GeForce 466.47 WHQL driver download & discussion

Discussion in 'Videocards - NVIDIA GeForce Drivers Section' started by valorex, May 18, 2021.

  1. kman

    kman Master Guru

    Messages:
    339
    Likes Received:
    22
    GPU:
    1080ti Strix
    Already off.The 1.1k spikes still happen.

    tty8k also mentioned trying this on an earlier comment with same results.
     
    Last edited: May 29, 2021
  2. BlackScout

    BlackScout Active Member

    Messages:
    62
    Likes Received:
    21
    GPU:
    GTX 1660 6GB
    Insider on 470.14
    [​IMG]

    Here's a repro step: Leave whatever you have open...
    Open GTA V, move your mouse as much as you can, FPS will tank and stutter heavily.
     
  3. EdKiefer

    EdKiefer Ancient Guru

    Messages:
    2,697
    Likes Received:
    283
    GPU:
    MSI 970 Gaming 4G
    You're not supposed to test LatencyMon with any loaded application, should be run with just OS running. This is per Dev.

    "Should I run my audio software while LatencyMon is testing my system ?
    In general, no. In particular, the interrupt to process latency that the software measures already simulates the workings of an entire audio process. You should run most tests without running any other software in the background. Running your audio software together with LatencyMon only makes sense if you wish to measure hard pagefaults of the audio process. "

    https://www.resplendence.com/latencymon_faq
     
    endbase and Cave Waverider like this.
  4. kman

    kman Master Guru

    Messages:
    339
    Likes Received:
    22
    GPU:
    1080ti Strix
    Yet its still detecting dpc latency spikes where previous drivers didn't.It may not be meant for a certain use but it can still be usefull to point you to a possible issue.
     

  5. EdKiefer

    EdKiefer Ancient Guru

    Messages:
    2,697
    Likes Received:
    283
    GPU:
    MSI 970 Gaming 4G
    Your still running the tool out of spec of what it was intended for, the text results will be invalid IMO.

    That said you can use it to test various things running which add latency as long as you change one thing at a time and the exact value will still be inflated because of the internal load of latencyMon uses.
    It would be nice if you had control of this internal load but then it may be needed for reading results (I have no idea what is possible here).
     
    endbase likes this.
  6. kman

    kman Master Guru

    Messages:
    339
    Likes Received:
    22
    GPU:
    1080ti Strix
    Does it matter if its not exactly used as intended if its still picking something up on 1 nvidia driver but not the other?Alot of ppl suggest latencymon to pick up issues like stutters and such so it has it uses either way intended or not.

    "That said you can use it to test various things running which add latency as long as you change one thing at a time and the exact value will still be inflated because of the internal load of latencyMon uses."

    I mean.1 driver doesn't trigger latencymon and the other does (both using the same testing scenario (unless you use 10bit mode on the monitor on the driver that triggers latencymon) For some reason switching to 10bit mode on the monitor gives you MUCH lower dpc latency spikes in general,that's all I need to know.Something has changed between drivers.
     
    Last edited: May 30, 2021
  7. endbase

    endbase Maha Guru

    Messages:
    1,015
    Likes Received:
    154
    GPU:
    TUF OC RTX 3080
    Man you are becoming the Don Quichot of Latency I agree with what @EdKiefer said about the proper use of LatencyMon
     
  8. EdKiefer

    EdKiefer Ancient Guru

    Messages:
    2,697
    Likes Received:
    283
    GPU:
    MSI 970 Gaming 4G
    What I am getting at is that while you might show certain drivers produce more latency than other (if properly tested), the magnitude of the value probably is not representative if your running other apps at the time of the test.
    There are a lot of people who use tools the wrong way so that by itself is not a justification of its the right way.

    Have you tried this one, maybe the results might be the same? I am not trying to argue just stating a point of view from the DEV tool.
    https://forums.guru3d.com/threads/simple-way-to-trace-dpcs-and-isrs.423884/
     
  9. Astyanax

    Astyanax Ancient Guru

    Messages:
    11,327
    Likes Received:
    4,251
    GPU:
    GTX 1080ti
    while you can't use it to prove an issue exists while the system is busy, you can prove that the numbers are different or higher than they were before.

    its just preferred that an actual ETL is made for such cases because it includes graphable sampling in the performance analyzer that can be pin pointed down to what thread and what core it was done on.
     
  10. SmokingPistol

    SmokingPistol Member

    Messages:
    39
    Likes Received:
    5
    GPU:
    RTX 3090
    I think that you are just giving useless imput altogether, though.

    This "latency" discussion is very informative and interesting and I for one appreciates the effort/feedback that tty8k, kman, Astyanax are sharing.

    Sometimes, it is best to be silent if you don't have nothing of value to say, you know?!
     

  11. endbase

    endbase Maha Guru

    Messages:
    1,015
    Likes Received:
    154
    GPU:
    TUF OC RTX 3080
    Thanks to freedom of speech I appreciate your input I have been silent on the object TBH but this babling aound hunting ghosts every day was getting on my nerves but it's all out now because I ventilated my thoughts in my post I even hesitaded to make it less harsh but no I stick with it! once again thx for your input

    ps Kman is not using latencymon as intended per DEV! (look a few posts up if you want)
     
    Last edited: May 31, 2021
  12. tty8k

    tty8k Master Guru

    Messages:
    549
    Likes Received:
    141
    GPU:
    Ati 5850
    @endbase
    This babbling around getting on your nerves might actually help nvidia solve some of the issues we have and also help us determine what we can do on a user level to improve the situation.
    While findings and opinions might differ from user to user (different systems) they're all welcome as long as it helps us build a bigger picture.

