Haswell vs Kaveri

Discussion in 'Architecture and Products' started by AnarchX, Feb 8, 2012.

  1. Andrew Lauritzen

    Moderator Veteran

    Joined:
    May 21, 2004
    Messages:
    2,526
    Likes Received:
    454
    Location:
    British Columbia, Canada
    Interesting idea - the screen is statically split into tiles across the two slices so you'll definitely get somewhat more scattered access in GT3 from that point of view (both ROP and memory reads from the shader). That said, the tiles are pretty small so I doubt it's going to thrash 4K pages or anything like that... how big are the "ranks" we're talking about in the DRAM?
     
  2. 3dilettante

    Legend Alpha

    Joined:
    Sep 15, 2003
    Messages:
    8,122
    Likes Received:
    2,873
    Location:
    Well within 3d
    A rank is the set of chips that are activated simultaneously by a chip select signal.
    The example 2-rank 4GB DIMM from the slide would have 2GB per rank.
     
  3. Andrew Lauritzen

    Moderator Veteran

    Joined:
    May 21, 2004
    Messages:
    2,526
    Likes Received:
    454
    Location:
    British Columbia, Canada
    So is that striped across the address space then? Any idea what the stripe size is? Just trying to figure out if really you'd get much more multi-rank stuff from a 2 slice vs. 1 slice part.
     
  4. 3dilettante

    Legend Alpha

    Joined:
    Sep 15, 2003
    Messages:
    8,122
    Likes Received:
    2,873
    Location:
    Well within 3d
    The exact mapping and policies for memory controllers are part of what I would use the words "special" and "sauce" for if I hadn't become sick of hearing that phrase over the last year.
    If there as least some locality to accesses, it would generally be better if addresses were not striped finely across multiple ranks, as hopping back and forth means the access pattern is now forcing a transition where it must unselect one set of chips and select the other with all the physical penalties, synchronization, and wait states needed.

    http://globalsp.ts.fujitsu.com/dmsp/Publications/public/wp-sandy-bridge-ep-memory-performance-ww-en.pdf

    According to this, at least for these workloads it is better for ranks to interleave with a granularity closer to an OS page.

    Channel interleaving can be done more finely, at granularities consistent with cache lines, since having multiple channels means chips aren't fighting over the same wires.

    I did see a presentation that gave a 256 byte granularity for GPU memory striping, but that might have been for an old architecture.
     
  5. kalelovil

    Regular

    Joined:
    Sep 8, 2011
    Messages:
    555
    Likes Received:
    93
  6. Wynix

    Veteran Regular

    Joined:
    Feb 23, 2013
    Messages:
    1,052
    Likes Received:
    57
    If demand for them is high they should just start using the 7700's as 7600's.
    Alternatively they could released a locked 7700 at ~$130.
     
  7. Blazkowicz

    Legend Veteran

    Joined:
    Dec 24, 2004
    Messages:
    5,607
    Likes Received:
    256
    Production is expensive or low yield?
    They have a back catalog of Trinity and Richland to sell as well. Releasing the 7600 makes them less desireable.

    It's damn annoying. At this point I'm wondering if they'll exit the highish performance business after Carrizo.
    VIA Nano X2 + DX10.1 chipset was like that : good, but you can't buy it. There was even that unobtainable Nano X4 that looked like a small Q6600, the swan song of the adventure then poof.. VIA/Centaur x86 CPU have ended.
     
  8. DSC

    DSC
    Banned

    Joined:
    Jul 12, 2003
    Messages:
    689
    Likes Received:
    3
    http://blogs.intel.com/technology/2...adobe-creative-cloud-intel-iris-pro-graphics/

    Nice turnaround for Intel compared to 4 years ago.

    http://www.anandtech.com/show/3972/nvidia-gtc-2010-wrapup/5

     
  9. Paran

    Regular Newcomer

    Joined:
    Sep 15, 2011
    Messages:
    251
    Likes Received:
    14

    There is a ~70% gap between GT3 and GT3e. GPU frequency difference roughly 15%. Bigger cache and faster CPU helps Iris Pro a bit of course but the gap is way too big. We know from Intel that the cache improves performance almost by 40% in average. This speedup is missing on the GT3 non edram version. With Skylake I hope 25-28W Iris will get some kind of edram too, otherwise it won't look appealing for the price.
     
  10. DSC

    DSC
    Banned

    Joined:
    Jul 12, 2003
    Messages:
    689
    Likes Received:
    3
  11. UniversalTruth

    Veteran

    Joined:
    Sep 5, 2010
    Messages:
    1,747
    Likes Received:
    22
    A8-7600 won’t be available until the second half of 2014.

    AMD is mum on the official reason for the delay. It could vary from high-demand for pre-built Kaveri-powered systems from emerging markets, or it could be trouble at Global Foundries. Regardless, more information will be available in the coming weeks at either a spring Tech Day or in the leadup to Computex.


    http://vr-zone.com/articles/amd-quietly-delays-mid-range-kaveri-chip/75820.html
     
  12. DSC

    DSC
    Banned

    Joined:
    Jul 12, 2003
    Messages:
    689
    Likes Received:
    3
    http://www.phoronix.com/scan.php?page=news_item&px=MTY2NTQ

     
  13. Paran

    Regular Newcomer

    Joined:
    Sep 15, 2011
    Messages:
    251
    Likes Received:
    14
    Assuming that GT3 has everything doubled in Broadwell it's no surprise. In Haswell GT3 the non-slice parts weren't doubled including the Media Engine. With Broadwell by the looks of it it's a true doubling this time. This should give Broadwell GT3 a huge transcoding boost in general over GT2.
     
  14. kalelovil

    Regular

    Joined:
    Sep 8, 2011
    Messages:
    555
    Likes Received:
    93
    Interesting that there is no mention of Carrizo in AMD's latest roadmap. I wonder if it been cancelled, or perhaps this particular roadmap is only designed to highlight their 'ambidextrous' strategy.

    [​IMG]
     
  15. DSC

    DSC
    Banned

    Joined:
    Jul 12, 2003
    Messages:
    689
    Likes Received:
    3
  16. DSC

    DSC
    Banned

    Joined:
    Jul 12, 2003
    Messages:
    689
    Likes Received:
    3
  17. Wynix

    Veteran Regular

    Joined:
    Feb 23, 2013
    Messages:
    1,052
    Likes Received:
    57
    #1378 Wynix, May 20, 2014
    Last edited by a moderator: May 20, 2014
  18. kalelovil

    Regular

    Joined:
    Sep 8, 2011
    Messages:
    555
    Likes Received:
    93
  19. Alexko

    Veteran Subscriber

    Joined:
    Aug 31, 2009
    Messages:
    4,490
    Likes Received:
    908
    Nice catch! :lol:

    It's pretty funny when naming schemes make so little sense that even internally, people get things completely wrong.
     
Loading...

Share This Page

  • About Us

    Beyond3D has been around for over a decade and prides itself on being the best place on the web for in-depth, technically-driven discussion and analysis of 3D graphics hardware. If you love pixels and transistors, you've come to the right place!

    Beyond3D is proudly published by GPU Tools Ltd.
Loading...