RTX 4090 killed by monitor swap?

[UndaDawg]

So i switched from 3 monitors (1140p@240, 1080p@240 and 1080P@144) down to 2 monitors (2 x 1440p @ 240 hz) and my 4090 overclock went from +1500 memory basically to non overclockable but only on the memory.
Any thoughts?

 

[EarthDog]

I'd gather it just doesn't like the 2x high Hz monitors/config somehow.

 

[UndaDawg]

Yea i mean that could be and Id say ok to that if the drop was like 200-300 on the memory side but now cant get anything on the memory when it was +1500 and that was with core pushed to +220.

Post magically merged: Nov 20, 2022

and just to add to that ive tried just using one monitor and it didnt help the OC is still killed.

 

[EarthDog]

Static electricity through the ports, maybe. No idea bud. Could be anything.

 

[Niku-Sama]

i wonder if its sort of like the "unhackable mining lockout" thing going on. and a dummy display got around it

what ports and what res/refresh rates for each port were you using?
have you tried it back with the three that it was before?

 

[Vishera]

The static electricity through the ports thing actually isn't terribly farfetched. I've heard crazier stories from people working in repair shops, or general IT. There's also the possibility that such an overclock was never truly stable, and running it at that setting screwed something up enough that now the card just immediately starts throwing errors.

 

[stompah]

I remember back in the day I would have a card that would play dead anytime I unplugged a monitor that was turned on. Few hours or days later the card came back to life. Weird but true.

 

[UndaDawg]

i wonder if its sort of like the "unhackable mining lockout" thing going on. and a dummy display got around it

what ports and what res/refresh rates for each port were you using?
have you tried it back with the three that it was before?

So originally it was 3 dp ports. 2 1080p (1@240hz, 1@144hz) and 1 1440p @ 240hz.

I tried just the iriginal monitor by itself 1440p @ 240hz. Did not help.

Post magically merged: Nov 22, 2022

I am actually thinking about removing the card for an hr or so and unplugging power from case to see if it helps. Waste of time or not???

 

[Niku-Sama]

yea thats an idea too.

also set everything back to the way it was before it went unstable at oc and try it again.
that includes drivers and such too.

 

[UndaDawg]

+300 mem crash file. Can someone take a look and tell me what the problem is please?

 

[EarthDog]

Have you tried the suggestions listed earlier.... like repeating the card, running it st all stock, etc...

 

[UndaDawg]

It runs fine at stock.

 

[EarthDog]

So the overclock is unstable. Simple.

Why it's now not stable and was before, no idea...