AMD: RDNA 3 Speculation, Rumours and Discussion

Discussion in 'Architecture and Products' started by Jawed, Oct 28, 2020.

  1. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,819
    Likes Received:
    3,976
    Location:
    Finland
    Well that thing is coming and supposedly will compete with 3060 Ti (cut down model against 3060 then?), but it's unlikely related to this as we've seen several Navi2X-chip codenames and none are related to Nashira, summit or otherwise.
     
  2. Bondrewd

    Veteran Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    1,130
    Likes Received:
    510
  3. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,819
    Likes Received:
    3,976
    Location:
    Finland
    It doesn't fit any of the previous codenames, though?
     
  4. Bondrewd

    Veteran Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    1,130
    Likes Received:
    510
    Cuz it's the board, not the GPU itself.
    Remember that Vega12 ES? Big and bright-red.
    Stuff like that.
     
    ethernity and Lightman like this.
  5. ethernity

    Newcomer

    Joined:
    May 1, 2018
    Messages:
    88
    Likes Received:
    207
    AMD apparently filed two provisional patents in 2019 for ML based chiplets
    "CHIPLET APPROACH FOR COUPLING GPU WITH MACHINE LEARNING ACCELERATION AT HIGH POWER EFFICIENCY," filed on Jul. 22, 2019
    "HIGH BW INTER-CONNECTED CHIPLETS AND GPU FOR HIGH PERFORMANCE GAMING AND MACHINE LEARNING WORKLOADS," filed on Jul. 22, 2019

    Full patent application is this (within 12 months of Provisional)
    Filed: July 20, 2020

    20210026686 CHIPLET-INTEGRATED MACHINE LEARNING ACCELERATORS

    Techniques for performing machine learning operations are provided. The techniques include configuring a first portion of a first chiplet as a cache; performing caching operations via the first portion; configuring at least a first sub-portion of the first portion of the chiplet as directly-accessible memory; and performing machine learning operations with the first sub-portion by a machine learning accelerator within the first chiplet.

    upload_2021-1-28_12-15-46.png

    Seems like a ML accelerator for gaming implemented in the Infinity Cache chiplets. Seems gaming oriented.

    https://www.freepatentsonline.com/20210026686.pdf
     
    Newguy, Lightman and Krteq like this.
  6. ethernity

    Newcomer

    Joined:
    May 1, 2018
    Messages:
    88
    Likes Received:
    207
    In continuation to the patent post
    The Memory in the chiplet is divided into two parts, one acting as LLC and another part use to sync between the CUs and the Accelerator

    upload_2021-1-28_12-50-7.png
     
    Newguy, Lightman, Krteq and 1 other person like this.
  7. CarstenS

    Legend Veteran Subscriber

    Joined:
    May 31, 2002
    Messages:
    5,365
    Likes Received:
    3,107
    Location:
    Germany
    Now, where have we seen something like this?
     
  8. Rootax

    Veteran Newcomer

    Joined:
    Jan 2, 2006
    Messages:
    2,095
    Likes Received:
    1,536
    Location:
    France
    This ML accelerator chiplet will never be found....
     
  9. dobwal

    Legend Veteran

    Joined:
    Oct 26, 2005
    Messages:
    5,715
    Likes Received:
    1,967
    Xbox 360?
     
  10. DegustatoR

    Veteran

    Joined:
    Mar 12, 2002
    Messages:
    2,244
    Likes Received:
    1,681
    Location:
    msk.ru/spb.ru
    Xe HPC more like
     
    3D0SH likes this.
  11. Megadrive1988

    Veteran

    Joined:
    May 30, 2002
    Messages:
    4,717
    Likes Received:
    238
    According to Paul at Redgamingtech, Navi 31 and 32 will use chiplets, Navi 33 will be a monolithic die. All coming in 2022.

    Navi 41 is early, no real silicon yet, but progressing well.

     
    Lightman likes this.
  12. Frenetic Pony

    Regular Newcomer

    Joined:
    Nov 12, 2011
    Messages:
    697
    Likes Received:
    382
    Like, I know the guy was right before. But isn't this the same leak from like a month ago? The one that doesn't make a lot of sense because why would anyone make a chiplet that big with that low of a yield if you could just cut it in half and see yields skyrocket, design costs plummet, and get whatever flexibility you want with binning. Besides which, they'd need to design multiple chiplets for this (rather than one and just reuse like with Zen), cut another 25% of power just to hit 360 watts for this "160cu" top end chip, and run a 512bit bus with 18gbps GDDR6 or HBM just to supply the thing.

