esram astrophysics *spin-off*

Discussion in 'Console Technology' started by astrograd, Aug 3, 2013.

Thread Status:
Not open for further replies.
  1. McHuj

    Veteran Regular Subscriber

    Joined:
    Jul 1, 2005
    Messages:
    1,433
    Likes Received:
    553
    Location:
    Texas
    The minimum given at hot chips was 109GB/sec which you can only derive using the new clock.

    1024bits x 853MHz ~ 109. The 204 still doesn't add up
     
    #221 McHuj, Sep 10, 2013
    Last edited by a moderator: Sep 10, 2013
  2. Hardknock

    Veteran

    Joined:
    Jul 11, 2005
    Messages:
    2,203
    Likes Received:
    53
  3. Brad Grenz

    Brad Grenz Philosopher & Poet
    Veteran

    Joined:
    Mar 3, 2005
    Messages:
    2,531
    Likes Received:
    2
    Location:
    Oregon
    He's also made several factually incorrect and self contradictory claims in that thread. He offers no explanation for how the simultaneous read/write is attained, or why it only apparently only occurs 7/8ths of the time.
     
  4. 3dilettante

    Legend Alpha

    Joined:
    Sep 15, 2003
    Messages:
    8,122
    Likes Received:
    2,873
    Location:
    Well within 3d
    He said the eSRAM can do read and write cycles simultaneously.
    The discussion so far has been the multiple ways that can be done, in keeping with the other numerical constraints that have also been disclosed.
     
  5. blakjedi

    Veteran

    Joined:
    Nov 20, 2004
    Messages:
    2,975
    Likes Received:
    79
    Location:
    20001
    You are conflating what was said here (7/8ths) with what he said (simultaneous read/write). He also doesnt have to provide ANY explanation as to how his engineers achieved that. The fact that you think he should is ludicrous.
     
  6. Lalaland

    Regular

    Joined:
    Feb 24, 2013
    Messages:
    596
    Likes Received:
    265
    I thought the 7/8ths comes from the fact that quoted bandwidth for the bidirectional ESRAM is not 2x unidirectional speed which is weird when the scaling is normally linear
     
  7. blakjedi

    Veteran

    Joined:
    Nov 20, 2004
    Messages:
    2,975
    Likes Received:
    79
    Location:
    20001
    depends on which numbers you use.

    if you take the original 102 GB/s and make it two ops per cycle then you get his number.

    the 133gb and 192gb/s are the question mark numbers.

    I understand he said those numbers and hes the PR but be stated categorically that the engineers will eventually answer all technical questions. If thats what GAF needs to be convinced then so be it.
     
  8. 3dilettante

    Legend Alpha

    Joined:
    Sep 15, 2003
    Messages:
    8,122
    Likes Received:
    2,873
    Location:
    Well within 3d
    If you ignore the clock bump that updated both the minimum and peak.
     
  9. blakjedi

    Veteran

    Joined:
    Nov 20, 2004
    Messages:
    2,975
    Likes Received:
    79
    Location:
    20001
    true
     
  10. Hardknock

    Veteran

    Joined:
    Jul 11, 2005
    Messages:
    2,203
    Likes Received:
    53
    Sorry for being obtuse but would a clock boost automatically warrant a boost in esram bandwidth?
     
  11. Shifty Geezer

    Shifty Geezer uber-Troll!
    Moderator Legend

    Joined:
    Dec 7, 2004
    Messages:
    40,729
    Likes Received:
    11,200
    Location:
    Under my bridge
    No, but when you have a BW increase that's proportional to the clock increase, it points a very strong finger to the eSRAM bus being on the same clock.
     
  12. Brad Grenz

    Brad Grenz Philosopher & Poet
    Veteran

    Joined:
    Mar 3, 2005
    Messages:
    2,531
    Likes Received:
    2
    Location:
    Oregon
    It's ludicrous to want an explanation for why a piece of hardware can achieve nearly twice the performance it should be mathematically capable of based on their own description of the design? And why it's not actually twice the performance and instead an additional 7/8ths? He claimed simultaneous read/write and quoted the 7/8ths number at the same time. If he wants anyone to believe either then he should be able to provide some details on how this is possible or it will just make him look like a carpet bagger.
     
