Question While gaming pc monitors turn off and wont come back on without a restart

[j20raspberry]

specs are: CPU = ryzen 5900x, gpu = rtx 3070 vision, PSU = Corsair CX750F, 32gb memory 3200mhz

Hi I'm generally quite knowledgeable when it comes to computers however this issue is one that i cant seem to replicate easily and am having issues fixing

while gaming after some time my monitors will turn off randomly and not come back on without a restart, while this happens i can still hear and talk in discord and the pc stays on. I have observed my gpu temps when this happened once and it was at 76c.

solutions I have tried:

fixing air flow in my case i recently brought 4 noctua fans and configured air flow correctly replacing my old fans with 3 intake on the front and one for exhaust, my CPU is cooled by an aio and to note i do not think the cpu is the issue.

Trying different games, this issue only occurs on rainbow six seige and no others, i play tarkov, rust etc and i have no issues gaming for hours at a time

i recently noticed a driver update for nvidia and since this only started occurring since i have been play siege ( last 3 weeks) i thought i would give this a go. However i have not managed to have enough time to play long enough to make the issue occur

i also checked the event log at the time this happened and the only critical alert is my using the power button to shutdown the computer as the screen is turned off ( i read somewhere that if the graphics card is overheating it will say something about resetting the pci bus )

issues i think it maybe:

PSU seems to be the cause most people are reporting on issues similar to this and with seige being the culprit also however my psu is only 2 years old and im unsure how this is the main issue

GPU: im hoping this is not the issue as on all over games apart from siege it performs well and this issue does not seem to occur

guys I'm really stumped with this one especially considering i can usually fix stuff like this pretty easily, like I said I have issues recreating this and for me i think my next step is to replace PSU, any feedback or advice would be great

 

[DSzymborski]

You have a good budget PSU, but it's entirely possible it's the culprit here. The 30 series cards above a GPU *really* loved quick power spikes, and a lot of good budget PSUs have run into trouble with those cards. It's certainly worth exploring if the PSU is a problem; I'd look for somewhere which has a good return policy and get a PSU from them if you don't have access to borrowing one.

 

[j20raspberry]

You have a good budget PSU, but it's entirely possible it's the culprit here. The 30 series cards above a GPU *really* loved quick power spikes, and a lot of good budget PSUs have run into trouble with those cards. It's certainly worth exploring if the PSU is a problem; I'd look for somewhere which has a good return policy and get a PSU from them if you don't have access to borrowing one.

I'm tempted to rebuy the same psu that way all the cables management can stay since its modular i can just plug the cables straight into the new one and test it. i have never seen the gpu temps go above 76 while on six siege for a couple of hours so i really don't think its the graphics card over heating plus when the monitors turn off i can still hear and talk to people i would think that if the gpu did overheat the whole pc would turn off or at least break

 

[DSzymborski]

I'm tempted to rebuy the same psu that way all the cables management can stay since its modular i can just plug the cables straight into the new one and test it. i have never seen the gpu temps go above 76 while on six siege for a couple of hours so i really don't think its the graphics card over heating plus when the monitors turn off i can still hear and talk to people i would think that if the gpu did overheat the whole pc would turn off or at least break

Seems odd to spend money on the same quality, especially as CX prices have narrowed with the better RMx prices. If you're worried about wasting cable management time, you can still leave all your old cables still routed until you know for sure the PSU is a problem.

And the other problem is that replacing with the exact PSU would help you rule out your current PSU being *defective* but it wouldn't rule out the possibility that it's a design problem with a particular GPU-use scenario and a budget PSU. Which means that if you replaced it with the same model and had the same thing happen, you still wouldn't know if it's the PSU or another thing.

 

[j20raspberry]

Seems odd to spend money on the same quality, especially as CX prices have narrowed with the better RMx prices. If you're worried about wasting cable management time, you can still leave all your old cables still routed until you know for sure the PSU is a problem.

Yeah that's true

 

[j20raspberry]

Seems odd to spend money on the same quality, especially as CX prices have narrowed with the better RMx prices. If you're worried about wasting cable management time, you can still leave all your old cables still routed until you know for sure the PSU is a problem.

And the other problem is that replacing with the exact PSU would help you rule out your current PSU being *defective* but it wouldn't rule out the possibility that it's a design problem with a particular GPU-use scenario and a budget PSU. Which means that if you replaced it with the same model and had the same thing happen, you still wouldn't know if it's the PSU or another thing.

you think this would be a good replacement PSU

Corsair RM850x 850W Modular 80+ Gold PSU | Ebuyer.com

Get a great deal on a Corsair RM850x 850W Modular 80+ Gold PSU as well as thousands of products at Ebuyer!

www.ebuyer.com

 

[DSzymborski]

An excellent one that is among the most recommended PSUs.

If it turns out the CX PSU is not the problem, you could always keep the RMx and sell the CX used to defray part of the cost.

The cables *may* be compatible. I forget off-hand, but Corsair tries to keep their modular cables with similar pinouts, but you'll have to check their site. Modular power cable pinouts is a severe enough issue that I would never suggest anyone re-use them based on my memory.

