(sorry for my bad english i'm french)
Sometimes when I launched GPU-Z/3DMark my computer freeze, I have to make a computer reset (with the reset button) after this, it's ok, the computer starts up normally.
I conclude with one thing : it freeze only with application get the SYSTEMINFO of the graphics cards.
is it possible to help me ?
Thank you !
I had a similar problem with my RX Vega 64 crossfire setup. 3dmark would freeze if I had msi afterburner running when I launched the benchmark, when systeminfo was gathering information. I fixed it by enabling stealth mode in rivatuner statics settings. The normal way msi afterburner was hooking the onscreen display was causing the crashes. You probably are experiencing the same thing I was going through with having gpu-z open at the same time as 3dmark. I am not sure how to help you except try and close gpu-z while running the benchmark.
Okay, "when I launched the benchmark, when systeminfo was gathering information" I also had this too.
Yesterday, I try with the old Driver (clean install with DDU - Crimson ReLive Edition 17.8.2)) and after open/close many times GPU-Z/3DMark I haven't freeze.
After, I retry with the most recent Driver : Crash.
I also remember that when launching GPU-Z sometimes I had a freeze.
You should also try and tell me
I install today the new drivers, I just launch 3DMark (No GPU-Z) and I had freeze, after 3 bench freeze again...
2 freezes today with no GPU-Z...
Yeah 3dmark and the scanning for system info is probably causing the problems with newer drivers. For me it was msi afterburner causing the issue. I also was crashing in games until I enable stealth mode for rivatuner statistics. So it seems like sometimes it crashes for you and other times it's fine. I really don't know what else to suggest except go back to a driver that worked fine with your graphics card and 3dmark.
I've had a vega liquid for about a week now and the machine has been an unstable mess since I got it. I've had a ton of freezes and instant-surprise-reboots, most of them happening while doing mundane desktop things. (ex, dragging a tool window in visual studio, opening a pdf, clicking something in firefox, etc..), some happening with "metricsware". I've not had any freeze/reboots while playing a game though.
I've found it is really easy to freeze the system by starting vulkan sdk cube sample and then repeat start-stopping gpu-z. This happens with oct6 driver as well as a few earlier drivers.
since I no longer use GPU-Z (or rarely) I do not freeze anymore.
Soon 2 weeks that the computer runs without any problems, Maybe your graphics card is failing?
I'm on the last Driver.
i just had a similar freeze tonight. got my vega 56 yesterday and its been running great until tonight. so not sure if its coincidence or actually related, but i just had a black screen + freeze + reboot after i loaded up gpu-z while playing pubg at the same time to check out temps with my vega 56. everything was running great since i installed the card yesterday before this incident. and it was quick. like the moment gpu-z started to launch (its splash screen) after i gave windows uac permission to load it, it froze.
i actually thought it was going to tdr as with my 580 i had similar problems, like loading up a youtube video while playing a game simultaneously, or loading up gpu-z would cause a tdr. but this didn't tdr, just rebooted. event viewer doesn't show anything useful. just a random warning un-schedule shutdown followed by a critical level shutdown. doesn't list a bsod or a tdr.
the reboot wasn't instant though. i heard the fan on my 56 spin up and down a few times for like 5-7 seconds before the reboot. do you think it could have been really gpu-z that caused it or something else?
evga g3 750 watt - connected with two dedicated 8 pin pci-express cables
crosshair vi hero
Did you ever figure this out? I am having the same issue and have been contemplating replacing the card under RMA. Mine is an RX Vega 64 air-cooled reference card. For me the game that always causes the crash is FFXV. GPU-Z shows HBM temps exceeding 95c in some instances. I suspect either the thermal paste from factory is inadequate on the HBM or the cooler is just not good enough at removing the heat at stock fan speed. I have been comparing this to a Sapphire Nitro+ RX Vega 64 and that card's HBM temps rarely go above 70c.
I figured out that it was the Seasonic Prime Titanium 1000w not being able to handle transient voltage spikes from Vega 64. After swapping to a Corsair HX1200 I no longer have any OCP shutdowns.