AMD: Pirate Islands (R* 3** series) Speculation/Rumor Thread

Discussion in 'Architecture and Products' started by iMacmatician, Apr 10, 2014.

Tags:
  1. Infinisearch

    Veteran Regular

    Joined:
    Jul 22, 2004
    Messages:
    739
    Likes Received:
    139
    Location:
    USA
    I figured it was backed in VRAM but my guess was there was an on-chip cache dedicated to it. It just makes sense from a performance perspective... adding a L2 cache access for every few framebuffer access's doesn't sound very nice.
     
  2. mczak

    Veteran

    Joined:
    Oct 24, 2002
    Messages:
    3,015
    Likes Received:
    112
    It is certainly possible (and indeed pretty much a must) there's some cache dedicated to it in the ROP, but I was just pointing out it wouldn't make sense to store all of it on chip at once (as did older chips), as the max cache size you'd need is simply too big. But we don't know much about how ROP cache works in general (tile data + metadata), and I have no idea how this works for newer chips (which need to access this from TMUs, so probably indeed through L2). But certainly it could be stuffed into ordinary cache lines (one cache line, no matter where, can store information about quite a few tiles of course (with those mentioned 2 bits per tile, for a rgba8 format that's only 1/1024 of the data you'd need for the uncompressed pixels for this metadata, so unless you have a really terrible case not achieving any compression it's not a big deal).
     
  3. Infinisearch

    Veteran Regular

    Joined:
    Jul 22, 2004
    Messages:
    739
    Likes Received:
    139
    Location:
    USA
    You're right I forgot about morton ordering... so that would be twice the number of tiles per byte size cache line. But I was told in another thread https://forum.beyond3d.com/threads/gpu-cache-sizes-and-architectures.56731/ that at least on GCN that the CB and DB caches don't go through the L2... did it change for tonga/GCN+framebuffercompression?
    Why would TMU's need access to this? If you mean render target to texture doesn't the driver resolve the compressed RT first manually? At least thats what I assumed.
     
  4. Miksu

    Regular

    Joined:
    Mar 9, 2003
    Messages:
    997
    Likes Received:
    10
    Location:
    Finland
    The Fury X should go on sale tomorrow so I think we should see reviews soon. Any word when exactly reviews are out?
     
    Grall likes this.
  5. Lightman

    Veteran Subscriber

    Joined:
    Jun 9, 2008
    Messages:
    1,804
    Likes Received:
    475
    Location:
    Torquay, UK
    Tomorrow as well. NDA lift and sales will coincide.
     
  6. Grall

    Grall Invisible Member
    Legend

    Joined:
    Apr 14, 2002
    Messages:
    10,801
    Likes Received:
    2,172
    Location:
    La-la land
    Further thinking about buffer compression... How do you store compressed pixels efficiently in memory, are they packed together into bundles to fit DRAM burst length, or how does it work? I can only assume it would be very inefficient to store a few bytes at most where a full 8 or 16 byte deep color pixel used to reside, and then repeat for all other pixels.

    I'm just thinking that if you re-pack the pixels, you would need some kind of mechanism to quickly isolate an individual pixel inside the bundle when you need to read it back again... *shrug* Instances like this, I really wish I'd had the brains and willpower to have studied graphics engineering stuff in university... :p
     
    Lightman likes this.
  7. DmitryKo

    Regular

    Joined:
    Feb 26, 2002
    Messages:
    696
    Likes Received:
    581
    Location:
    55°38′33″ N, 37°28′37″ E
    Yes, fixed-length encoding - probably pixel differences in a 8x8 block.

    http://graphics.stanford.edu/~mhous...all/HoKo_compression_in_graphics_pipeline.pdf
    http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.60.8187&rep=rep1&type=pdf
    http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.81.412&rep=rep1&type=pdf

    Inefficient for what exactly - perceived quality, compression ratio, or decoding complexity?
     
    #2327 DmitryKo, Jun 23, 2015
    Last edited: Jun 23, 2015
  8. gongo

    Regular

    Joined:
    Jan 26, 2008
    Messages:
    582
    Likes Received:
    12
    [​IMG]

    450GB/s, non-overclockable HBM...lolx
     
  9. dskneo

    Regular

    Joined:
    Jul 25, 2005
    Messages:
    517
    Likes Received:
    20
    rumors were wrong?
     
