NVIDIA GeForce 517.48 WHQL driver download & Discussion

Discussion in 'Videocards - NVIDIA GeForce Drivers Section' started by valorex, Sep 27, 2022.

  1. schloppy

    schloppy Member Guru

    Messages:
    106
    Likes Received:
    1
    GPU:
    ASUS 3080 TUF V2
    With 517.48 I am experiencing random instability; desktop appears to have locked up and mouse cursor frozen but 5-10 seconds later the issue stops. (5950x CPU, 3080 GPU, 64GB RAM, Windows 11 22H2 build 22622.608)

    Event log at the time of problem shows this.

    'The description for Event ID 0 from source nvlddmkm cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.
    If the event originated on another computer, the display information had to be saved with the event.
    The following information was included with the event:
    \Device\Video3
    Restarting TDR occurred on GPUID:b00
    The message resource is present but the message was not found in the message table'

    There are 3 of these errors with 'Resetting TDR occurred' then 'Reset TDR occurred' and then 'Restarting TDR occurred'.
     
  2. Mineria

    Mineria Ancient Guru

    Messages:
    5,540
    Likes Received:
    701
    GPU:
    Asus RTX 3080 Ti
    Maybe they think that LOD and texture filtering should be optimized via game engine and not via the driver?

    Another thing with HQ is that it taxes some systems on the CPU, making it not worth it over Q.
    Anisotropic sample optimization is disabled when you set it to Q?

    As for FSR 2... maybe Nvidia wants us to use DLSS instead....

    Didn't AMD always have better filtering when it comes to rasterization btw.? I recall that some people claim that AMD GPU's have better image quality.
     
  3. aufkrawall2

    aufkrawall2 Ancient Guru

    Messages:
    4,361
    Likes Received:
    1,821
    GPU:
    7800 XT Hellhound
    It's mip bias texture LOD. The bias is requested by the application (and is negative with DLSS and FSR 2) and how well it is filtered is determined by the driver and TMU hardware.

    The CPU doesn't care at all about texture filtering, not the tiniest bit.

    It is off for both Quality and High Quality.

    Was just an example, shimmers the same with DLSS and texture filtering set to quality like FSR 2.
    So yeah, Nvidia unnecessarily gives users visibly worse texture filtering by default, especially with DLSS and FSR 2 that use negative LOD bias. And the Nvidia Vulkan driver doesn't support proper filtering for negative LOD bias not even with filtering set to HQ.

    AMD notoriously sucked at texture filtering until Vega, but since then things are different. Not related to other quality or color settings like dithering though. But yes, AMD also enables dithering by default for 8 bit. Also the Nvidia Linux driver does it, just Nvidia Windows devs think differently for whatever bad reason.
     
  4. XantaX

    XantaX Master Guru

    Messages:
    706
    Likes Received:
    316
    GPU:
    RTX 4070 TI 12G
    Hahah LOL the "game" Vega is a Gigabyte Radeon RX Vega LOL, but i think you knew that already haha :):):)
     

  5. DuncanZXCore

    DuncanZXCore Guest

    Messages:
    9
    Likes Received:
    2
    GPU:
    2070Super
    I got the same event log when call of duty mw random freezes and crashes to desktop.
     
  6. Astyanax

    Astyanax Ancient Guru

    Messages:
    17,017
    Likes Received:
    7,353
    GPU:
    GTX 1080ti
    fix your memory and cpu stability.
     
    Xtreme512 and PapaJohn like this.
  7. nndnnd81

    nndnnd81 Guest

    Messages:
    1
    Likes Received:
    0
    GPU:
    asus tuf rtx 3060ti
    Don't think it has anything to do with cpu or memory stability, this happened to me yesterday. Pc was just idling with steam store page opened, when i tried to resize Steam's window, screen has gone black for a second and driver crashed with "nvlddmkm cannot be found" error in event viewer. Out of bad habit after closing steam page i refreshed the desktop and the driver crashed again. Never happened before or during gaming until this occurrence. For now I'm back on previous driver.
     
  8. XantaX

    XantaX Master Guru

    Messages:
    706
    Likes Received:
    316
    GPU:
    RTX 4070 TI 12G
    A while ago i had exact the same issues, too much Oc Cpu and Memory ,Set it to default in the bios and the problems are solved.
    With too much OC windows can run normal but errors occur when playing demanding Games.
    Or you have faulty hardware.
     
  9. DuncanZXCore

    DuncanZXCore Guest

    Messages:
    9
    Likes Received:
    2
    GPU:
    2070Super
    When call of duty mw crash for me, it did had a black screen also. It was only call of duty mw crashing I notice other games were fine.

    My bios is on all default only thing I got on is XMP for my memory to run at its proper speed, I did had my 2070 super overclocked a bit at the time thought I could try get bit more out of it lol, notice there's no worth it, So its back to factory stock speeds so far its been fine no crashes, thanks.
     
  10. Hyppolyte

    Hyppolyte New Member

    Messages:
    5
    Likes Received:
    5
    GPU:
    MSi 2080 Ti
    @ManuelG

    Just installed 517.48 before 30mins. Tried to play some Mass Effect: Andromeda, and game crashed in 5mins. Tried again, crashed also in 2 minutes. Error was something like that:

    "DirectX Error: "Dx11Renderer::tryMap", failed with "DXGI_ERROR_DEVICE_REMOVED" (''The video card has been physically removed from the system, or a driver upgrade for the video card has occurred. ") This error is usually caused by the graphics driver crashing; try installing the latest drivers.Also make sure you have a supported graphics card with at least 512MB".