 

[Eximo]

You are in luck, the CX750F is a type 4 PSU just like the RM850x, so the cables should plug right in.

The CX750 is not type 4 compatible though, so double check.

 

[scout_03]

use os systen to shut down computer also hwinfo could made log of your system behavior you could check sensor page and try new occt to stress test psu and parts .

 

[DSzymborski]

You are in luck, the CX750F is a type 4 PSU just like the RM850x, so the cables should plug right in.

The CX750 is not type 4 compatible though, so double check.

Thanks Eximo! I didn't have the chart in front of me and I didn't want to give advice that could lead someone to damage their PC.

 

[DaleH]

specs are: CPU = ryzen 5900x, gpu = rtx 3070 vision, PSU = Corsair CX750F, 32gb memory 3200mhz

Hi I'm generally quite knowledgeable when it comes to computers however this issue is one that i cant seem to replicate easily and am having issues fixing

while gaming after some time my monitors will turn off randomly and not come back on without a restart, while this happens i can still hear and talk in discord and the pc stays on. I have observed my gpu temps when this happened once and it was at 76c.

solutions I have tried:

fixing air flow in my case i recently brought 4 noctua fans and configured air flow correctly replacing my old fans with 3 intake on the front and one for exhaust, my CPU is cooled by an aio and to note i do not think the cpu is the issue.

Trying different games, this issue only occurs on rainbow six seige and no others, i play tarkov, rust etc and i have no issues gaming for hours at a time

i recently noticed a driver update for nvidia and since this only started occurring since i have been play siege ( last 3 weeks) i thought i would give this a go. However i have not managed to have enough time to play long enough to make the issue occur

i also checked the event log at the time this happened and the only critical alert is my using the power button to shutdown the computer as the screen is turned off ( i read somewhere that if the graphics card is overheating it will say something about resetting the pci bus )

issues i think it maybe:

PSU seems to be the cause most people are reporting on issues similar to this and with seige being the culprit also however my psu is only 2 years old and im unsure how this is the main issue

GPU: im hoping this is not the issue as on all over games apart from siege it performs well and this issue does not seem to occur

guys I'm really stumped with this one especially considering i can usually fix stuff like this pretty easily, like I said I have issues recreating this and for me i think my next step is to replace PSU, any feedback or advice would be great

The age of your PSU or any other component is not so important. Even new things fail. If they were designed so that they never fail, no one could afford them.

 

[j20raspberry]

Okay so after wiping windows and reseating gpu and using a different pcie cable from PSU to graphics card it happened at around the 2 hour mark and this time i have a load of event viewer errors

i seriously don't know what's wrong

Graphics Exception on (GPC 5, TPC 2): VSC_ERR
Graphics Exception: ESR 0x52d084=0xffffffff
Graphics Exception on (GPC 5, TPC 2): GSPILL_ERR
Graphics Exception on (GPC 5, TPC 2): GPULL_ERR
Variable String too Large
Graphics SM Global Exception on (GPC 5, TPC 2, SM 1): Multiple Warp Errors
Error occurred on GPUID: 700
Graphics Exception on (GPC 5, TPC 2): PIN_ERR
Graphics Exception on (GPC 5, TPC 2): PIF_ERR
Graphics Exception: ESR 0x52d614=0xffffffff 0x52d610=0xffffffff
Graphics Exception on GPC 1 ZROP 1: Graphics is hung, FATAL!!
Graphics Exception on GPC 1 ZROP 1: CACHE_HIT_FROM_OTHER_GPC
Graphics Exception on GPC 1 ZROP 0: ZWRRAM_ORPHANS
Graphics Exception on GPC 1 ZROP 0: HUNG_ON_IL_THREAD_RDAT
Graphics Exception on GPC 1 ZROP 0: RDAT_ORPHANS
Graphics Exception on GPC 1 ZROP 0: EOZP_MISMATCH
Graphics Exception on GPC 1 CROP 1: CRD_CACHE_HIT_FROM_OTHER_GPC
Graphics Exception on GPC 1 CROP 1: CRD_LONG_WAIT_ON_WRITEACK
Graphics Exception on GPC 1 CROP 1: CRD_NO_FREE_CLEAN_CACHE_ENTRY
Graphics Exception: ESR 0x50b838=0xffffffff
Graphics Exception on GPC 1 CROP 1: CRD_DEAD_TAGS
Graphics Exception on (GPC 5, TPC 3): VSC_ERR
Graphics Exception on GPC 0 CROP 0: CRD_L2ACK_BUT_WACKFIFO_EMPTY
Graphics Exception on (GPC 5, TPC 3): ACACHE_ERR
Graphics Exception on (GPC 5, TPC 3): STRI_ERR


pretty much these same errors thousands of times when the monitors turn off

any help would be amazing im out of my depth here

 

[j20raspberry]

I have also tried reseating GPU and using different pcie slot in psu and different pcie cable nothing works