Crashing while gaming

[wookiee87]

Event ID Error 4101 persists.
Display driver nvlddmkm stopped responding and has successfully recovered.

which monitering software do you have running?, too many can cause problems too

i typically run msi afterbuner all the time and when i want to check cpu temps i open hwinfo, i dont have 2-3 programs running as it can cause issues especially with amd cpu's making them run super hot.

just out of intrest what are temps on the cpu and gpu when you get the error?

 

[Joxeon]

Try running -100 on the GPU core and -200 on the VRAM.

 

[nicklog]

Try running -100 on the GPU core and -200 on the VRAM.

I swear this is what I did before i read your comment and I think it's fixed (no errors so far)

 

[nicklog]

which monitering software do you have running?, too many can cause problems too

i typically run msi afterbuner all the time and when i want to check cpu temps i open hwinfo, i dont have 2-3 programs running as it can cause issues especially with amd cpu's making them run super hot.

just out of intrest what are temps on the cpu and gpu when you get the error?

me too just afterburner, temps are WAY normal this is what makes it weird could be like 60 average on both

 

[wookiee87]

me too just afterburner, temps are WAY normal this is what makes it weird could be like 60 average on both

anyway to check psu voltages when you are in game, download a program called hwifo 64 and it will give you all the info you need including psu voltages under load

scroll down to the part i've just posted and check the lowest reading, if it is too low then your psu may be on the fritz, if the error persists

 

[Joxeon]

Try running -100 on the GPU core and -200 on the VRAM.

If it is indeed stable then try setting the core back to normal but keep -200 on the VRAM and vice-versa.

If I overclock my card to far I get the same crash error so if that's what's happening at stock then it could be a dodgy card, could also be the psu but usually that causes a reboot.

 

[nicklog]

If it is indeed stable then try setting the core back to normal but keep -200 on the VRAM and vice-versa.

If I overclock my card to far I get the same crash error so if that's what's happening at stock then it could be a dodgy card, could also be the psu but usually that causes a hard reset.

Look, on the games that I keep receiving the error, I could never even go to the opening screen when I tried to do even the tiniest GPU OC. So I kept it normal in order to get in game , I never thought it would need underclock ..wow
So i returned them to default and I still got the error, went my clock -100 and kept VRAM default and it runs perfectly for now

 

[wookiee87]

Look, on the games that I keep receiving the error, I could never even go to the opening screen when I tried to do even the tiniest GPU OC. So I kept it normal in order to get in game , I never thought it would need underclock ..wow
So i returned them to default and I still got the error, went my clock -100 and kept VRAM default and it runs perfectly for now

in that case the card is a bit of a wet lemon, if it works fine with a underclock on the core then you could offset the slight drop with a memory oc, but again if you then get the same problem occoring with a memory overclock then i'd look at rma'ing the card as it's potentually faulty as it wont run at stock speed, bad news given the gpu shortage but you may get lucky and get a replacment as it should be under warranty

 

[Joxeon]

Look, on the games that I keep receiving the error, I could never even go to the opening screen when I tried to do even the tiniest GPU OC. So I kept it normal in order to get in game , I never thought it would need underclock ..wow
So i returned them to default and I still got the error, went my clock -100 and kept VRAM default and it runs perfectly for now

Will probably need to RMA at some point although if it's going to be long turn around time I'd live with for now until stocks return to normal.

 

[nicklog]

@wookiee87 @Joxeon
I see.. although first of all it's only 100mhz is that, that big of an issue u reckon?
and also unfortunately I bought the card from an auction site I'm not the original buyer so... Not sure how easy will be to RMA

 

[wookiee87]

@wookiee87 @Joxeon
I see.. although first of all it's only 100mhz is that, that big of an issue u reckon?
and also unfortunately I bought the card from **** I'm not the original buyer so... Not sure how easy will be to RMA

the 100mhz offset wont really drop performance by a lot (loose 5fps maybe), but buying from such sites you should always get the origianl invoice just in case something like this happens, that way you are covered, contact the person who you bought from and ask to have the origianl invoice.

 

[nicklog]

the 100mhz offset wont really drop performance by a lot, but buying from such sites you should always get the origianl invoice just in case something like this happens, that way you are covered, contact the person who you bought from and ask to have the origianl invoice, btw mentioning auction sites is not allowed so change that part quick before you get suspended

Thanks for that. I totally forgot.

What I mean is I'm getting the expected performance of a 3080 anyway I don't think I'll see such a bad difference, ofcourse it's a bummer I bought every single piece brand new and built it couple weeks ago and I want everyrhing to be perfect as all of us. But with this situation in the market now.. I don't think I'm gonna change something .

Thanks for the help to all of you guys honestly

 

[Joxeon]

Yeah see if the seller will send you the original invoice if not you have a couple of options, either live with and maybe try to find the point where it crashes and set it just below or you could start a return procedure at the auction site where you brought it from which may encourage the seller to cooperate and send you the invoice or email you a copy etc.

 

[SkeeterUK]

Unless the card can be rma with manufacturer by serial number only which i think some do dont they like gigabyte?

Maybe worth stress testing the gpu with the underclock for a few loops then gradually increase to stock untill u get crashes then u can narrow the underclock down maybe by 50hz nearer to stock than -100hz?

Could always check if a bios was available for the gpu or flashable to a diff one which might solve the issue? Tho im not sure of this idea just somit i read ppl do sometimes on the forum.

 

[pc-guy]

Some 3080s run hot due to poor pads. That maybe your solution. Try to see if bringing more airflow to the card help with stability.


As always, building a system you need to incrementally test each component for stability to ensure your bits work in the first place individually.