    My gpu is MSi 2080 Ti Lightning Z, and never had such problems until today as soon as i installed 517.48. Anyone else getting similar driver crashes in another game(s)?
    Thanks in advance.

     

  11. Astyanax

    Astyanax Ancient Guru

    Messages:
    17,017
    Likes Received:
    7,353
    GPU:
    GTX 1080ti
    except it does.
     
    PapaJohn likes this.
  12. PapaJohn

    PapaJohn Master Guru

    Messages:
    402
    Likes Received:
    136
    GPU:
    Asus 6700XT OC 12Gb
    3080 12gb here and to some of the above posts, I've played both Andromeda and MW for hours since installing this driver - both at 4k and no crashes. I agree with Astyanax, check stability on your system.
     
  13. XantaX

    XantaX Master Guru

    Messages:
    706
    Likes Received:
    316
    GPU:
    RTX 4070 TI 12G
    Try to disable XMP in the Bios ,XMP is as "safe" overclocking memory / cpu speed see what's happens then, if there are no errors in the game set cpu speed manual on xmp value speed and for memory too for memory raise the voltage a bit. from 1.3 to 1.4v some DDR brands need more voltage for example Corsair Memory, mine is Corsair set @1.45v and never had a system crash.

    You can try to shorten the bios reset with a jumper on the motherboard or removing the battery and in both cases reflash the bios latest version, B4 doing this take a picture first of the bios settings,you need it after reflashing the Bios...... or you can skip this all and install an earlier nVidia driver and see what happens then.....happy hunting ....;)
     
  14. Astyanax

    Astyanax Ancient Guru

    Messages:
    17,017
    Likes Received:
    7,353
    GPU:
    GTX 1080ti
    sometimes its as simple as bumping the command rate from 1 to 2.
     
    XantaX likes this.
  15. FR4GGL3

    FR4GGL3 Master Guru

    Messages:
    235
    Likes Received:
    125
    GPU:
    GB 4090 Gaming OC
    Some of these issues come from memory tuning. If I tune certain subtimings, most Tests gain Speed and seem to be Stable (even Tests like LinX or Karhu). But in my case my weak spot Test is Firestrike. For some reason in such cases it finishes and shows the scores but a few seconds later (without load) I will get a hard shutdown crash (also unknown Reason in the event Log). There is something going on with this "tuning" and going into low energy states in idle after load. I don't care anymore. Stability over speed for me. So no tuning attempts anymore.

    Don't blame the drivers in the first place. Look after Temperatures, Settings, Hardware defects and so on in the first place. If you can sort that out and it still crashes, you could look after drivers - but also after recent Windows Updates. I'm afraid it isn't a simple task to sort out such instability issues.
     
    XantaX likes this.

  16. Astyanax

    Astyanax Ancient Guru

    Messages:
    17,017
    Likes Received:
    7,353
    GPU:
    GTX 1080ti
    :cool: if theres a major driver side issue, we'll all be talking about it
     
  17. OnnA

    OnnA Ancient Guru

    Messages:
    17,852
    Likes Received:
    6,743
    GPU:
    TiTan RTX Ampere UV
    IMhO it's only for 4090 ;) (21.xx)
     
  18. dr_rus

    dr_rus Ancient Guru

    Messages:
    3,886
    Likes Received:
    1,015
    GPU:
    RTX 4090
    A new common driver will probably be released around A Plague Tale: Requiem's launch date (18th)
     
  19. OnnA

    OnnA Ancient Guru

    Messages:
    17,852
    Likes Received:
    6,743
    GPU:
    TiTan RTX Ampere UV
    -> https://babeltechreviews.com/geforce-517-48-driver-performance/

    Based on our previous results and findings, we recommend Ampere and Turing users update to the latest GeForce 517.48 driver.
    Its raw performance and smoothness or frametimes consistency level are overall on par with our previously recommended driver 516.94, using both NVIDIA GPU architectures.

    From a qualitative point of view, there are also important reasons that make it advisable to upgrade to version 517.48.
    These reasons include its higher level of driver bug fixes, the optimizations for the latest games, and support for the latest NVIDIA technologies.
     
  20. escape75

    escape75 Member Guru

    Messages:
    104
    Likes Received:
    39
    GPU:
    MSI RTX 3060 12G
    There's some weird TDR's I'm experiencing as well with 517.xx that I've never had with 472.xx for example.
    They only happen when the card is idle, and usually when the monitor is sleeping after inactivity,- PC is set to never shut off.

    I can fix it by changing to "Prefer max performance", and I've already tried MB BIOS update, GFX BIOS update, BIOS setting changes (XMP, etc).

    It seems like there's (maybe only on some chipsets ...) a power saving feature that kicks-in and causes a TDR in these new drivers ...
    I'm running RTX 3060 (GA106-302) and a MSI H510M PRO motherboard with an i5-11400F and Corsair LPX RAM and Corsair 650w PS.

    The system is actually a pre-built but I've replaced the RAM and Power Supply, they were both used on an H410M PRO previously
    and never had any issues, but I've tried another power supply and it's still causing the same issues ...

    I was going to double check all the pins on the CPU next, but if the system is stable when set to "Max Perf" I'm beginning to think it may be the drivers itself ...
     

Share This Page