Article: Intel's Long Awaited Return to the Memory Business

Article: Intel’s Long Awaited Return to the Memory Business
By: aaron spink (aaronspink.delete@this.notearthlink.net), April 23, 2013 6:30 pm
Room: Moderated Discussions
Formula350 (burban502.delete@this.gmail.com) on April 23, 2013 2:58 pm wrote:
When you run a graphics card
> of higher performance with less RAM then that of a slightly lower performance card that is equipped with
> more RAM, the latter will actually pull ahead in instances where there is a call for texture storage space,
> such as when running high resolutions or with Anti-Aliasing being enabled. I believe it was the GTX480
> going up against the HD5870 where this occured, though it might have been GTX580 vs HD6970, but nevertheless...
> The nVidia card came equipped with 1.5GB, where as AMD was outfitting theirs with 2GB. Despite nV having
> a quicker product most of the time, when multi-screen (high resolution) gaming or increase levels of AA
> at even 1080p was tested, that "mere" 512MB was allowing the underdog to achieve higher frame rates. Or
> rather, more importantly, it was able to achieve playable frame rates!
>

You do understand that the case you are describing actually plays into the hand of this design correct? For the multi-screen space, the increasing constraint is the back buffer which for something like GT3e will almost assuredly be in the embedded dram which means that performance with screensize will be very good.

Space left over that is not needed for back buffer will likely be used as a high speed texture cache. The week point will likely be applications with large texture footprints that are heavily randomly accessed that can fit within the memory footprint of a discrete card but not the embedded dram. For applications with data set sizes that can neither fit in the embedded dram nor a discrete cards memory, a GT3e will likely pull ahead due to the higher performance to main memory.


< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
Article: Intel's Long Awaited Return to the Memory BusinessDavid Kanter2013/04/23 07:14 AM
  Article: Intel's Long Awaited Return to the Memory BusinessVarun2013/04/23 08:26 AM
    L4 Cache on Haswell-ETed J.2013/07/08 03:06 PM
      L4 Cache on Haswell-EDwight2013/07/09 03:18 PM
    Cache Chips Embedded in Package SubstrateAlbertH2013/07/10 03:50 PM
      Stacked Cache ChipsJason Lee2014/02/21 09:49 PM
        Stacked Cache ChipsDoug S2014/02/23 11:09 AM
          Stacked Cache ChipsMaynard Handley2014/02/23 05:55 PM
          Stacked Cache ChipsShiv2014/02/23 08:59 PM
  Article: Intel's Long Awaited Return to the Memory Businesssomeone2013/04/23 11:11 AM
    Article: Intel's Long Awaited Return to the Memory BusinessDavid Kanter2013/04/23 11:15 AM
      Article: Intel's Long Awaited Return to the Memory Businesssomeone2013/04/23 11:22 AM
      MIM capslacker2013/04/23 09:30 PM
        MIM capDavid Kanter2013/04/23 10:49 PM
          MIM captarlinian2013/04/24 07:40 AM
            This is my guess as well (NT)slacker2013/04/24 04:19 PM
        MIM capRob Thorpe2013/04/24 02:23 AM
  Article: Intel's Long Awaited Return to the Memory BusinessUngo2013/04/23 11:33 AM
    Article: Intel's Long Awaited Return to the Memory BusinessDavid Kanter2013/04/23 11:48 AM
      Article: Intel's Long Awaited Return to the Memory BusinessRohit2013/04/23 12:53 PM
        Article: Intel's Long Awaited Return to the Memory Businessdefault2013/04/24 08:29 AM
          Article: Intel's Long Awaited Return to the Memory BusinessRohit2013/04/25 07:21 AM
      Article: Intel's Long Awaited Return to the Memory BusinessUngo2013/04/23 09:08 PM
    Article: Intel's Long Awaited Return to the Memory BusinessAlpha2013/04/24 08:10 AM
      Article: Intel's Long Awaited Return to the Memory BusinessJames2013/04/24 08:39 AM
        DRAM power - sanity checkEduardoS2013/04/24 08:53 AM
          DRAM power - sanity checkMark Roulo2013/04/24 09:38 AM
            DRAM power - sanity checkEduardoS2013/04/24 10:32 AM
              DRAM power - sanity checkKevin G2013/04/25 06:36 AM
  Article: Intel's Long Awaited Return to the Memory BusinessFormula3502013/04/23 01:58 PM
    Article: Intel's Long Awaited Return to the Memory BusinessMark Roulo2013/04/23 02:20 PM
      Article: Intel's Long Awaited Return to the Memory BusinessAntti-Ville Tuunainen2013/04/23 04:25 PM
        Article: Intel's Long Awaited Return to the Memory BusinessJukka Larja2013/04/24 05:21 AM
        Article: Intel's Long Awaited Return to the Memory BusinessKevin G2013/04/25 06:48 AM
    Article: Intel's Long Awaited Return to the Memory BusinessDavid Kanter2013/04/23 03:04 PM
    Article: Intel's Long Awaited Return to the Memory Businessaaron spink2013/04/23 06:30 PM
      Article: Intel's Long Awaited Return to the Memory Businessanon2013/04/24 03:24 AM
        Article: Intel's Long Awaited Return to the Memory Businessaaron spink2013/04/24 10:01 AM
          Article: Intel's Long Awaited Return to the Memory Businessanon2013/04/24 03:23 PM
  Article: Intel's Long Awaited Return to the Memory Businessanon2013/04/23 05:56 PM
    Article: Intel's Long Awaited Return to the Memory Businessp.asta2013/04/24 09:13 AM
  Article: Intel's Long Awaited Return to the Memory BusinessVogel2013/04/24 06:07 AM
    Article: Intel's Long Awaited Return to the Memory Businessp.asta2013/04/24 08:57 AM
  Update: High frequency linkDavid Kanter2013/04/28 11:08 PM
    Update: High frequency linkGabriele Svelto2013/04/29 04:56 AM
      Update: High frequency linkAlbertH2013/07/10 04:21 PM
  Article: Intel's Long Awaited Return to the Memory BusinessMr. Camel2013/05/01 01:12 PM
  Haswell DRAM DIMM latencyAlbertH2013/07/10 04:37 PM
    Haswell DRAM DIMM latencyEduardoS2013/07/10 05:08 PM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell green?