AMD Radeon Software Crimson ReLive 17.5.2 driver Download & Discussion

Discussion in 'Videocards - AMD Radeon Drivers Section' started by Hilbert Hagedoorn, May 18, 2017.

  1. Passus

    Passus Ancient Guru

    Messages:
    1,618
    Likes Received:
    446
    GPU:
    RX 5700 Mech OC
    It is MS fault mate, they will have bundled 17.1.1 with the creators update

    it happened to me with a few insider builds, I don't use the test versions anymore but since that after any big windows update i uninstall old drivers first knowing it will install it's own then use DDU after and install the latest
     
  2. ReaperXL2

    ReaperXL2 Member Guru

    Messages:
    178
    Likes Received:
    18
    GPU:
    EVGA 1080ti SC2


    Wait what !!! There is a working 17.5.1 driver :gape: Where can I find this? or have I misread?

    Fyi: I, like yourself, am disgusted with regard to the FH3 debacle. MS & AMD should hang their heads in shame !

    DX12 !!! Sh*t, FH3 should fly given that MS is the birth mother of dx12 & amd the father of related native hardware support... Sheesh :bang:
     
    Last edited: May 20, 2017
  3. passenger

    passenger Master Guru

    Messages:
    401
    Likes Received:
    59
    GPU:
    Sapphire RX580 8GB
    It's better to go back to whql 17.4.4 until 'graphics division' get their sh't together. Many bugs it's like one person workin' in there, they really need to ryze up their game.
     
  4. Dch48

    Dch48 Ancient Guru

    Messages:
    1,822
    Likes Received:
    1
    GPU:
    Sapphire Nitro+ RX 470 4g
    No reason unless you have R or RX series. 99.9% chance it won't improve anything for 7000 cards.
     

  5. Chastity

    Chastity Ancient Guru

    Messages:
    2,646
    Likes Received:
    862
    GPU:
    Nitro 5700 XT
    Been playing SWTOR as of late, and the in-game AA options are performance hogs. Using driver overrides works very well in alleviating this bottleneck, which, unfortunately, do not work since 17.5.1.

    I've gotten around this by using VSR @ 2880 x 1620.

    I've submitted bug reports to AMD about this.

    On another note, for people looking for help with Strix BIOS issues, you should try contacting Gary Key. He's Asus' Technical Marketing Manager and is more responsive to such issues. I was with Asus Customer Care, and I know that the current Asus Customer Support is inadequate for handling this, as they're a 3rd party contractor. Customer Care division no longer exists, and we would have been the ones to handle this kind of issue, as we could report it to the TPM's, who would then confirm and notify the engineers in Taiwan. Gary Key also has direct contacts with AMD and could get the ball rolling on this too.

    Sorry, but I can no longer help out like I used to.
     
    Last edited: May 20, 2017
  6. OnnA

    OnnA Ancient Guru

    Messages:
    12,429
    Likes Received:
    3,090
    GPU:
    Vega XTX LiQuiD
    That's why i said: really ATI :bang:
    It's a god da!mn shame

    But we have to forgive them, becouse:
    1. ZEN (it is really good)
    2.Vega will have good drivers at launch

    But shame is shame - period!

    17.5.1 is still live IMO [for Fiji HBM is one of the best WDDM 2.2 drivers now, all is working for me NP]
    Here -> http://support.amd.com/en-us/download/desktop/previous/detail?os=Windows 10 - 64&rev=17.5.1
     
    Last edited: May 20, 2017
  7. ReaperXL2

    ReaperXL2 Member Guru

    Messages:
    178
    Likes Received:
    18
    GPU:
    EVGA 1080ti SC2

    Ok, thanx dude :thumbup: I'm on the same page as you now :banana:
     
  8. Passus

    Passus Ancient Guru

    Messages:
    1,618
    Likes Received:
    446
    GPU:
    RX 5700 Mech OC
    Me him basically have the same card 7950=280 and i see improvements in new drivers so why wouldn't he
     
  9. Undying

    Undying Ancient Guru

    Messages:
    14,606
    Likes Received:
    3,770
    GPU:
    Aorus RX580 XTR 8GB
    Exactly.

    AMD didnt drop support on gcn1.0, yet so just update your drivers.

    If we had no improvements the 7970 would not be this far from his Kepler competitor.
     
    Last edited: May 20, 2017
  10. Dch48

    Dch48 Ancient Guru

    Messages:
    1,822
    Likes Received:
    1
    GPU:
    Sapphire Nitro+ RX 470 4g
    If you do fine. I personally have never seen much improvement with any driver in any generation of cards but I have seen problems with older hardware trying to use newer drivers more than once. Having said that, I forgot that some 7000 series cards aren't much different from the R series that replaced them so maybe you could see improvements. I don't see any difference in performance with my card between 17.2 and 17.5 except for in one game, FH3, where anything newer than 17.2 is basically unusable.
     

  11. siriq

    siriq Master Guru

    Messages:
    791
    Likes Received:
    14
    GPU:
    Evga GTX 570 Classified
    GCN 1 has async compute back on the table, it is enabled already in prev drives as well.
     
  12. Romulus_ut3

    Romulus_ut3 Master Guru

    Messages:
    670
    Likes Received:
    138
    GPU:
    AMD RX 570 4GB
    Apparently you don't have issues that all other GCN 1.0 consumers are suffering from either.
     
    Last edited: May 21, 2017
  13. Rambo

    Rambo Master Guru

    Messages:
    274
    Likes Received:
    140
    GPU:
    RX 560 w8 4 Vega 56
    HD 7000 and RX XXX are basiclly the same so improvements should be noticeable under every of GCN cards.
     
