Page 253 of 368 FirstFirst ... 153203243251252253254255263303353 ... LastLast
Results 2,521 to 2,530 of 3678

Thread: Vega 10, Vega 11. What we know

  1. #2521
    Senior Member
    Join Date
    Jun 2010
    Posts
    4,228
    Quote Originally Posted by Nerdmaster View Post
    Vega has to be at least 20% faster than 1080 otherwise it will be failure. It is a much larger gpu with many more transistors and it has higher clocks than the older amd cards. It should be able to compete with pre OC 1080 cards and beat them.
    It will also depend on the price, but yes, the architecture would be bad.
    In any case gtx 480 (529mm^2 and 300w) was 10% faster (less at higher resolutions) than 5870 (334mm^2 and 184w), the comparison here would be Vega and gtx 1080. And to be worse it should also produce more noise, be hotter, cost 100$ more, etc.

    Quote Originally Posted by TESKATLIPOKA View Post
    1. I am not confusing It with speculation. In this thread you can find a lot of hype up comments and even more speculations.
    2. I think you are thinking about something else.
    Not sure what you call hype, in any case some people with the data AMD made public expects or expected more performance than gtx 1080ti, others than 1080, and a few ones than 1070.
    We'll see because currently we don't have any idea of the performance of RX Vega, there could have been hype or downplaying, who knows, I don't know the future.
    Sure:
    http://semiaccurate.com/forums/showp...postcount=2379
    There are more of those.
    It's also very different to post a card that has not launched is a winner or a disaster confronting all people than to say you think it will be this or that because of reasons. Still I have not seen anyone saying it is a winner, but a disaster... very curious for a thread full of "hype".

    Quote Originally Posted by french toast View Post
    Vega 10 has to be compared to GPUs in its price bracket and/or size.
    In this case Titan xp and 1080ti, both of which are last year's architecture, which is heavily based on 2014/15 Maxwell dx11 architecture.
    Vega is larger than both of those GPUs, is AMDs latest architecture, has the advantage of hbm2, consumes about 20℅ more power.

    It's largely beats out Pascal in compute (loses in some) but that has to be balanced against its perf/mm2 and perf/watt which somewhat closes the average somewhat.
    In gaming- assuming new drivers don't bring large performance gains performance is between a gtx 1070! And a gtx1080! ...that's shocking.

    Honestly I can't believe that is representative of final game performance, it's nigh on impossible for AMD to create a card like this with the resources and technology available to Vega, so drivers surely have to make a difference.
    If not it's the biggest flop in gpu history I'm sad to say.
    I only compared it to the gtx 480, a comparison than other people brought here.
    I expect at least Fury X perf/FLOP with new drivers, now it seems far behind but I don't trust PCPER, if RX Vega improves less than 25% (GTX 1080 performance) over Fury X it would be one of the worst gpu architectures ever.

  2. #2522
    Quote Originally Posted by DCO View Post
    It will also depend on the price, but yes, the architecture would be bad.
    In any case gtx 480 (529mm^2 and 300w) was 10% faster (less at higher resolutions) than 5870 (334mm^2 and 184w), the comparison here would be Vega and gtx 1080. And to be worse it should also produce more noise, be hotter, cost 100$ more, etc.


    Not sure what you call hype, in any case some people with the data AMD made public expects or expected more performance than gtx 1080ti, others than 1080, and a few ones than 1070.
    We'll see because currently we don't have any idea of the performance of RX Vega, there could have been hype or downplaying, who knows, I don't know the future.
    Sure:
    http://semiaccurate.com/forums/showp...postcount=2379
    There are more of those.
    It's also very different to post a card that has not launched is a winner or a disaster confronting all people than to say you think it will be this or that because of reasons. Still I have not seen anyone saying it is a winner, but a disaster... very curious for a thread full of "hype".


    I only compared it to the gtx 480, a comparison than other people brought here.
    I expect at least Fury X perf/FLOP with new drivers, now it seems far behind but I don't trust PCPER, if RX Vega improves less than 25% (GTX 1080 performance) over Fury X it would be one of the worst gpu architectures ever.

    Are you another flatliner? It has clearly been determined that the drivers on the F.E. Vega are primitive. They are actually FIJI drivers with little exception. They are not using the new hardware in Vega. The final drivers should be some 25% better than the current state of F.E. drivers. Plus the clock speed on RX Vega wil undoubtedly be at least 100mhz higher than F.E. That would mean at least 30% better performance.