  10. Lightman

    Veteran Subscriber

    Joined:
    Jun 9, 2008
    Messages:
    1,804
    Likes Received:
    475
    Location:
    Torquay, UK
    Are you laughing because there is an error on the box or because you think Fury X will have 450GB/s memory bandwidth?

    Other vendor for instance has put Fury X for pre-orders, but don't buy it as it has GDDR5! (sarcasm)
     
  11. gongo

    Regular

    Joined:
    Jan 26, 2008
    Messages:
    582
    Likes Received:
    12
    Sapphire is AMD premium partner and ...i think they also assembles some cards for them...it would be a silly printing mistake..no?
     
  12. Lightman

    Veteran Subscriber

    Joined:
    Jun 9, 2008
    Messages:
    1,804
    Likes Received:
    475
    Location:
    Torquay, UK
    On the other hand AMD designs chip and puts out specification. All Fury X cards will be AMD made for now (manufactured by a partner, most likely Sapphire). AMD states in official data-sheet 500MHz HBM clock so please do the math and make your own conclusion. Besides HBM is overclokable as is any other memory, but AMD will not allow overclocking from Overdrive, at least initially. There is no saying 3rd party tools will not break the lock and TBH there were leaks from couple of months ago claiming HBM1 on Fiji can go as high as 700MHz.

    Anyway, this is non issue for now as there is no point having massive bandwidth if your engine can't utilize it. We will have to wait and see if Fiji can be memory bandwidth limited at all.
     
    Grall likes this.
  13. Xmas

    Xmas Porous
    Veteran Subscriber

    Joined:
    Feb 6, 2002
    Messages:
    3,299
    Likes Received:
    137
    Location:
    On the path to wisdom
    I don't know what AMD implemented, but a resolve pass required on colour render targets might often actually increase total memory bandwidth requirements.

    Note that bandwidth savings aren't everything and it is sometimes possible to achieve higher performance by increasing the total number of memory accesses, if that leads to a more even spread of memory accesses over time. I.e. if you could do the resolve (which is a streaming operation thus doesn't need to pollute any cache) while the memory subsystem is mostly idle. But that seems unlikely.
     
  14. lanek

    Veteran

    Joined:
    Mar 7, 2012
    Messages:
    2,469
    Likes Received:
    315
    Location:
    Switzerland

    I think the error is on the box, because the 512GB/s is coming from the AMD slides.
     
  15. pharma

    Veteran Regular

    Joined:
    Mar 29, 2004
    Messages:
    2,933
    Likes Received:
    1,629
    We should know shortly from other AIB packaging.
     
  16. lanek

    Veteran

    Joined:
    Mar 7, 2012
    Messages:
    2,469
    Likes Received:
    315
    Location:
    Switzerland
    Faster way, there's the meeting webinar with AMD t today about Fury and other graphics lineup. if some attempt to it can ask them, im registered, but not sure i could be online at this time.

    HH have his sample it seems. ready for a review tomorrow.
    http://www.guru3d.com/news_story/are_you_ready_for_some_fury_tomorrow.html
     
  17. Rurouni

    Veteran

    Joined:
    Sep 30, 2008
    Messages:
    950
    Likes Received:
    216
    Maybe the HBM is clocked at 440MHz? I think AMD promising up to 512GB/s and to achieve that they must use 500MHz. Since it's up to, I wouldn't be surprised if 450GB/s on early card is correct. It will still be the highest bandwidth GPU available.
     
  18. Grall

    Grall Invisible Member
    Legend

    Joined:
    Apr 14, 2002
    Messages:
    10,801
    Likes Received:
    2,172
    Location:
    La-la land
    Thanks for your reply! :)

    Pardon my imprecise language - I meant memory bandwidth-wise, as writing a few bytes spread out here and there is quite wasteful.
     
  19. gongo

    Regular

    Joined:
    Jan 26, 2008
    Messages:
    582
    Likes Received:
    12
    How many hours more before NDA Reviews are up...?
    Exciting times...Fury X (450GB/s) or 980Ti G1....i need an upgrade.
     
    pharma and Grall like this.
  20. Grall

    Grall Invisible Member
    Legend

    Joined:
    Apr 14, 2002
    Messages:
    10,801
    Likes Received:
    2,172
    Location:
    La-la land
    I know that upgrade demon intimately. It starts haunting me strongly every two years or so!
     
    Lightman likes this.
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...