Page 7 of 9 FirstFirst ... 456789 LastLast
Results 97 to 112 of 134

Thread: AMD Polaris GPU: detailed slides leaked

  1. #97
    Banhammer in peace PeterB kalniel's Avatar
    Join Date
    Aug 2005
    Posts
    31,038
    Thanks
    1,878
    Thanked
    3,379 times in 2,716 posts
    • kalniel's system
      • Motherboard:
      • Gigabyte Z390 Aorus Ultra
      • CPU:
      • Intel i9 9900k
      • Memory:
      • 32GB DDR4 3200 CL16
      • Storage:
      • 1TB Samsung 970Evo+ NVMe
      • Graphics card(s):
      • nVidia GTX 1060 6GB
      • PSU:
      • Seasonic 600W
      • Case:
      • Cooler Master HAF 912
      • Operating System:
      • Win 10 Pro x64
      • Monitor(s):
      • Dell S2721DGF
      • Internet:
      • rubbish

    Re: AMD Polaris GPU: detailed slides leaked

    Quote Originally Posted by watercooled View Post
    The AMD slides state they used an i7 4790k in the test systems - an unusual choice as it's not Haswell's most efficient bin, but again I suppose it comes down to the 'if you think we're deliberately CPU-limiting the GPU to skew the results somehow, you're wrong' thing.
    Given the fact nVidia cards tend to be less demanding on the CPU pre DX12 it'd be particularly stupid for AMD to demo on slow CPUs

  2. #98
    Senior Member watercooled's Avatar
    Join Date
    Jan 2009
    Posts
    11,478
    Thanks
    1,541
    Thanked
    1,029 times in 872 posts

    Re: AMD Polaris GPU: detailed slides leaked

    Yeah I'd thought of that when looking at the power consumption numbers - if anything you'd expect CPU power draw to be higher on the Polaris system than the 950 one.

    But WRT the 4970k thing, I don't imagine picking a 4970 non-k or an i5 would have made much of a difference to performance, but system power consumption might have been lower (though it depends on what clocks they end up running at under partial load), making for a more impressive difference on the slide. But again, picking something else might have led to people nit-picking the choice as something suspicious.

    It seems like AMD have given away a really small amount of detail to get people interested, but not enough to reveal much about performance/positioning of the card - e.g. the FPS cap prevents us from seeing the card's actual performance.

  3. #99
    Moosing about! CAT-THE-FIFTH's Avatar
    Join Date
    Aug 2006
    Location
    Not here
    Posts
    32,042
    Thanks
    3,909
    Thanked
    5,213 times in 4,005 posts
    • CAT-THE-FIFTH's system
      • Motherboard:
      • Less E-PEEN
      • CPU:
      • Massive E-PEEN
      • Memory:
      • RGB E-PEEN
      • Storage:
      • Not in any order
      • Graphics card(s):
      • EVEN BIGGER E-PEEN
      • PSU:
      • OVERSIZED
      • Case:
      • UNDERSIZED
      • Operating System:
      • DOS 6.22
      • Monitor(s):
      • NOT USUALLY ON....WHEN I POST
      • Internet:
      • FUNCTIONAL

    Re: AMD Polaris GPU: detailed slides leaked

    It could be a cunning ploy to force Nvidia's hand too,to see what their response is.

  4. #100
    Registered User
    Join Date
    May 2012
    Posts
    11
    Thanks
    0
    Thanked
    0 times in 0 posts

    Re: AMD Polaris GPU: detailed slides leaked

    I hope AMD do get a win out of this. I think if they fail with the 400-series it might end up being very bad in the long run for us consumers.

  5. #101
    Senior Member
    Join Date
    Nov 2007
    Posts
    1,061
    Thanks
    10
    Thanked
    39 times in 38 posts

    Re: AMD Polaris GPU: detailed slides leaked

    Well it looks like Polaris will deliver the goods, and then some within the laptop and notebook markets so I think they have two wins already. If AMD can transform the power saving into performance with the high spec Polaris chips, then we should see very large jump in processing power as the power use increases. I'm fully expecting AMD to run away with DX12/Vulkan and VR performance with Polaris.
    Last edited by jigger; 15-01-2016 at 07:38 PM.