  3. #2523
    Senior Member
    Join Date
    Dec 2012
    Posts
    7,390
    Quote Originally Posted by os2wiz View Post
    Are you another flatliner? It has clearly been determined that the drivers on the F.E. Vega are primitive. They are actually FIJI drivers with little exception. They are not using the new hardware in Vega. The final drivers should be some 25% better than the current state of F.E. drivers. Plus the clock speed on RX Vega wil undoubtedly be at least 100mhz higher than F.E. That would mean at least 30% better performance.
    While I am happy if this is true, were was the confirmation? Thank you
    -Q

  4. #2524
    Senior Member
    Join Date
    Jan 2012
    Posts
    827
    Quote Originally Posted by os2wiz View Post
    Are you another flatliner? It has clearly been determined that the drivers on the F.E. Vega are primitive. They are actually FIJI drivers with little exception. They are not using the new hardware in Vega. The final drivers should be some 25% better than the current state of F.E. drivers. Plus the clock speed on RX Vega wil undoubtedly be at least 100mhz higher than F.E. That would mean at least 30% better performance.
    Whilst I'd love to see that, we have been given NO indication that is true, all amd said was Rx would be faster and cheaper, that's it.
    According to rys the driver is an older driver.but not gimped, it's s recent one though from.whst I've read, amd didn't tell pcper that anything was especially wrong with the performance of Vega Fe.

    It's possible that they have been saving a massive launch day driver for Rx Vega, but that is guess work and blind hope on our part, amd has not hinted at that being the case which you would think they would after this bad press.

  5. #2525
    Senior Member
    Join Date
    Jan 2010
    Posts
    1,882
    Question for you "pros" here.

    If you were writing drivers for a card like the Vega FE would you include all the individual per game optimizations? Or would you write the drivers to make the cards features work with code that was written accurately as per the API? (Hopefully I've described this so it makes sense.)

    I think I would do the latter. It's not for playing games. It's for designing them.

  6. #2526
    Senior Member
    Join Date
    Dec 2010
    Location
    Switzerland Geneva Cern
    Posts
    2,904
    Quote Originally Posted by Relayer View Post
    Question for you "pros" here.

    If you were writing drivers for a card like the Vega FE would you include all the individual per game optimizations? Or would you write the drivers to make the cards features work with code that was written accurately as per the API? (Hopefully I've described this so it makes sense.)

    I think I would do the latter. It's not for playing games. It's for designing them.
    It makes sense for amd to use non optimised drivers for vega FE to avoid crashes but the argument is that the gaming vega launch is only 1 month away so proper drivers should have already be almost prepared by now.
    Current rig: i5 3570 @3.8GHz, Asus 7750, Dell 24'' 2560X1440, Corsair Vengence 16GB Ram DDR3 @1600MHz, Chieftec 550W PSU, 480GB OCZ SSD + 500GB Seagate + 1TB WD hd, Windows 10 64-bit

  7. #2527
    Senior Member
    Join Date
    May 2012
    Posts
    445
    Quote Originally Posted by testbug00 View Post
    While I am happy if this is true, were was the confirmation? Thank you
    Gaming drivers have been decompiled by a couple of different people now.

    It's been confirmed that the code path is Fiji with very few modifications, and the OS sees the device as Fiji with more memory.

    Given they were using something similar 4-5 months ago when it was first shown, I'm fairly confident the driver team does at least have beta Vega-specific RX drivers by now.

    Performance uplift ought to be enormous.

  8. #2528
    Senior Member
    Join Date
    Jan 2010
    Location
    London
    Posts
    3,211
    Guys, don't force me to ditch my holiday at FOS in Goodwood and come down here to wooop your arses. FFS it's like all of you just had a major brainfarts and can't think straight I see only couple of people who did not have a sunstroke. FE is a frikkin content creation card with generic drivers. Try installing generic MS drivers from back in a day when MS used their own basic drivers. FE driver is similar. It has pro drivers with all the bells and whistles to wooop Titan XP around and on a side it has generic stable "gaming" driver which lets you load the game you develop and test the code. Most of you sound so silly it's not even funny anymore.
    Now if you let me get back to my daily festivities at Goodwood. Subaru is about to wooop everyone's arse

  9. #2529
    Senior Member
    Join Date
    Sep 2011
    Posts
    866
    Quote Originally Posted by DCO View Post
    Sure:
    http://semiaccurate.com/forums/showp...postcount=2379
    There are more of those.
    It's also very different to post a card that has not launched is a winner or a disaster confronting all people than to say you think it will be this or that because of reasons. Still I have not seen anyone saying it is a winner, but a disaster... very curious for a thread full of "hype".
    As I already said, I was talking about something else. I wasn't talking about his deleted posts by the moderators, I don't even know what he wrote to cause them to be deleted, so I can hardly comment them. I was talking about a situation which happened days before he had those deleted posts.

    If someone says It's a winner or disaster, I think It's better If he also writes his reasoning. On the other hand, If he doesn't, what does It change? It's just his own opinion. You can agree or simply disagree with his opinion.

    Yes, yes, in this thread with ~253 pages, there is not a single hype post, because you didn't see It.
    If you focus only on the last ~10-20 pages then I am not surprised you only see posts about disaster.

  10. #2530
    Senior Member
    Join Date
    Nov 2013
    Posts
    1,012
    https://twitter.com/GFXChipTweeter/s...70308694822913

    Raja has confirmed Vega 10's die size to be 484mm^2 large. Looks like PCPer was off by a large margin even though so many people took their guestimate as objective truth.

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
WordPress Appliance - Powered by TurnKey Linux