Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

This question already has an answer here:

I'm having problems with my desktop computer. It's 5 years old already, but still works well even with the latest games so I'd really like to keep it as long as possible.

Specs listed below:

  • motherboard: ASUS P5N72-T
  • CPU: Intel Core 2 Quad Q9550
  • memory: OCZ 4GB Dual DDR2
  • graphics: Gigabyte GeForce GTX280
  • power unit: Corsair 620W
  • case: Thermaltake Kandalf with decent cooling
  • CPU is cooled by a separate radiator and fan, not the standard box one

The problem started a couple of years ago. When playing GPU-intensive games it kept freezing every hour or so. The intervals began to be smaller and smaller with time. Recently, when playing Farcry 3 I had to restart the machine every 15 minutes which pretty much rendered the game unplayable for me. Also, this summer, when the days were very hot (the room was probably 40+ degrees Celsius) it started freezing even when working normally (i.e. not playing).

When the machine crashes, it literally freezes - the screen stops at the last frame, all sounds stop and it's totally unresponsive. I've tried checking the temperature - the graphics card is definitely the hottest, but remains within acceptable limits, the CPU and motherboard temperatures are OK.

It's been suggested to me that the motherboard I have is known to be problemmatic because of the memory bus bandwidth or so. I've tried downclocking the CPU and it works to a certain extent - freezes don't happen so often, but they still do.

Another interesting things is that when I boot linux on this machine, kernel bootup sometimes freezes when initialising mdev.

Would you have any hints as to how I might diagnose the part that's causing the problems?

share|improve this question

marked as duplicate by Ƭᴇcʜιᴇ007, Heptite, Kevin Panko, Moses, Mokubai Apr 12 '14 at 12:21

This question was marked as an exact duplicate of an existing question.

What is the Temperature reported by the CPU and GPU when it freezes? Run something like RealTemp on a second screen until it freezes, and then see what temperature is displayed. This sounds a LOT like an overheating CPU. – Darth Android Aug 12 '13 at 19:58
@DarthAndroid I don't agree; an overheat should trigger a thermal protection shutdown or maybe a restart, not a freeze. I've had outdated drivers produce a similar result, but that wasn't linked to any kind of thermal issue as observed by the asker. – Aaron Miller Aug 12 '13 at 20:04
@DarthAndroid I've recorded the CPU temperatures using RealTemp and at the moment of the crash the hottest core was at 60 degrees Celsius which is perfectly fine I guess. – Mikolaj Radwan Aug 16 '13 at 21:17
@techie007 I've actually come across the threads you mentioned before but after your post I looked through the answers more carefully. I tried FurMark's burn-in test. It ran for ca. 12 minutes, GPU warmed up to as high as 102 degrees Celsius and then got a freeze. To be sure, 102 is too hot, but then again - does it certainly prove that graphic card overheating is my problem? I remember experiencing freezes with GPU temps at around 80 degrees. I'll try and look into some cooling solutions and see what happens. – Mikolaj Radwan Aug 16 '13 at 22:17

Here are some tips:

  • Please do not install the vendor's OEM drivers that ships with the Gigabyte GTX box. Go to Nvidia and install the official GTX drivers:
  • Check whether or not the wire that connects the GPU fan is connected correctly and working. It may be loose or the GPU fan is not working while in operation.
  • If you still cannot play any 3d games, bring your video card for replacement or repairs.
share|improve this answer

Not the answer you're looking for? Browse other questions tagged .