  6. #102
    Senior Member watercooled's Avatar
    Join Date
    Jan 2009
    Posts
    11,478
    Thanks
    1,541
    Thanked
    1,029 times in 872 posts

    Re: AMD Polaris GPU: detailed slides leaked

    If they can bolt a couple of CPU cores on that sort of GPU they'd have a heck of a mobile APU! 14nm could allow some impressive performance from APUs but as others have said I guess it will come down to yield and pricing, especially at first, which is why we probably won't see 14nm APUs for a while after the discrete parts.

  7. #103
    Senior Member
    Join Date
    Dec 2013
    Posts
    3,526
    Thanks
    504
    Thanked
    468 times in 326 posts

    Re: AMD Polaris GPU: detailed slides leaked

    We probably won't see Polaris in an APU until ZEN hits the market towards the end of the year, that's if nothing goes wrong or gets delayed, all the signs are looking good for AMD so fingers crossed.

  8. #104
    Banhammer in peace PeterB kalniel's Avatar
    Join Date
    Aug 2005
    Posts
    31,038
    Thanks
    1,878
    Thanked
    3,379 times in 2,716 posts
    • kalniel's system
      • Motherboard:
      • Gigabyte Z390 Aorus Ultra
      • CPU:
      • Intel i9 9900k
      • Memory:
      • 32GB DDR4 3200 CL16
      • Storage:
      • 1TB Samsung 970Evo+ NVMe
      • Graphics card(s):
      • nVidia GTX 1060 6GB
      • PSU:
      • Seasonic 600W
      • Case:
      • Cooler Master HAF 912
      • Operating System:
      • Win 10 Pro x64
      • Monitor(s):
      • Dell S2721DGF
      • Internet:
      • rubbish

    Re: AMD Polaris GPU: detailed slides leaked

    Quote Originally Posted by jigger View Post
    Well it looks like Polaris will deliver the goods, and then some within the laptop and notebook markets so I think they have two wins already. If AMD can transform the power saving into performance with the high spec Polaris chips, then we should see very large jump in processing power as the power use increases. I'm fully expecting AMD to run away with DX12/Vulkan and VR performance with Polaris.
    I'm not sure they will run away with it. CPU usage seems good in DX12, but DX12 is a very small market. But look at what nVidia managed with Maxwell - if they get anything close to the same power improvements from finfet that AMD claim then pascal is going to be quite amazing.

  9. #105
    Senior Member
    Join Date
    Nov 2007
    Posts
    1,061
    Thanks
    10
    Thanked
    39 times in 38 posts

    Re: AMD Polaris GPU: detailed slides leaked

    The R9 290 pritty much made £1000 Maxwell card obsolete for a third of the price. The second round of Maxwell cards come a year? after Hawaii, cost a lot more and the DX11 performance is nothing special and they are still getting a hard time from the R9 290 cards today. Apart from the low power use, I'm not sure Nvidia managed that much with Maxwell. The next gen API performance is very questionable compared to Hawaii never mind Fiji.

  10. #106
    Banhammer in peace PeterB kalniel's Avatar
    Join Date
    Aug 2005
    Posts
    31,038
    Thanks
    1,878
    Thanked
    3,379 times in 2,716 posts
    • kalniel's system
      • Motherboard:
      • Gigabyte Z390 Aorus Ultra
      • CPU:
      • Intel i9 9900k
      • Memory:
      • 32GB DDR4 3200 CL16
      • Storage:
      • 1TB Samsung 970Evo+ NVMe
      • Graphics card(s):
      • nVidia GTX 1060 6GB
      • PSU:
      • Seasonic 600W
      • Case:
      • Cooler Master HAF 912
      • Operating System:
      • Win 10 Pro x64
      • Monitor(s):
      • Dell S2721DGF
      • Internet:
      • rubbish

    Re: AMD Polaris GPU: detailed slides leaked

    It's the power usage I'm talking about - quite incredible really. If they get the same kind of gains from finfet that AMD did then they've got a lot of scope for all sorts of good things.

