AMD: Navi Speculation, Rumours and Discussion [2019]

Discussion in 'Architecture and Products' started by Kaotik, Jan 2, 2019.

  1. pTmdfx

    Regular Newcomer

    Joined:
    May 27, 2014
    Messages:
    341
    Likes Received:
    283
    Those HBM GPUs have power-of-two numbers in slices & channels though. So sharing a slice is relatively easy.

    16 slices to 20 channels would require, say, groups of 4 L2 slices bound to 5 channels (20 / GCF), each of which uses a local 4/5 crossbar for memory blocks to be stripped at the same granluarity (256B?)

    That's how I read it too. The whitepaper also said:

     
    #2621 pTmdfx, Aug 20, 2020
    Last edited: Aug 20, 2020
    dobwal likes this.
  2. 3dilettante

    Legend Alpha

    Joined:
    Sep 15, 2003
    Messages:
    8,472
    Likes Received:
    4,410
    Location:
    Well within 3d
    The Xbox One X is an example of having a power of two number of cache slices mapped to a non-power of two number of channels.
    8 L2 slices were connected to 12 channels. There were four main controller clusters covering a power of two number of channels, and the remaining 4 channels were split into pairs that were each shared by two clusters.
    https://en.wikichip.org/wiki/microsoft/scorpio_engine
    That decision could easily be a fluke, but a similar scheme could happen with the Series X. Four controller clusters each hosting 4 channels, then the remaining four could be split into two pairs that each straddle two main clusters.

    The data fabric has crossbar nodes, although the broad GPU bus makes the overall interconnect a mesh. This might simplify the sharing arrangement, since the network could naturally route packets along the mesh. The 2x wider fabric from Renoir may also leave more slack since it can roughly 2x the bytes per clock than a single channel can deliver.

    If it were Vega there might be further reason to do this due to possible L2 alignment issues with the power of two count of RBEs. Flushes could still occur if for some reason the L2 slices and ROP data somehow did not align naturally. Perhaps Navi's L1 removes this concern, although I did see speculation that the 4-banked L1 would have a bank per set of L2 slices, and 20 slices would indicate another bank or an expansion of the L1-L2 interconnect.
     
    Silent_Buddha and Lightman like this.
  3. Krteq

    Newcomer

    Joined:
    May 5, 2020
    Messages:
    36
    Likes Received:
    63
    All aboard a hypetrain



     
  4. Bondrewd

    Veteran Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    1,042
    Likes Received:
    441
    wheeeeeeee, krumpin' time
     
  5. SimBy

    Regular Newcomer

    Joined:
    Jun 21, 2008
    Messages:
    641
    Likes Received:
    313
    Were there any leaks about AMD using HBM on consumer cards this gen?
     
  6. Bondrewd

    Veteran Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    1,042
    Likes Received:
    441
    Yeah.
     
  7. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    10,881
    Likes Received:
    1,102
    Location:
    New York
    Oh man, the combination of solid arch, high unit counts and market timing should result in a good old rumble at the high end. I can't remember the last time AMD checked all of those boxes at the same time.
     
  8. SimBy

    Regular Newcomer

    Joined:
    Jun 21, 2008
    Messages:
    641
    Likes Received:
    313
    Wasn't there someone from AMD (could be Raja) regretting the whole HBM on consumer cards idea?
     
  9. Bondrewd

    Veteran Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    1,042
    Likes Received:
    441
    Yeah.
    Arguably Hawaii.
    This time they have the full stack (from a meagre funne APU to N21) with quick rollouts so smells like Evergreens to me.
    Was him, very recently.
    You just need performance chops to back it up.
     
  10. Rootax

    Veteran Newcomer

    Joined:
    Jan 2, 2006
    Messages:
    1,717
    Likes Received:
    1,080
    Location:
    France
    I believe Raja said, in an Intel video, that he learned that hbm was costly, something like that. But I didn't remember the word "regret".
     
  11. Krteq

    Newcomer

    Joined:
    May 5, 2020
    Messages:
    36
    Likes Received:
    63
    There are some Sienna related commits to AMDGPU kernel driver clearly mentioning HBM interface for that GPU.

    Question is... will be customer cards based on Sienna Cichlid GPU or not?
     
  12. Bondrewd

    Veteran Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    1,042
    Likes Received:
    441
    It's literally N21, they're just using funny codenames to obfuscate their ninja ways.
     
  13. yuri

    Newcomer

    Joined:
    Jun 2, 2010
    Messages:
    249
    Likes Received:
    231
    So "GA104 can't beat it". Big Navi is faster than RTX 3070, mkay.
     
  14. Bondrewd

    Veteran Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    1,042
    Likes Received:
    441
    N22, yes.
    Mobile will smell foul once we exit January'21.
    Faster than many a thing.
     
  15. madhatter

    Newcomer

    Joined:
    Jul 23, 2020
    Messages:
    21
    Likes Received:
    17
    When's N22 coming for desktop?
     
  16. Bondrewd

    Veteran Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    1,042
    Likes Received:
    441
    1-4 weeks after N21.
     
  17. Krteq

    Newcomer

    Joined:
    May 5, 2020
    Messages:
    36
    Likes Received:
    63
    Well, that seems to be a reason why NV jumped to GA102 for x80 card this time and also pumped TDP so high. So I'm really excited for upcoming days :)
     
  18. madhatter

    Newcomer

    Joined:
    Jul 23, 2020
    Messages:
    21
    Likes Received:
    17
    Nice, I was worried it might be longer than that.
     
  19. Bondrewd

    Veteran Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    1,042
    Likes Received:
    441
    Oh yeah, makes me wish AMD was less lazy with their marketing.
    Nah, they gotta crap all things RDNA2 out and move on.
     
  20. ToTTenTranz

    Legend Veteran Subscriber

    Joined:
    Jul 7, 2008
    Messages:
    11,499
    Likes Received:
    6,260
    Who is this twitter user and what's their track record?
     
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...