Question Losing display during gaming (sometimes), maybe a PSU issue?

[Freeze Aim]

Hey all!

So, as the title leads, I'm having issues where, sometimes, when I play, I lose display, but can still hear audio and my PC is still running and all. To get display signal I have to restart my PC (either quick restart or hard restart).

After doing some searching, I found that it could be the voltages. I monitored during full load, and found that my 8-pin connector voltages specifically are rather low. 8-pin #1 is around 11.8 avg during full load, and 8-pin #2 is about 11.7-11.6 V. The other voltages, like PCIE voltage hover around ~12V. However, during idle load or light load, both 8-pin connectors are around 12V

Is that far too low? My PC specs are the following:


Specs:

CPU: ryzen 5800x3d
GPU: msi gaming x trio rtx 3060 ti
PSU: Seasonic Focus Plus Gold 850W
MB: msi pro carbon ac b450
RAM: 32 gb @3200mhz CL16

Here are the *removed by moderator* by GPU-Z, if anyone's interested.

 

[thestryker]

Do the GPU fans go to maximum RPM when this happens?

Usually this type of thing in general is related to power in some fashion or other so unplugging/reconnecting power cables is a good first place to start. Also looks like your card is pulling ~240-250W so if you've raised the power limit (I believe that card is one of the 220W models) you might also drop that back to 100% to see if there's a change.

 

[Freeze Aim]

Do the GPU fans go to maximum RPM when this happens?

Usually this type of thing in general is related to power in some fashion or other so unplugging/reconnecting power cables is a good first place to start. Also looks like your card is pulling ~240-250W so if you've raised the power limit (I believe that card is one of the 220W models) you might also drop that back to 100% to see if there's a change.

No, the RPM pretty much stays the same. I originally thought it was a driver issue, but using DDU and stuff did not really help.

I haven't tried plugging and unplugging. However, the cables does seem to be properly connected, from what I can see.

As for the wattage, that's odd! I have not touched any setting such that it would go above TDP. The only power setting I've changed is windows one to favour ultimate performance.

I just now played some Smite with friends, and looking at the connectors, they did not drop below 12.0 V, though the load was light, only comsuming around 25% TDP

 

[thestryker]

As for the wattage, that's odd! I have not touched any setting such that it would go above TDP. The only power setting I've changed is windows one to favour ultimate performance.

You've probably got the Gaming X Trio, not the Gaming X, it's 240W TDP so if that's the case it's fine.

I haven't tried plugging and unplugging. However, the cables does seem to be properly connected, from what I can see.

Reconnect all of the cables just the same whether they seem properly connected or not.

Also are you using a daisy chained PCIe connector or two separate cables from the PSU? While this should be okay with the power draw, but if you're only seeing the voltage drop when under high load this is a potential failure point.

 

[Freeze Aim]

You've probably got the Gaming X Trio, not the Gaming X, it's 240W TDP so if that's the case it's fine.

Oh well. Didn't know there was a Gaming X non-trio, my bad!

Also are you using a daisy chained PCIe connector or two separate cables from the PSU? While this should be okay with the power draw, but if you're only seeing the voltage drop when under high load this is a potential failure point.

I'm indeed using a daisy-chained PCIe connector. Do you think that that is the culprit of my weird issue?

 

[thestryker]

Oh well. Didn't know there was a Gaming X non-trio, my bad!

No worries it's just dumb naming and it didn't dawn on me MSI would use multiple different TDP across the models.

I'm indeed using a daisy-chained PCIe connector. Do you think that that is the culprit of my weird issue?

It shouldn't matter, but it's a pretty easy place to start just in case.

 

[--SID--]

Seasonic recommands not to use a daisy chained cable for a >225w GPU.

Is the motherboard BIOS up to date?

What's the brand and model (SKU) of the RAM?

 

[Freeze Aim]

No worries it's just dumb naming and it didn't dawn on me MSI would use multiple different TDP across the models.

It shouldn't matter, but it's a pretty easy place to start just in case.

Yeah. I also just saw, by the comment from @--SID-- that Seasonic actually recommends two separate 8-pin connectors for cards with TDPs higher than 225W

Seasonic recommands not to use a daisy chained cable for a >225w GPU.

Is the motherboard BIOS up to date?

What's the brand and model (SKU) of the RAM?

The brand is Seasonic and model is focus plus (850W).

 

[TerryLaze]

I lose display, but can still hear audio and my PC is still running and all.

Not a power issue, if it was then the PC would shut down or the GPU or pc would crash.
It could just be your monitor or even more probable your hdmi cable, if it's not good quality or a bit worn out it can lose the connection to the PC, you can try pushing the input select button on the monitor, cycle it through to see if it finds the signal again.
Or just change out the hdmi cable.

 

[--SID--]

Yeah. I also just saw, by the comment from @--SID-- that Seasonic actually recommends two separate 8-pin connectors for cards with TDPs higher than 225W

The brand is Seasonic and model is focus plus (850W).

The brand and model of the PSU you mentioned in the topic start. I asked for the brand and model of the RAM!

 

[Freeze Aim]

The brand and model of the PSU you mentioned in the topic start. I asked for the brand and model of the RAM!

Damn I can't read. Anyways, it's Corsair Vengeance Pro.

Not a power issue, if it was then the PC would shut down or the GPU or pc would crash.
It could just be your monitor or even more probable your hdmi cable, if it's not good quality or a bit worn out it can lose the connection to the PC, you can try pushing the input select button on the monitor, cycle it through to see if it finds the signal again.
Or just change out the hdmi cable.

It's neither the monitor nor the cables I believe. I have two monitors, and what are the odds that both cables or both monitors are faulty? Additionally, it only seems to happen during gaming, and even then it's rare, but it happens. Moreover, to get display signal back I have to restart my PC, which wouldn't be required if it were only the cables, I think

 

[--SID--]

Damn I can't read. Anyways, it's Corsair Vengeance Pro.

What's the exact partnumber and version number?

 

[Freeze Aim]

What's the exact partnumber and version number?

Not certain, but why does the RAM interest you? Do you think they are faulty?

 

[--SID--]

Ryzen and Corsair RAM often have issues, depending on the MIC. The part number and version number tells us what the RAM really is.

What BIOS version is the motherboard running?

 

[Vic 40]

Could do this for parts and nrs,

download hwinfo,
install and open it=click run,
close the top window which is the system summary,
in the main window at the left top click "save report",
at the bottom of the next window check "Summary for Clipboard",
after that you'll see what's in the pc,
copy by clicking "copy to clipboard" and rightclick+paste in your next respons

Is a handy tool for knowing temps and such as well including bios version.

 

[Freeze Aim]

Ryzen and Corsair RAM often have issues, depending on the MIC. The part number and version number tells us what the RAM really is.

What BIOS version is the motherboard running?

Oof I’m unsure of the BIOS version, but it should be the latest as I had to recently flash it to support the 5800x3d

Could do this for parts and nrs,



Is a handy tool for knowing temps and such as well including bios version.

I have checked temps on both cpu and gpu, and they are all good. The gpu hotspot is around ~81c, and gpu temp is around ~65c during full load. The cpu is around 68-75c and does not rise above that temp during full load. I do not believe this is a temp issue, as the system would completely shut down and not keep running, I think

 

[Vic 40]

I have checked temps on both cpu and gpu, and they are all good. The gpu hotspot is around ~81c, and gpu temp is around ~65c during full load. The cpu is around 68-75c and does not rise above that temp during full load. I do not believe this is a temp issue, as the system would completely shut down and not keep running, I think

Wasn't the main reason i put that there this time, was for a good list of parts and bios nr.

 

[Freeze Aim]

Ryzen and Corsair RAM often have issues, depending on the MIC. The part number and version number tells us what the RAM really is.

What BIOS version is the motherboard running?

The BIOS version is 7B85v1H1. I think it's the latest, beta version. Could it be that, maybe? But how would that explain the low voltages on 8-pin #1 and 8-pin #2?

Wasn't the main reason i put that there this time, was for a good list of parts and bios nr.

Yeah just saw that. Unfortuantely cannot attach the file... But if you want any specific info, I'll give it!

What's the exact partnumber and version number?

I found the part number, which is CMW32GX4M2C3200C16. I'll have to find out about the other numbers soon.

 

[dafman89]

Hi

I also have the same problem as you with the video graphics although I have an XFX RX 7600 XT with a 650 watt power supply. I have read the comments and I also have the power supply connected to my graphics card with a single pci exprees cable with two 8 pin connectors (my graphics card requires a 16 pin connection), I think it's called a daisy chained cable. Did you manage to solve the problem?