GPU Keeps Going Idle (p8 state)

Apr 8, 2018
3
0
10
0
A few weeks ago my power supply became unstable and had to be replaced. During that time it caused my GPU to start idling (405mhz mem and core - p8 state) while playing games, causing them to drop from 60-70fps down below 30fps. Now, my power supply has been replaced, but my GPU continues to throttle down after about 15 mins of playing any game. (up until the bad power supply, it ran my game without issues)

Specs:
Sony Vaio (SVF15A18CXB)
i7 3537u
12gb Ram
GeForce 735m 2GB

I have tried the following solutions to fix the issue

GPU-Z shows the perfcap as idle,
My temps are normal (60-70c)
I have used DDU to roll back drivers
clean installed the most current drivers
clean installed windows and the GPU itself
Used Nvidia Inspector and MSI AB to try and manually set the default clocks
I have set Nvidia control panel to all of the correct "prefer max performance" settings but the issue remains.

I have scoured google, and these forums for a solution for the past two weeks, and nothing seems to work. The only thing I can think of is that when the wonky power supply was connected it triggered something in the GPU to go into some kind of safe mode, and I can't find anything on how to check or fix that. Please help
 

Tolis_GR

Estimable
Mar 13, 2017
1,486
11
3,665
315
How do you record gpu temps? Are those temps under heavy load? Your issue looks like more gpu thermal throttle than power supply fail. Try a free gpu stress/stability software like Unigine Valley. Set every setting to max and your native resolution. Use a trust worthy temp monitoring tool. Double check with a second software. I use Afterburner and GPU-Z sensors. See what temps you get after 10-15 mins of benchmarking with the software.
 
Apr 8, 2018
3
0
10
0
I have used GPU-Z, Nvidia Inspector, Throttle stop and HDW Monitor for temps and they're all reading the same 56-60c at idle and 62-71c under heavy load. I ran Unigine Heaven benchmark for a good 20 mins or so last night and it never went over 71c, it was able to maintain a p0 state for the whole test which I thought was a good sign, but when I went to play fortnite the same p8-405mhz issue returned. I thought it was temp throttling at first as well, but every temp software I've used indicates otherwise. Plus, I've never had a temp issue before, I keep everything SUPER clean and well ventilated, and the issue started when my power supply was causing my whole system to just shut off mid game.
 
Apr 8, 2018
3
0
10
0
I have the GPU-Z logs that show fortnite running at 940mhz and 900mhz and then dropping to 405-405. The log shows the temps at the time as well and they're definitely not too hot.
 

ASK THE COMMUNITY

TRENDING THREADS