  14. Romulus_ut3

    Romulus_ut3 Master Guru

    Messages:
    670
    Likes Received:
    138
    GPU:
    AMD RX 570 4GB
    The amount of misinformation spread by die hard AMD fanboys are nauseating. The Kepler cards didn't stop receiving support despite numerous uninformed claims.

    Here's a review done by Steve Walton who's also at Techspot.

    https://www.youtube.com/watch?v=BfSi-Z8r12M


    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    As you can see from the marked date that on the 11th of April, Unigine Superposition was launched.

    [​IMG]

    And it can also be seen from the date the Youtube video was published featuring the GTX 770, it was done on the same date/immediately after the Benchmark was launched and the nvidia drivers likely doesn't have any optimization for it whatsoever. Despite this, my GTX 770 4GB scores 1500+ with an unoptimized driver, so a 100% increase of VRAM on a dated "gimped" Kepler results in 300 extra points. And also, Boost 2.0 and Boost 3.0 aren't the same, and at stock, Kepler cards aren't always going to run at their maximum boost clock. The above comparison is very much invalid. But comparisons like these are what bolstered claims that GCN 1.0 is vastly superior than their Kepler counterparts. GTX 970 is seen often outperforming the R9 390/390X. What actually bothers me is the lead RX 480/580 has gotten over the R9 390/390X in newer games, not to mention the abysmal driver support. So whose gimping?

    :infinity:

    So much for HD 7970 being superior than it's Kepler counterpart. Where are the improvements? All I see is delusion.
     
    Last edited: May 21, 2017
  15. prazola

    prazola Member Guru

    Messages:
    176
    Likes Received:
    16
    GPU:
    R9390XSOC / R9290DCU2OC
    lol 9/11 confirmed. What an argument :wanker:
     

  16. Romulus_ut3

    Romulus_ut3 Master Guru

    Messages:
    670
    Likes Received:
    138
    GPU:
    AMD RX 570 4GB
    9/11 Confirmed!

    Indeed. I am sure Hardware Unboxed hates AMD, as clearly the benchmark and the numbers taken above are fake and made up too like 9/11 was.

    Do a DDU and then install 17.4.1 and then see how that affects your scores vs. 17.5.1

    :stewpid:

    TechPowerUp has this nice thread on Unigine Superposition Benchmark scores.

    Here's a result from it which I found interesting:

    [​IMG]

    The R9 280 result at 1150/1550 with dated drivers from the aforementioned thread:

    [​IMG]

    (Source)

    And my HD 7950 result at 1080/1500 with ReLive 17.5.1:

    [​IMG]

    Of course, Swamp Monster has a weaker CPU than I do. But as it's evident, that having an optimized driver for a certain benchmark or a game matters as you can clearly see his GPU is running 100/50 MHz faster than mine but the scores are pretty close despite that. Driver optimization can make a night and day difference, but apparently that is something that has to be discarded when you're comparing GCN 1.0 to Kepler.

    So yeah, 9/11 confirmed. :banana:

    I wish I didn't need to Bruteforce CUDA on my other rig pretty much 24/7, or I would've posted some Kepler scores of my own. :bang:

    That would've literally spilled :puke2: over some of the bold claims I see in this part of the forum.
     
    Last edited: May 21, 2017
  17. prazola

    prazola Member Guru

    Messages:
    176
    Likes Received:
    16
    GPU:
    R9390XSOC / R9290DCU2OC
    17.4.1? Here we are, wit my old rig.
     
  18. Romulus_ut3

    Romulus_ut3 Master Guru

    Messages:
    670
    Likes Received:
    138
    GPU:
    AMD RX 570 4GB

    [​IMG]

    Foreceware 376.33. Much older than even 17.4.1, surely R9 390 craps all over Kepler, right?

    (Source: Here)

    And as you say, CPU doesn't matter in this case, if true, bolsters my argument above that a person with a higher overclock on the same card as mine is achieving the same scores as me and the only variable apart from the CPU is the driver?

    For a more fair comparison:

    Xeon X5650 4 GHz | R9 290X @ 1100 MHz (actual boost clock) | 1250 MHz

    [​IMG]

    Source: Here

    VS.

    Kanan|i7 3960X 4.5 GHz|Asus GTX 780 Ti DCU II @ 1150-1220 MHz|1925 MHz

    [​IMG]

    Source: Here

    Such GCN Superiority (complex) much wow.
     
    Last edited: May 21, 2017
  19. prazola

    prazola Member Guru

    Messages:
    176
    Likes Received:
    16
    GPU:
    R9390XSOC / R9290DCU2OC
    Gtx780ti has almost same TF of Hawaii but with better GPU usage. Is not strange it's not lagging behind. Nice 1300 MHz BTW :D
    Now show me gtx680 ty.
     
  20. Virs

    Virs Member Guru

    Messages:
    164
    Likes Received:
    0
    GPU:
    R9 390
    In the past I tried tweeting to @AMDCare @Roy_techhwood @GFXChipTweeter @CatalystMaker. I don't expect help throubleshooting from them, I'm only hoping for one of them to note it down and send the issue down to the people who can investigate it. AMDCare answered me the first time, but gave some copy/paste answer, which is to be expected. I just really wish the issue could be acknowledged, if the "known issue" section simply said "Rare loss of monitor signal" to let us know they're aware and working on it, that'd be great.

    For me, this issue makes upgrading to a new AMD card a big question mark. I prefer AMD to Nvidia, but since I've seen this issue being reported on different cards and brands, and since the issue is not acknowledged, I just don't know if upgrading to (for example) the new Vega cards would come with the same problem or not.
     

Share This Page