    I used latencymon on a specific trigger (webbrowser/youtube open) and a very short time span specifically to avoid other latency events.

    You can help with your own findings rather than arguing for the sake of posting.
     
    eGGroLLiO likes this.
  13. endbase

    endbase Maha Guru

    Messages:
    1,015
    Likes Received:
    154
    GPU:
    TUF OC RTX 3080
    I allready did some posts back tbh I posted 2 out comes of latencymon so it's not that I don't want to contribute but in my perception the program is not designed to run with open different programs with it. If you guys think different allright with me but I stand behind my argument tbh and that is that in the disclosure of latencymon is clearly mentioned that the program is designed to run idle state of the OS as the developer clearly mentions (Should I run my audio software while LatencyMon is testing my system ?
    In general, no. In particular, the interrupt to process latency that the software measures already simulates the workings of an entire audio process. You should run most tests without running any other software in the background. Running your audio software together with LatencyMon only makes sense if you wish to measure hard pagefaults of the audio process.) and second the program is used to find a pin point problem where the latency is coming from. Further I think it is the end user responsebility to find the exact cause on his system and not a compagny as NVidia because it should be their driver related thing. As stated before by different members it is caused on the user's system and should be examined on the end user end IMO
     
    Last edited: May 31, 2021
  14. Astyanax

    Astyanax Ancient Guru

    Messages:
    11,327
    Likes Received:
    4,251
    GPU:
    GTX 1080ti
    i semi agree, but theres a point where the user can't go further because of things like the nvidia symbols being unavailable publicly.


    Hyperthreaded CPU's
    Hyperthreaded/Multicore cpu's with C states enabled
    Certain Motherboards with digital/adaptive EPU managers (some can be swapped back to analog/static with an available utility to resolve this)
    Faulty Monitors / Cables
    Faulty storage
    Faulty IO Controller (SB/NB/PCIEonCPU) (not necessarily all slots/lanes are affected - seen it with a P6X58D)
    Unusual error correction on the bus that could be any of the devices that make up the chain.

    Outdated Realtek audio and lan drivers
    Bad sata driver
    Windows itself
    A certain Kingston A series SSD (yep, im going there :D)
    PS4 controllers (these are notorious for DPC in the hid driver)


    It would be safe to assume that only a couple of those listed have any possibility of being interactive with the system when a depth format changes, i've bolded those.

    atleast one of the affected users has a 6700K, so ReBAR can be ruled out as a factor.

    There is the chance of some sort of bad interaction or errors on the pcie bus, but i don't have any personal history with 6700k's experiencing that.
    Motherboard configuration of the PCIE bus, features enabled, among other things, are a possibility.

    Hey, did we all forget nvidia enabled gpu passthrough on recent drivers? has anyone tested disabling Intel VTd/AMD IOMMU?

    did you happen to make a trace etl?
     
  15. endbase

    endbase Maha Guru

    Messages:
    1,015
    Likes Received:
    154
    GPU:
    TUF OC RTX 3080
    I can dig that ofcourse! there are some exceptions to what end users are capable of them self drivers and code wise.
     
    Last edited: Jun 1, 2021

  16. tty8k

    tty8k Master Guru

    Messages:
    549
    Likes Received:
    141
    GPU:
    Ati 5850
    Nope, can do at some point if needed.

    Yesterday I tested a friend's system (10700k - 1080 card) and it doesn't have the color depth issue.
     
  17. kman

    kman Master Guru

    Messages:
    339
    Likes Received:
    22
    GPU:
    1080ti Strix
    Was he on recent driver when you tested?
    Remember certain drivers like 460.89/457.51 don't trigger the dpc latency spike.
     
    Last edited: Jun 1, 2021
  18. tty8k

    tty8k Master Guru

    Messages:
    549
    Likes Received:
    141
    GPU:
    Ati 5850
    He is on 466.11 which has the bug on my system.
    Maybe is something that affects RTX cards not the 1k series ...
     
  19. kman

    kman Master Guru

    Messages:
    339
    Likes Received:
    22
    GPU:
    1080ti Strix
    If I remember correctly neil has a 2k card and has been vocal about having similar spikes in certain drivers after 457.51
     
  20. Dragondale13

    Dragondale13 Maha Guru

    Messages:
    1,484
    Likes Received:
    212
    GPU:
    GTX 1070 AMP! • H75
    There was a user over at Geforce forums a few weeks ago describing what you guys are discussing.
    Basically high latency in any driver past 457 and possibly a mem leak in 465, 466, which doesn't occur with 462.
    Maybe that's why Nvidia Studio, Quadro & Tesla are still 462.
    I get a very slight pop on my phones whenever I'm selecting options in the menu of GTA 5, which started with 466.
    So could very well be the driver.
     

Share This Page