Review: AMD Radeon R9 Fury X

Discussion in 'Frontpage news' started by Hilbert Hagedoorn, Jun 24, 2015.

  1. Loophole35

    Loophole35 Guest

    Messages:
    9,797
    Likes Received:
    1,161
    GPU:
    EVGA 1080ti SC
    Just poking a bit. lol
    You are one person I do trust to be totally honest on here.

    I hope that drivers do make this card stronger. So far it seems that GCN doesn't scale well. (edit: I take that back it seems to scale normal 390X - FuryX is a 32% increase in core count the performance is about 25-30% difference)

    Have you had a chance to see what your VRM temps are at?
     
    Last edited: Jun 27, 2015
  2. mcfart

    mcfart Guest

    Messages:
    311
    Likes Received:
    2
    GPU:
    HD 5970
    If the Fury series (mostly talking Nano)is significantly better in DX12, then AMD stand to gain a lot since Pascal's still aways away.
     
  3. Fox2232

    Fox2232 Guest

    Messages:
    11,808
    Likes Received:
    3,371
    GPU:
    6900XT+AW@240Hz
    It's hard to make me angry over something :) And I am really happy that we found root cause for that another jump. And cowie could have been bit more direct as he noticed immediately.

    Here is something interesting on drivers:
    API overhead Test:
    API-Overhead
    1st is 15.200.1023.5 (no optimization driver on W10) HD7970 @925/1375
    2nd is 15.200.1040.0 (optimized driver on W10) HD7970 @925/1375
    3rd is 15.150.0.0 (no optimization driver for W7/8.1 hacked into W10) Fury X @1050/500
    4th is 15.200.1040 (optimized driver on W10) But not Fury X code - using Tonga code instead. Fury X @1050/500 - But not reaching even HD7970 scores.
    5th is 15.200.1040 (optimized driver on W10) But not Fury X code - using Tahiti code instead. Fury X @1050/500 - Got even closer to HD7970 scores.
    It is quite clear that release driver is not able to feed Fury X properly even with Tonga code path. I hope to see Fury X code in W10 test driver soon.
    Apparently Tahiti code path is better than Tonga's. And I rerun Firestrike, but result is same as one with Tonga. (that smaller boost, not that erroneous big one)
     
  4. cowie

    cowie Ancient Guru

    Messages:
    13,276
    Likes Received:
    357
    GPU:
    GTX
    I thought everyone knew fox no biggie

    can you run no tess on hwbot heaven bench?
    oh no matter we take it to the owners thread
     

  5. Ragingun

    Ragingun Guest

    Messages:
    982
    Likes Received:
    0
    GPU:
    295x2
    I see a lot of firestrike comparing going on here. I find it "interesting" to put it positively that there are still people whom this this is a good measure of graphics ability.
     
  6. Hootmon

    Hootmon Guest

    Messages:
    1,231
    Likes Received:
    6
    GPU:
    XFX THICC III Ultra
    All these comparisons depend on what you care about. Fury is about 4k and DX12. Unless that is your frame of reference these arguments miss each other's point.
     
  7. ---TK---

    ---TK--- Guest

    Messages:
    22,104
    Likes Received:
    3
    GPU:
    2x 980Ti Gaming 1430/7296
    Maybe bring that up to every single website that reviews video cards so they can remove firestrike from their benchmarking suites.
     
  8. Mineria

    Mineria Ancient Guru

    Messages:
    5,540
    Likes Received:
    701
    GPU:
    Asus RTX 3080 Ti
    Last edited: Jun 27, 2015
  9. mcfart

    mcfart Guest

    Messages:
    311
    Likes Received:
    2
    GPU:
    HD 5970
    We won't know that until there's some DX12 games out.
     
  10. Evildead666

    Evildead666 Guest

    Messages:
    1,309
    Likes Received:
    277
    GPU:
    Vega64/EKWB/Noctua
    Or various TechDemo's from various places.
    I would hope Nvidia and AMD, maybe MS has some DX12 Techdemo goodness ahead :)

    Maybe even a patch or two if we're lucky.
    Would have thought they would have said something by now, but then NDA's could be in force.

    The Beta testers would be first to know.
     

  11. Hootmon

    Hootmon Guest

    Messages:
    1,231
    Likes Received:
    6
    GPU:
    XFX THICC III Ultra
    Precisely. I certainly will not be buying any card until I see those results.

    Its too close...
     
    Last edited: Jun 27, 2015
  12. xIcarus

    xIcarus Guest

    Messages:
    990
    Likes Received:
    142
    GPU:
    RTX 4080 Gamerock
    In fact the 980Ti is a castrated version of the TX, while the 970 is a castrated version of the 980.
    The 960 is pretty much an impotent version of the 980.
     
  13. Noisiv

    Noisiv Ancient Guru

    Messages:
    8,230
    Likes Received:
    1,494
    GPU:
    2070 Super
    [​IMG]

    [​IMG]

    [​IMG]
     
  14. Ragingun

    Ragingun Guest

    Messages:
    982
    Likes Received:
    0
    GPU:
    295x2
    Most sights use 3dmark as just one out of 10 graphics tests and most are games. The 980ti hits what, 20k or so? My 295x2 hits 17,500 but it blows the doors off the 980ti in any game that allows multi GPU use. I'm not comparing the 980ti to the 295x2, I'm simply recognizing the fact that 3dmark and gaming are 2 completely separate things and 3dmark isn't even near a factual reflection of anything game related.

    I'm not putting down the guys that have fun with 3dmark, I was once in the top 30 with my 980's but it certainly isn't comparable to anything but itself.
     
    Last edited: Jun 27, 2015
  15. Ragingun

    Ragingun Guest

    Messages:
    982
    Likes Received:
    0
    GPU:
    295x2
    Correct, that was my point.
     

  16. ---TK---

    ---TK--- Guest

    Messages:
    22,104
    Likes Received:
    3
    GPU:
    2x 980Ti Gaming 1430/7296
    You did not say that at all. You said a 980 and 980ti are the same gpu. The 980ti is much closer in performance to a titan x. About 3% difference.
     
  17. -Tj-

    -Tj- Ancient Guru

    Messages:
    18,095
    Likes Received:
    2,601
    GPU:
    3080TI iChill Black
    This time Maxwell looks the same across Gx204 and Gx200, so they had some time and I agree FuryX driver will improve more.
    Every new chip did, btw 1000points is what 2-3fps in firestrike, this can easily be driver optimized
     
  18. ---TK---

    ---TK--- Guest

    Messages:
    22,104
    Likes Received:
    3
    GPU:
    2x 980Ti Gaming 1430/7296
    1000 points were due to tesselation manipulation. There were extra fps in the firestrike tests that have tesselation.
     
  19. xIcarus

    xIcarus Guest

    Messages:
    990
    Likes Received:
    142
    GPU:
    RTX 4080 Gamerock
    Exactly.
    To add to that they're basically the same chip aswell, the GM200. 980/970 are a smaller, GM204 chip.
     
  20. GreenAlien

    GreenAlien Active Member

    Messages:
    63
    Likes Received:
    5
    GPU:
    Sapphire Nitro+ RX 480 OC
    They called it Fury and made it out to be a better Titan. What a joke. There was no reason to give this card a new name. They should have just named it 390X. This card needs to lose another 100€ to be even a bit competitive.

    AMD did it again. They adopted 4k too early just like they did 8 cores, and they will continue to fail if they dont learn from their mistakes.
     
    Last edited: Jun 28, 2015

Share This Page