    I'm just going to go ahead and doubt this one a bit, at least until concrete information on how this is supposed to be supported at all emerges.
     
  13. Leoneazzurro5

    Newcomer

    Joined:
    Aug 18, 2020
    Messages:
    226
    Likes Received:
    249
    Why they'd need to make multiple chiplets? They could have an "80 CU" chiplet and one I/O scalable die, both with adequate cache amount on die, so they could go for 80-160 CU with two chiplets and one I/O, and 40-80 CUs with one chiplet and I/o with less RAM bus size, and monolithic for everything below.
     
  14. Rootax

    Veteran Newcomer

    Joined:
    Jan 2, 2006
    Messages:
    2,095
    Likes Received:
    1,536
    Location:
    France

    I was thinking the same thing about the chiplet size. But then I thought, it's their first gpu chiplet design, maybe they will keep it simple ? Anyway, we'll see :)
     
  15. Jawed

    Legend

    Joined:
    Oct 2, 2004
    Messages:
    11,286
    Likes Received:
    1,551
    Location:
    London
    GDDR bus size needs to scale with the count of graphics chiplets. An I/O chiplet providing GDDR doesn't do that.
     
  16. sir doris

    Regular

    Joined:
    May 9, 2002
    Messages:
    700
    Likes Received:
    157
    Could an I/O chiplet be designed for the max bandwidth, then disable a proportion of the Infinity Fabric and GDDR channels for lesser designs?
     
  17. no-X

    Veteran

    Joined:
    May 28, 2005
    Messages:
    2,377
    Likes Received:
    364
    7nm Navi 21 is ~520 mm². Let's say the IO die (PCIe/DP/HDMI/UVD/VCE/ETC) will reduce the die by ~60 mm² (just a quick guess). That's 460 mm². At 5 nm the resulting chiplet size could be around 255 mm². Maybe a bit bigger because of the interface for chiplet interconnection. Is it really that big? At the time of RV770 AMD called it "sweet spot".
     
  18. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,819
    Likes Received:
    3,976
    Location:
    Finland
    Yes, but no.
    edit:
    To clarify, yes, it would be possible, but it would be beyond strange and stupid.
    I/O wants to sit on the edge of the chip(let), scalable I/O die for GPU would need to be quite big just to accommodate wide enough bus for the high end GPUs, which would make it unpractical in anything lower end.
    By having memory controller(s) in each compute chiplet the bus width would scale with GPU performance in sensible way.
     
    #138 Kaotik, Feb 9, 2021
    Last edited: Feb 9, 2021
    BRiT, SpeedyGonzales and sir doris like this.
  19. Bondrewd

    Veteran Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    1,130
    Likes Received:
    510
    Ugh broski it's 40 gfx11 WGPs on N5 without the fancy uncore even being there.
    That's small.
    That was before costs exploded but yeah.
    Pointless, instead they're throwing N6 tapeouts at the problem.
    See: Genoa.
     
    Lightman likes this.
  20. Frenetic Pony

    Regular Newcomer

    Joined:
    Nov 12, 2011
    Messages:
    697
    Likes Received:
    382
    It would be around 160-240mm or so as a good guess. That's 2-3x the size of a zen 3 chiplet on a node that ideally shrinks things by almost half. But still, yields don't actually go up that much if you cut it in half. So you have a point.

    What's more, just found TSMC's tiny sram bragging for their 5nm. 256mb cache is tiny tiny, 5mm. Suddenly I can see why AMD went with SRAM cache on RDNA2. It doesn't make a lot of sense at the moment, but as a future investment for 5nm it seems sensible.

    Ok, bandwidth and power are still problems. They'd need to 18gbps+ on a 512bit bus, or HBM. They'll also need to increase architecture power efficiency again to get that huge chip in a reasonable tdp, as realistically the 5nm shrink won't be enough alone. Still, it all seems more reasonable now, and a potential monster of a chip, which would explain why Nvidia is urging on their "Ada" arch ASAP.

    For future RDNA though I can see AMD going partially, or wholly, Samsung for their GPUs. They make more money on CPUs, so it'd make more sense for them to stick with the best foundry there even though supply from them is limited. But Samsung's gaafet is a transition that's coming one way or another. And it should put up a competitive fight against TSMC's 5nm and even their first 3nm, which looks to be such a disappointment for their customers that they immediately rushed out an announcement that they'd be transitioning to gate all around quickly after that as well.
     
    #140 Frenetic Pony, Feb 9, 2021
    Last edited: Feb 10, 2021
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...