Search results

Here is my LatencyMon result while playing Cyberpunk 2077. It did not change after 5 or more minutes btw.

I hope the 8-pin splits to two separate tiers on the power supply?

Unfortunately i can not say anything about that power supply. There are no really convincing tests to be found. I have reservations on this.

It is ok. You have some other problem, i guess it is your power supply.

Sounds like a problem regarding power surge. What GPU and power supply, how did you connect it?

Exactly. Just let it run about 5 minutes not doing anything. Then press stop.

There have to be done 2 tests. Idle (a few minutes after boot), and load (your game or app) :)

Looks good to me. Even though i want to know why the CLASSPNP.SYS had a high DPC. However, more important would be the results when you are running GPU intensive tasks.

That would be nice, but my DPC goes up to 300 idle and is stable at around 1000 under load. Besides that, i fully agree :cool: BTW: you may have noticed ntoskrnl, but for me it is wdf01000.sys which spike every now and then. Although it doesn't produce issues, so it is normal.

I respect the effort that has gone into this thread.

Present that system and its problems to me and i will try to fix it. The thing is, i can't do much without getting my hands on a system. I don't remember many times in my life where i couldn't find a solution to a PC problem. Even if it costs money. But it depends on a lot of details.

I hoped you would not ask, because i didn't think it is important, and already guessed. There are only a few guys here who actually had success at the topic, by downgrading their NVIDIA driver. I am at the same camp. I have an RTX 30XX and a recent PC setup therefore. That camp has a specific...

Did you check msinfo32 for any IRQ conflicts? Realtek NIC should not introduce problems if it not shares the same interrupt with another device, just to my thinking. But if it works for you using WiFi, that's great :cool:

Just out of curiosity: what chip / brand is your onboard NIC? And do you have any other PCIe devices attached besides your graphics card?

Interesting. Never try this before, but now i set my buffer size to 128 (which is the lowest value my interface offers), and again, still no issues. The DPC value of nvlddmkm.sys keeps around the same as before, which is pretty cool!

Here is my 5 min idle screenshot with driver version 512.95 With 516.94, the DPC time for nvlddmk.sys is almost double. While using my audio app, i still get DPC times around 1000, but no audio issues. Edit: i have to mention that i'm running my audio at 96 Khz with a buffer size of 512. This is...

Anyone having issues that are not related to the NVIDIA driver version, already tried these: https://www.xbitlabs.com/how-to-get-better-latency-in-windows/ It is about disabling: - HPET - Synthetic Timers - Dynamic Tick I've not been aware of these, and my issues are already gone by downgrading...

:)

Not many views at the NVIDIA forum so far. https://forums.developer.nvidia.com/t/driver-version-516-94-introduces-high-dpc-latency-and-serious-audio-problems/238290/1 Edit: a bot has removed the topic for reviewing by an NVIDIA personal. That's great

It is a Motu Traveler MK3 firewire interface. But i had the same issue with ASIO4All and an USB sound card. With the 512.95 driver all problems are gone and it performs like a champ again.