  11. #107
    Senior Member
    Join Date
    Nov 2007
    Posts
    1,061
    Thanks
    10
    Thanked
    39 times in 38 posts

    Re: AMD Polaris GPU: detailed slides leaked

    The power use was very good and much better than first round of Maxwell cards, but people made too much of the TDP numbers Nvidia marketed. In truth, the real world difference between Hawaii and Maxwell wasn't that big of a deal especially considering the price difference and time it took Nvidia to get the card to market.

    Anyway, I was talking about the high spec cards, and I'm not sure how power use is related to DX12/Vulkan and VR performance?
    Last edited by jigger; 15-01-2016 at 10:59 PM.

  12. #108
    root Member DanceswithUnix's Avatar
    Join Date
    Jan 2006
    Location
    In the middle of a core dump
    Posts
    13,009
    Thanks
    781
    Thanked
    1,568 times in 1,325 posts
    • DanceswithUnix's system
      • Motherboard:
      • Asus X470-PRO
      • CPU:
      • 5900X
      • Memory:
      • 32GB 3200MHz ECC
      • Storage:
      • 2TB Linux, 2TB Games (Win 10)
      • Graphics card(s):
      • Asus Strix RX Vega 56
      • PSU:
      • 650W Corsair TX
      • Case:
      • Antec 300
      • Operating System:
      • Fedora 39 + Win 10 Pro 64 (yuk)
      • Monitor(s):
      • Benq XL2730Z 1440p + Iiyama 27" 1440p
      • Internet:
      • Zen 900Mb/900Mb (CityFibre FttP)

    Re: AMD Polaris GPU: detailed slides leaked

    Quote Originally Posted by kalniel View Post
    It's the power usage I'm talking about - quite incredible really.
    I think that was totally necessary to have any chance of Tegra working, and Nvidia did throw a lot behind Tegra. So far it hasn't worked out for them, I think that is a shame specially as a lot of the image pain was from their old DX9 shaders in the older Tegra devices (like the HTC One X I used to have) so Maxwell really needed to be power efficient. I think it is too late, tablet and phone markets are now tied up and gone to Nvidia leaving things like Chromebooks for which Tegra is really very good (I got my wife a Tegra based HP Chromebook for Christmas).

    There seem to be claims that gaming laptops are going to be the next big thing, but I see that as less likely than the Ultrabook revolution that never actually happened.

    So yes Nvidia have superb power usage, but I don't know anyone who really cares, if they cared about power or heat they wouldn't buy 970s or water cooled SLI dual 980ti setups. They want the speed, and they trust the drivers.

    I'm sure the designers of the PS4 and XB1 would have loved more power efficiency, but it seems that wasn't a deal breaker either. I think those are the most power sensitive mass production devices I can think of atm.

  13. #109
    Senior Member
    Join Date
    Mar 2010
    Posts
    2,567
    Thanks
    39
    Thanked
    179 times in 134 posts

    Re: AMD Polaris GPU: detailed slides leaked

    if you look at the factory overclocked gtx 980ti - they are right up there with the highest end amd cards in power draw

    as cat said - only the gtx 750ti and the gtx 980 are low power cards , the others? not as great as nv want you to believe

  14. #110
    Moosing about! CAT-THE-FIFTH's Avatar
    Join Date
    Aug 2006
    Location
    Not here
    Posts
    32,042
    Thanks
    3,909
    Thanked
    5,213 times in 4,005 posts
    • CAT-THE-FIFTH's system
      • Motherboard:
      • Less E-PEEN
      • CPU:
      • Massive E-PEEN
      • Memory:
      • RGB E-PEEN
      • Storage:
      • Not in any order
      • Graphics card(s):
      • EVEN BIGGER E-PEEN
      • PSU:
      • OVERSIZED
      • Case:
      • UNDERSIZED
      • Operating System:
      • DOS 6.22
      • Monitor(s):
      • NOT USUALLY ON....WHEN I POST
      • Internet:
      • FUNCTIONAL

