Been having an strange issue with my video card lately:
Any pg game which is more demanding (PS360 port or such things) hangs up after just a few minutes of gameplay. First the screen goes black, then it goes back to desktop and windows says the game crashed and/or the video drivers recovered from a problem.
This is a little bit weird, because my graphics card is somewhat new (bought it at the end of 2012) so it's not damaged or dirty, and some games, such as L.A. Noire and Dragon Age 2, works just fine. Also, I played Zelda Skyward Sword from beginning to end in Dolphin, @1080p and double internal resolution, and no problems occurred.
But any other game I throw at it hangs the same way (Crysis, Bioshock Infinite, Tomb Raider, others). At first I though it was some sucky drivers from nvidia, but after testing 3 different drivers, I'm sick of installing and re-installing this thing.
I started to get suspicious at one thing: My PSU. I took a picture of it's specs so you guys can see and tell me if it's bottlenecking the video card in stressful situations. (It's the BRP 500 model, Sentey)
I'm currently using windows 7 and 334.89 drivers from nvidia, just upgraded from 320.
Maybe because of the overclocking (by factory or user), the video card components degraded over time, to the point that it can't endure that GPU/Memory speed. So underclocking option makes sense here.
Confirmed: The crashes occur every time the GPU reaches above 75ºC.
I though that was a safe temperature?
Used the benchmark in Tomb Raider as the test, which was always crashing even before reaching the results.
Both EVGA Precision and MSI Afterburner shows that my video card is picking up heat pretty quickly (1~2 minutes), easily reaching the 75ºC threshold. The crash happens exactly after that (I got 76ºC just once ).
Curiously, my video card fan was always sticking to 50% usage, so I set up a more aggressive profile, so that it reaches 100% when hitting 70ºC. Even though, the card reaches 76ºC after just a few minutes.
Underclocked both the graphics card processor to 870Mhz down from the original 980Mhz and memory of the card to 1900Mhz down from 2250Mhz and I got a stable peak of 72ºC, at least for half an hour. But it's hardly ideal, as my card is running it's fan at full speed and I'm under the impression that using it like that might damage it even more.
Dropped the clock down a bit more, to 849Mhz in the processor and 1841Mhz in the memory. Going to test those values for a bit and will update later.
This setup of 849Mhz in the processor and 1841Mhz in the memory is stable enough, I just left the card at maximum load for more than three hours; it peaked and then stayed at 72ºC.
Processors which crash at a certain temperature do so because the silicon has degraded allowing voltage to leak into surrounding transistors which causes invalid results to occur
what kind of power supply you running and how old is it? that can cause crash's they too can degrade overtime i ran into a similar problem working on a whitebox setup
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Related Threads
?
?
?
?
?
Next Generation Emulation
2.2M posts
459.8K members
Since 2001
A forum community dedicated to all emulation enthusiasts. Come Join discussion on all platforms from Nintendo, Microsoft Xbox, Sony Playstation, to PC. Coding, tips, builds, specs, tricks and more.