View Single Post
Old 09-08-10, 03:56 AM   #11
YCH
Registered User
 
Join Date: Sep 2010
Posts: 1
Default Re: very random going on's (480 not detected/showing on post or in windows randomly)

Hi there, I signed up just to post here because I experienced a similar problem recently.

I had not changed any BIOS settings recently but one day my PC appeared to freeze in Windows 7. Then it started to freeze while the PC was still booting in the BIOS stage. It even froze while I was browsing through the motherboard BIOS trying to ascertain what low-level problem this could be.

I started to underclock my CPU and RAM and everything non-GPU related but still the freezes happen.

I began to suspect my video card because even when the display froze at bootup I could still hear the hard drive clicking softly like it does during a normal bootup. The keyboard still worked too (I could press the Numlock keys and it would toggle so I know the CPU is still responding).

So I transferred my GTX480 to my second PC and it appeared to work even under stressful benchmarks like 3Dmark Vantage, so I returned it to my first PC but the problem occured again. I finally managed to get it to complete a few runs of Vantage on my first PC but I noticed that benchmark numbers were lower than my typical runs (for example the GPU particles test is like 50% lower so it's way beyond the margin of error). Once again I transferred it to the second PC and now the same display freezing problem cropped up as well. So I figure it's the video card for sure.

I put in my RMA to MSI and they have processed it. Should be receiving it this week but I have no idea if that actually found a problem with it like I have or just sent it back to me without any changes. These kinds of problems are annoying because they are intermittent so if for some reason the video card doesn't act up and the manufacturer only tests for 10 minutes it's entirely possible they don't see the fault.

Although I don't have hard proof, this intermittent failure may point to something like a intermittent connection on the video card itself. Maybe with all the heat cycles it's not taking it well. To be fair though my PC is well ventilated and it idles around 45C and during hard benchmarks it goes up to about 85C (case is a roomy Thermaltake FT02 for the first PC, and a relatively well cooled Antec P182 for the second PC). This should be well within the tolerance of the video card. But you know...with something like 3 billion transistors on a huge die with hundreds (thousands?) of pins all it takes is a few pins to decouple from repeated heat cycles and you've got some failure there.
YCH is offline   Reply With Quote