  13. warb

    Veteran

    Joined:
    Sep 18, 2006
    Messages:
    1,057
    Likes Received:
    1
    Location:
    UK
    You're doubting the possibility? I don't see why'd they bother lying to X1 devs first. People who have the hardware.
     
  14. bkilian

    Veteran

    Joined:
    Apr 22, 2006
    Messages:
    1,539
    Likes Received:
    3
    You don't have to believe him. The only folks who have to believe him are the ones actually writing code on the box, and they have all the tools they need to verify his statement.
     
  15. ROG27

    Regular

    Joined:
    Oct 27, 2005
    Messages:
    572
    Likes Received:
    4
    Looking at the leaked documentation on VGLeaks (http://www.vgleaks.com/durango-memory-system-overview/), I'm not sure how MS now arrives at 109 GB/s (102.4 GB/s prior to uptick) being the minimum bandwidth available for the ESRAM and 204 GB/s (192 GB/s prior to uptick) being the maximum, when diagrams included in their prior documentation seem to indicate that the bidirectional read and write bus had a 51.2 GB/s max throughput for read and 51.2 GB/s max throughput for write? How can both read and write's throughput seemingly double to now handle 102.4 GB/s each (factoring in the uptick)?
     
  16. temesgen

    Veteran Regular

    Joined:
    Jan 1, 2007
    Messages:
    1,536
    Likes Received:
    327
    I hope at some point we get some answers at any rate here is an article from ARS Tech on the subject:

    http://arstechnica.com/gaming/2013/...s-xbox-one-from-accusations-its-underpowered/
     
  17. taisui

    Regular

    Joined:
    Aug 29, 2013
    Messages:
    674
    Likes Received:
    0
    The leak was based on that you can't do R/W simultaneously.

    Therefore when doing ESRAM->ESRAM, BW was divided in the middle, 102.4/2 = 51.2.

    When doing ESRSM<->DRAM, the limit is bound by the DRAM, therefore, 68.2, even when the theoretical BW on the ESRAM is 102.4.

    If you are curious about how the design could be and how the math can line up correctly, go back previous page and read #212.
     
  18. blakjedi

    Veteran

    Joined:
    Nov 20, 2004
    Messages:
    2,975
    Likes Received:
    79
    Location:
    20001
    Believe him? I don't have a problem with the inquiry, I have an issue with discriminatory tone from some people stating that Penello is either shading the truth or outright lying. As if his job is to prove anything to them.

    You do realize that except in very brief circumstances ps4 will not achieve 176gb/s utilization with any regularity? Why don't you ask Cerny what their real world bandwidth figures look like while we are making demands. :roll:
     
  19. Rangers

    Legend

    Joined:
    Aug 4, 2006
    Messages:
    12,322
    Likes Received:
    1,120
    Semiaccurate Charlie said they are achieving 140-150GB/s with the ESRAM regularly. Which I suspect is a similar percent of BW GDDR5 or any other type of memory achieves of it's peak.

    This really implies it's a real, healthy figure and not some contrived case.

    http://semiaccurate.com/2013/08/30/a-deep-dive-in-to-microsofts-xbox-one-gpu-and-on-die-memory/

     
  20. ROG27

    Regular

    Joined:
    Oct 27, 2005
    Messages:
    572
    Likes Received:
    4
    The ~133 GB/s number was pre-uptick. Between 140 and 150 GB/s seems to make sense post-uptick. It's strange how Leadbetter's article seems to imply it is not a normal throughput (i.e., an outlier for certain post-processing operations) where Charlie's says otherwise. Charlie also seemed to think that they overclocked the CPU at ~1.9 GHz when in fact it was a more modest 1.75 GHz (see semiaccurate).
     
Loading...
Thread Status:
Not open for further replies.

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...