Stutters in game

[Niekao]

Help, something went wrong all of the sudden. I dont know what happend this was fine yesterday and now I can barely play any game at all.

[Nord]

This isn't stutter, this is artifacting.

 

Most likely cause is that your GPU broke. Or rather it's VRAM or cooling solution.
If you still got warranty on it - RMA it.

 

If you want to verify that it's the GPU, though im 99% certain it is:
Try a different PCI slot on the motherboard.
Try a different PSU.
Try a different GPU (if you got an old one)
Clean-install drivers for that 0.1%  chance it's that.
Check GPU temperature with afterburner, hwmonitor or hwinfo. If it's above 82°C, the card is overheating which (maybe) could be fixed with new thermal paste.

Make sure the card is seated properly into the motherboard and all cables are connected properly. Also check if theFans spin under load.

 

If you just absolutely need the GPU right now you can maybe get it to somewhat work still, depending on what exactly broke, by uing afterburner to downclock the Memory by as low as it goes and also reduce core by -100 while lowering the powertarget by - 10 or 20%.
Obviously tanks performance but might be enough to make it stable-ish, still just a temporary solution.

[kelvinhall05]

17 minutes ago, Nord said:

This isn't stutter, this is artifacting.

 

Most likely cause is that your GPU broke. Or rather it's VRAM or cooling solution.
If you still got warranty on it - RMA it.

 

If you want to verify that it's the GPU, though im 99% certain it is:
Try a different PCI slot on the motherboard.
Try a different PSU.
Try a different GPU (if you got an old one)
Clean-install drivers for that 0.1%  chance it's that.
Check GPU temperature with afterburner, hwmonitor or hwinfo. If it's above 82°C, the card is overheating which (maybe) could be fixed with new thermal paste.

Make sure the card is seated properly into the motherboard and all cables are connected properly. Also check if theFans spin under load.

 

If you just absolutely need the GPU right now you can maybe get it to somewhat work still, depending on what exactly broke, by uing afterburner to downclock the Memory by as low as it goes and also reduce core by -100 while lowering the powertarget by - 10 or 20%.
Obviously tanks performance but might be enough to make it stable-ish, still just a temporary solution.

That does *not* look like artifacting caused by a GPU. His hud is fine, his player model is fine, the ground close to him is fine, etc.

 

 

OP, try reinstalling the game and/or Windows.

[Nord]

A GPU has not only 1 but several VRAM modules/chips (w/e they are called properly) on the PCB, depending on how much VRAM the card has in total it can be that - made up example incoming - out of the 4 modules on the card, 3 are fine, 1 lost it's thermal pad. Coincidentally the one that lost the thermal pad is fully loaded with all the textures of the surrounding areas while the others, or just one of them - does not matter - have the remaining textures in it, results in what you are seeing.
I’ve actually seen similar happen on older nvidia cards. 

Anyway, even if the character model is fine, the fact that it gets blacked out or “overdrawn” by artefacts is still there but I have to agree, it is weird that the HUD has no issues and is visible all the time.

 

Though a windows reinstall suggest seems a bit excessive to me. Even if I would not believe it to be a faulty GPU, the first step would still be to check on temps, proper card seating, see if other games are affected or if it's only AC odysee and re-install the driver (like you suggested) and possibly rollback to a previous version first. Only after each of these have been ruled out a OS reinstall becomes a reasonable approach as all of the above maybe take 20min while an OS reinstall takes around half a day, including installing & setting up everything again.

 

 

[Niekao]

1 minute ago, Nord said:

A GPU has not only 1 but several VRAM modules/chips (w/e they are called properly) on the PCB, depending on how much VRAM the card has in total it can be that - made up example incoming - out of the 4 modules on the card, 3 are fine, 1 lost it's thermal pad. Coincidentally the one that lost the thermal pad is fully loaded with all the textures of the surrounding areas while the others, or just one of them - does not matter - have the remaining textures in it, results in what you are seeing.
I’ve actually seen similar happen on older nvidia cards. 

Anyway, even if the character model is fine, the fact that it gets blacked out or “overdrawn” by artefacts is still there but I have to agree, it is weird that the HUD has no issues and is visible all the time.

 

Though a windows reinstall suggest seems a bit excessive to me. Even if I would not believe it to be a faulty GPU, the first step would still be to check on temps, see if other games are affected or if it's only AC odysee and re-install the driver (like you suggested) and possibly rollback to a previous version first. Only after each of these have been ruled out a OS reinstall becomes a reasonable approach.

 

 

Temps dont go over 68 celsius. I tried other games, Payday 2 fine, skyrim fine, anno 2205 not fine, rainbow six siege not fine. 

[Nord]

What GPU do you have?

Exact model, example: ASUS Strix 1060.

[Epimetheus]

Maybe try to uninstall and reinstall the drivers?

[Niekao]

Just now, Nord said:

What GPU do you have?

Exact model, example: ASUS Strix 1060.

Gigabyte RTX 2080ti GAMING OC 11 G

[Nord]

I’m fairly certain that card can run payday 2 and skyrim (unless you modded it to hell and beyond) w/o even going to full clock speed, which would explain why these two games are unaffected and GPU heavy ones like AC, anno and Siege (where I would suppose you got uncapped FPS) are not fine. 

 

I’ll stick to my previous statement of faulty VRAM.
A driver reinstall can not hurt though just to be sure. Also if you got the card overclocked remove the OC for now, otherwise refair to my previous posts on how to troubleshoot or rather nail the cause down.
 

FYI. It could also be possible, though very unlikely, that it's the PCI-E slot or PSU’s fault, so if you happen to have a spare PC, PSU or motherboard lying around you can double check there, though those are the last thing I would check on.

[Niekao]

4 minutes ago, Nord said:

I’m fairly certain that card can run payday 2 and skyrim (unless you modded it to hell and beyond) w/o even going to full clock speed, which would explain why these two games are unaffected and GPU heavy ones like AC, anno and Siege (where I would suppose you got uncapped FPS) are not fine. 

 

I’ll stick to my previous statement of faulty VRAM.
A driver reinstall can not hurt though just to be sure. Also if you got the card overclocked remove the OC for now, otherwise refair to my previous posts on how to troubleshoot or rather nail the cause down.
 

FYI. It could also be possible, though very unlikely, that it's the PCI-E slot or PSU’s fault, so if you happen to have a spare PC, PSU or motherboard lying around you can double check there, though those are the last thing I would check on.

But how could this happen overnight? Yesterday everything was fine...

[Nord]

My PSU once broke while watching youtube, after a 30min “pause” straight after 5hours of gaming.
The VRAM of an old EVGA 570 broke entirely while playing a game that barely made the card work at half power. (like instant BSOD if you opened a 3d application).
Another PSU broke overnight, worked day before, completely dead the next day 4months after purchase.
I even had the right speaker of a headset break overnight once.
Once my car would simply not start because the battery utterly died overnight, inside the garage, after being driven the entire week, on a warm day.

 

Bottom line is, stuff just breaks and electronics can do so at any given point for a lot of different reasons, some knowen some not.