    Re: AMD Polaris GPU: detailed slides leaked

    Quote Originally Posted by kalniel View Post
    I'm not sure they will run away with it. CPU usage seems good in DX12, but DX12 is a very small market. But look at what nVidia managed with Maxwell - if they get anything close to the same power improvements from finfet that AMD claim then pascal is going to be quite amazing.
    But the problem is Nvidia has ripped a lot out to get there. AMD has primarily higher power consumption due to their use of hardware scheduling and Maxwell ripped out a lot of compute from Kepler to help too. This is why the GK210 is still being used in commerical markets.

    Nvidia has advertised Pascal has having a renewed focus on compute and if they do move to adding back more hardware functionality its going to mean more power consumption,especially if they like AMD are having a renewed push towards things like VR.

    There is no free lunch here. I can see them converging in some ways and some of their figures are some what marketing based(The GTX970 is meant to be a lower power consumption and TDP class than a GTX770 when it really is not for example).

    Nvidia has probably trialled as much power saving tech as they can with Maxwell,beyond using HBM/HBM2,and adding functionality in hardware is probably going to add to power consumption.

    AMD has a lot of functionality still in hardware(especially regarding some of the stuff added for VR which they had a year before Nvidia and still have more functionality now),and the same even goes for support for adaptive sync at the GPU level,etc.

    OTH,they have probably not as refined as Nvidia with regards to power saving tech,so probably have caught up in Polaris,in this regard and might even start to remove some stuff which is not really required in hardware,to cut on power consumption.

    Basically,I don't think Nvidia has anywhere as much of a scope to drop power consumption on a purely design level as AMD does,since AMD has simply more functionality they could drop to reduce power consumption. Nvidia has done as much as they can and thats why they are forced to still use Kepler for some markets.

    Edit!!

    Do I think the next 300MM2 to 400MM2 Nvidia GPU is going to consume a bit less power than the next AMD 300MM2 to 400MM2 one??

    Yep,since the AMD GPU will be dual use with transistors and portions dedicated towards the GPU towards pure DP performance,a bit like the HD7970 vs the GTX680,and the Nvidia GPU won't be top end but since they will wait until they can get a 500MM2 monster out first.

    People were quick to judge that GCN1.0 was worse off in efficiency than Kepler,but failed to realise than it was not really a valid uarch comparison.

    If you looked down the range and equivalent GPUs like Pitcairn and the GK106,they traded blows and I think for most cases it will be the same with Polaris against Pascal.
    Last edited by CAT-THE-FIFTH; 16-01-2016 at 12:27 AM.

  15. Received thanks from:

    Corky34 (16-01-2016),kalniel (16-01-2016)

  16. #111
    Senior Member watercooled's Avatar
    Join Date
    Jan 2009
    Posts
    11,478
    Thanks
    1,541
    Thanked
    1,029 times in 872 posts

    Re: AMD Polaris GPU: detailed slides leaked

    Quote Originally Posted by Corky34 View Post
    We probably won't see Polaris in an APU until ZEN hits the market towards the end of the year, that's if nothing goes wrong or gets delayed, all the signs are looking good for AMD so fingers crossed.
    Yeah I was also referring to the Zen-based HEDT CPU as one of those discrete parts I mentioned.

  17. #112
    Registered+
    Join Date
    Dec 2015
    Posts
    20
    Thanks
    0
    Thanked
    1 time in 1 post

    Re: AMD Polaris GPU: detailed slides leaked

    As a casual gamer I don't really get the top end cards but I'm really excited by this and whatever nvidia follow up with. Low power, smaller cards with the meat to play the top titles whenever you need it

Page 7 of 9 FirstFirst ... 456789 LastLast

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •