No Signal at 4k/120Hz after 10-20 minutes of gameplay.

[Chribolski]

Hi all,

I just wanted to post regarding my PC & LG C1 OLED not being able to cope with 4K@120Hz.

Everytime I play a game at this resolution/frame rate, about ten minutes in the TV turns black and says no signal. Windows hasn't crashed, I can still control media with the keyboard keys for example, but no signal will return with ctrl+shift+win+b and I have to hard reset.

 

I took my card/motherboard into the shop to see if there was a problem, unplugged everything, tried different drives/ports, and no dice, same error.

 

The game I had been testing with is RE:Village, and I could not get past the Heisenberg boss fight without the same black screen/no signal error.

 

I have put it to 60Hz @ 4K and it's been totally fine. This problem initially began with FFXIV in Windows Fullscreen whilst my desktop was at 120/4K.

Just wanted to see if there was a fix or if anyone has experienced similar problems. Is this kind of false advertising if even HDMI 2.1 can't handle this? - I have a 3080 ASUS TUF OC and a Belkin 2.1 cable (Certified by HDMI.org as being legit 2.1) for clarity.

 

I have tried different HDMI cables, cleaning the PC, reinstalling windows, replacing the PSU, sending the motherboard, RAM, CPU & GPU back to the store I bought it from.

 

I can tell you that I have a ASUS TUF 3080 OC, 16.0GB Dual-Channel DDR4 @ 1599MHz - Corsair Vengeance and AMD Ryzen 5 3600 OC'd to 4.0GHz again by the store.

 

I have been told by a few friends that 120Hz/4K is simply unstable via HDMI but if the cable/TV have them advertised as being able to do it, I'm a bit concerned as to whether it might be something else. It could be the drivers for all I know. Previously I was getting livekernelevent 141 and 117, but this has now stopped since the computer came back from the shop, though this same error does occur. Desktop window manager and the game itself will crash usually however, and event viewer does mention how Display driver nvlddmkm stopped responding and has successfully recovered, although it clearly hasn't. Weirdly when this issue happened with FFXIV, it sometimes took longer than 20 minutes, upto an hour, hour and a half, and when the signal went, the sound was still audible, and it didn't seem like the game had crashed.

 

Anything anyone can do to help, any advice that can be given, would be greatly appreciated.

DESKTOP-N6LQUBN.speccy

[xg32]

whats ur psu

[ejs14]

Possible Overheating

[Chribolski]

A brand new Corsair RM750x. - I think I may have found the problem however, I had one cable from my PSU going into the GPU, via the piggyback cable, I've now changed it to two PCIE 6+2 cables for the two connectors. Not tested yet, but do you think this might be the cause?

[stage]

It's a single rail so it doesn't matter. But it could be a GPU issue (as in transient spikes) and you might need a different power supply.

 

Any errors in event viewer? Edit, nvm I reread that part. You mention reinstalling windows, did gou try reinstalling the drivers with DDU and using the option to block windows automatic install if video drivers? 

[Chribolski]

8 pin pcie cable has a rating of 150w if you use both connectors of a single pcie cable the max draw is still 150w, split rail single rail has nothing to do with it, I think I should be using two cables as opposed to one.

 

Yes, I did.

[xg32]

Don't piggyback the cables, try 2, if that doesn't fix it,  it's very likely a transient spike issue, you can test this by setting the power limit to 70% with msi afterburner or evga x1, at stock, the spikes may not may not crash ur game each time they happen, but they happen often enough.

 

If the crashing stops at 70% power limit, consider a 1000w psu (not 750 or 850w)

[Chribolski]

9 hours ago, xg32 said:

Don't piggyback the cables, try 2, if that doesn't fix it,  it's very likely a transient spike issue, you can test this by setting the power limit to 70% with msi afterburner or evga x1, at stock, the spikes may not may not crash ur game each time they happen, but they happen often enough.

 

If the crashing stops at 70% power limit, consider a 1000w psu (not 750 or 850w)

I tried it with two seperate pci-e plugged into the GPU and the same issue happened. Might it be worth refunding this RM750x and getting a 1000W PSU?

 

As I said, it doesn't seem to happen at 60/4k or 120/1440

[xg32]

1 hour ago, Chribolski said:

I tried it with two seperate pci-e plugged into the GPU and the same issue happened. Might it be worth refunding this RM750x and getting a 1000W PSU?

 

As I said, it doesn't seem to happen at 60/4k or 120/1440

try lowering the power limit of the card first before buying anything.

[Chribolski]

Just a quick update, I got a 1000W PSU, cause I could afford it (Was my birthday recent, and I returned the other PSU) and still, the signal cuts out.

 

I have turned off HDR to see if that would help, and it basically, it runs for longer but in the end it does cut out still.

 

I am RMAing the card AGAIN because I feel like it can only be an issue with the card at this point. I've attached the GPU-Z Logs, and it looks like around 2022-08-23 01:53:53 - It's like the card just turns off?GPU-Z Sensor Log.txt