants to enable ECC they can just allocate RAM for the storage of ECC data. When ECC is enabled the available RAM will be reduced by 1/8th (to account for the 9th ECC bit) and then ECC data will be distributed among the RAM using that reserved space. This allows NVIDIA to implement ECC without the need for additional memory channels, at the cost of some RAM and some performance.

On the technical side, despite this difference in implementation NVIDIA tells us that they’re still using standard Single Error Correction / Double Error Detection (SECDED) algorithms, so data reliability is the same as in a traditional implementation. Furthermore NVIDIA tells us that the performance hit isn’t a straight-up 12.5% reduction in effective memory bandwidth, rather they have ways to minimize the performance hit. This is their “secret sauce” as they call it, and it’s something that they don’t intend to discuss at in detail at this time.

Shifting gears to the consumer side, back in January NVIDIA was showing off their Eyefinity-like solutions 3DVision Surround and NVIDIA Surround on the CES showfloor. At the time we were told that the feature would launch with what is now the GTX 400 series, but as with everything else related to Fermi, it’s late.

Neither 3DVision Surround nor NVIDIA surround are available in the drivers sampled to us for this review. NVIDIA tells us that these features will be available in their release 256 drivers due in April. There hasn’t been any guidance on when in April these drivers will be released, so at this point it’s anyone’s guess whether they’ll arrive in time for the GTX 400 series retail launch.

The GF100 Recap Tessellation & PhysX
Comments Locked

196 Comments

View All Comments

  • yacoub - Friday, March 26, 2010 - link

    I think he's got it right. If you want high yields, a larger chip size is the enemy, because you get fewer chips per die, and thus lower yields.
  • Rebel44 - Friday, March 26, 2010 - link

    IMO HardOCP review was better because they showed real world differences between those Nv and AMD cards - 470 didnt allow better setting than 5850 and 480 was only little bit better than 5870. So 470 is IMO epic fail at that price.

    When you add extra power and noise fom 470 and 480, I wouldnt pay for them more than for 5850 and 5870.
  • stagen - Friday, March 26, 2010 - link

    With the 470 and 480 generating so much heat and noise, and consume more power than even the dual GPU Radeon HD 5970, even thinking of dual GPU 470/480 (495?) is a scary thing to do.
  • yacoub - Friday, March 26, 2010 - link

    Agreed. And considering the $350 470 is no faster than a $150 5770 at 1680x in BF:BC2, and only 23% faster at 1920x, that's pathetic. Considering how much better it does in other games, it must be a driver optimization issue that hopefully can be worked out.
  • mcnabney - Saturday, March 27, 2010 - link

    Fermi has existed for months, so the driver work should be as far along as AMD. The delay allowed for better stepping and higher clocks, but the drivers aren't going to improve any more quickly than AMD.
  • uibo - Friday, March 26, 2010 - link

    What was the ambient temperature?
  • Ryan Smith - Saturday, March 27, 2010 - link

    20C.
  • mindbomb - Friday, March 26, 2010 - link

    Firmly in AMD's hands?
    i dont know about that.
    Although it can't bitstream true hd and dts-MA, I would argue that's not really as debilitating as not being able to bitstream level 5.0 h264 video, since you can output as LCPM.
  • Galid - Friday, March 26, 2010 - link

    FIRMLY in AMD's hand...it is... not only Nvidia doesn't do true HD and dts-ma for a card that doesn't fit in a HTPC but they won't even when they get the smaller cards out... Firmly? yeah....
  • mindbomb - Friday, March 26, 2010 - link

    let me be more clear.
    I'm saying although the ati cards have better handling of audio, the nvidia cards do in fact have better handling of video since they can handle level 5.0 and 5.1 h264 video (and i guess mpeg 4 asp, but thats irrelevant)
    so i wouldn't say the ati cards have a definite lead in this area.

Log in

Don't have an account? Sign up now