Haak

A new HW info after a 3d mark and some gaming. Also the stress test with all heaters on max on did not cause a crash so far.

Heaven will not run for some reason after i swapped the gpu back, but i am downloading 3d mark to stresstest

Not been able to find the rails in hwinfo but here is a link to the product https://www.gigabyte.com/Power-Supply/GP-AP850GM#kf If that helps There is quite alot on the circuit (heaters etc) But i can try to turn everything on while running a stress test. That should act as a worst case scenario for circuit draw.

Cant swap into the old system withouth bringing the PSU along as the old system runs a 500w psu atm. Also if it was a power problem. Shouldnt it be rather easy to reproduce during stress tests when the draw is high? Downloading HWinfo to check the rails

I have an old GPU from my old build that i threw in for trubble shooting . But the build with the 3080 is a completely new build with all new components.

Thanks for your swift reply I will try this I can also not that it was a fresh build, if that matters.

I can also add that i cant cause a crash to happend making it hell to get fixed or send in as rma. The tech shops may not find a problem and then send it back while charging me for for testing a card they cant find a problem on... So i really hope you have some good ideas!

Finally got my hands on a 3080. However i am having problems with random crashes in game but also from desktop. When the crashes occur the screen freezes and then often goes black sometimes also showing no input detected. If sound is played when the crash occures its often stuck on 1 note but sometimes the sound comes back for sec at the time. However the PC have never managed to resolve the crash withouth a restart. Then after a while the pc will try to restart or else i have to force a shutdown. Can also ad that when it freezes the picture sometimes get grainy = you cant read text in word for instance. 1 time the GPU fans also boosted to 100% after the crash. I have the newest drivers through ge force experience. A fresh windows with very little on it. Tried both DP and HDMP, Tested stock and with reduced clock frequency. Have 2 seperate power cables from the psu to the GPU (not 2 connectors on the same psu output). Still crashes. The card can complete benchmarks without crashing, but then crash later during low demanding tasks when its around 40c I tried swapping in my old GPU and the problems have stopped so far. Are there any known problems with the 3080 atm? Could i expect this to be resolved through a driver update or is it likely hardware related? My specs are x570 tomahawk 3900xt 3080 tuf non oc hyper X 3200 4x16gb gigabyte aorus ap850gm

Hdmi did not resolve the problem. Still crashes. Was also paying attention to the clocks and the max clock was about 1860. The card is now under clocked however the problem occurs at base clock and under clock so that does not seem to help. If anything its just more confusing as it more or less eliminates the high boost clock problem.

The card is powered by 2 separate 8 pins (2 separate from the psu). I have not tried a hdmi yet, i have seen 3070 owners having a lot of trouble with that. Going to try that next, just need to find one

I was thinking about this myself. But as you said this problem was "supposed" to be fixed by a drive update some time ago. I do however have the newest driver available (at least when using geforce experience.) This problem was though to occur when the cards boosted to high, but my crashes have also occurred while the card was under clocked. I followed guides on how to reduce frequency curves to reduce power draw at minimal performance loss (there are several videos out on this). I dont know enough about the technical stuff to understand if this also reduces the max boost frequency, or if the boosting algorithm still could push it into problematic territory. Is there anything i can do to solve something like this? i dont mind under clocking as im not running the most demanding games atm. But as i said this have not solved the problem so far.

If was of course to good to be true. The problem still persists. This time i managed to snatch a picture of Powerpoint showing how the resolution gets all distorted when the crash happens. This time the GPU also started to ramp up its fans to 100% after the crash and it failed to connect to the monitor so i had to force a shutdown. With the fans on the GPU ramping up to 100% after the crash can i assume that the GPU is the one to blame?

ill do that. a funny thing though is that stress test did not reliably manage to cause the crashes to happend. have been testing with cinbenbench r20, 3d mark and heaven. And only 1 time did heaven fail when i was testing undervolting performance.

It just freezes or goes black, often due to not detecting the monitor. However when its not detecting the monitor i know its also frozen as i can hear that its stuck if i played some sort of sound while the crash happend. Update. I uninstalled asus armory crate and additional software thinking it would be problematic with more than 1 of these rbg/package softwares ( i only used it to control the little rgb on the gpu). So far the crashes seems to have stopped. been running for 30 min and counting with 15 min in game. Are these programs known to cause issues? or am i grasping at straws?

Built a new build with the following specs x570 tomahawk 3900xt 3080 tuf non oc hyper X 4x16gb 3200 I had some problems getting the computer up and running, crashing when trying to install windows etc. But when i got it working it thought i was ok, and i was for a time. Then i suddenly started having crashes while gaming. Nothing hard to run, only WOW. Have had hwmonitor running in the background and the temps seems fine, never going over 70 and most of the time staying in the 40-60 range. I have turned off xmp, still crashes i have returned the gpu from an under volt state to normal clocks, still crashes Now it even crashed during browsing and even at idle with the temps in the 30s. If anything the crashes seems to occur more often and at lighter loads. Have tried both GPU bios, still crashes When the crash occur the screen freezes and goes black. The monitor can also show no signal. The sound (if any sound is played)is stuck on one note. Some of the times it tries to restart, with varying degrees of success. As far as programs go i have very little on the computer. Windows, word, Dragon center and asus armory, cinebench, heaven, hwmonitor and a few blizz games. Can also note that i have updated windows and have got the newest mobo bios and GPU drivers installed At this point i am at a loss and would really appreciate any help you can provide. During the time it took me to write this it has crashed 2 times from a clean desktop so the problems seems to be escalating.