Therefore, the counter test with another graphics card in your PC or this graphics card in another PC would have made sense. But it looks like the VRAM or it´s cooling (maybe booth) is indeed defective.
Tomorrow, i'm sending back this PC to the store where I have bought it
They'll run the tests and i'll see.
hotspot is at 75°C in game, ran a stress test in furmark but memory temp as reported in GPU-Z was quite high at 105°C for a 5 min stress test. No artifact detected in Furmark though.
They'll run the tests and i'll see.
Post automatically merged:
Normal ram and cpu is fine.
hotspot is at 75°C in game, ran a stress test in furmark but memory temp as reported in GPU-Z was quite high at 105°C for a 5 min stress test. No artifact detected in Furmark though.
I think this would be the best bet. Something is definitely not working correctly. The stretched geometry can happen in games if the installation itself contains corrupted data. Normally, a simple verify / repair of the game itself will fix that. But for that to have started appearing now on top of everything else means that you may, in fact, have a defunct card. It happens.
For something like this, yes. What we'd want to do is put his existing card in a completely different system to see if it created the same artifacting there, as well.
If so, we could see that something is probably borked with that card itself, but not necessarily what.
If not, we would be able to see that it's definitely something in the other system causing it. While we could rule out the card being the culprit, we'd still have a job of work to do figuring out what is causing the problem.
105* should probably not be possible -- the GPU manufacturers should have the cards automatically throttle down at around 85*C - 95*C. I don't think I've ever heard of a triple-digit throttle point. (It may actually be the case, though. Normal throttling now is in the 90*C range, mostly. So 105*C might actually be a thing. Still...yikes. That's getting really, awfully hot.)
105c is quite high, im guessing its not making proper contact with the heatsink but hand it in instead. 3080s memory can handle 110c i think before it starts too downclock. Opening the gpu would void warranty so its better too let them handle it.
For memory chips its quite a bit higher. The core would fry at that temp tho.
I admit, as an irl corpo, I didn't read all of the above, but might be able to help if someone is still troubled and could provide a summary.
Something that kills many builds in high-load scenarios is the PSU headroom.
Especially the high-end 30-series cards have short bursts of requiring way more power than officially stated. Here is a video from GamersNexus on the topic:
I had this in mind when I acquired my AX1600i in 2020. I didn't even think that I'd wind up with a 3090 that has the base TDP of 400+ watts, and hits 450+ OC'd. The cure is more PSU headroom, or perhaps lowering GPU's power draw, e.g. via MSI afterburner.
Else, overclocking RAM creates some ground for apps to die (no reboot/shutdown). Add summer weather to it, and that is the recipe for app crashes with sprinkles on it. Cure: use base freq and see if the issue persists.
Most folks don't want to deal with those aspects of PC gaming. Sending cares and hugs to you.
Something that kills many builds in high-load scenarios is the PSU headroom.
Especially the high-end 30-series cards have short bursts of requiring way more power than officially stated. Here is a video from GamersNexus on the topic:
I had this in mind when I acquired my AX1600i in 2020. I didn't even think that I'd wind up with a 3090 that has the base TDP of 400+ watts, and hits 450+ OC'd. The cure is more PSU headroom, or perhaps lowering GPU's power draw, e.g. via MSI afterburner.
Else, overclocking RAM creates some ground for apps to die (no reboot/shutdown). Add summer weather to it, and that is the recipe for app crashes with sprinkles on it. Cure: use base freq and see if the issue persists.
Most folks don't want to deal with those aspects of PC gaming. Sending cares and hugs to you.
Last edited:
Yea sadly its quite the problem with modern cards. The recommended is allways just on average. Doesnt say that anywhere but thats how it is. Also people get told stuff thats not true, higher watt doesnt mean your stuff will break. It just makes the PSU work at a lower load. It will not deliver more power unless your part asks for it! Also people cheap out on PSUs since they think its all the same, it isent! Buy as good as you can afford! Brands and/or ratings are very important! Buying a cheap one might cost x2 in broken hardware so dont be stupid please!
Well, I'm sure the shop will test out the GPU on another setup.
Don't have one myself to test it anyway.
Maybe even the GPU is powered by only one PCIe cable, don't know, tried to check but the case is so tightly packed that I couldn't verify.
Anyway since i'm sending it tomorrow i'll tell them to check.
I've send it like a month ago to the shop for deep cleaning because of somewhat high CPU temp (around 80°C in game) and perhaps they didn't plugged back the gpu correctly.
Don't have one myself to test it anyway.
Maybe even the GPU is powered by only one PCIe cable, don't know, tried to check but the case is so tightly packed that I couldn't verify.
Anyway since i'm sending it tomorrow i'll tell them to check.
I've send it like a month ago to the shop for deep cleaning because of somewhat high CPU temp (around 80°C in game) and perhaps they didn't plugged back the gpu correctly.
Should be able too se that easily just by opening the side panel unless its a mess off wires and stuff? 80c ingame aint that bad tbh, this game loads up the cpu more then any other game i have atleast but a good cleaning every now and then helps.
Great that it worked out for you. Did you get specifications on what was wrong?