Dear All, My pc crashes while playing certain games. I can go on playing Bioshock 2, fifa 2010 for hours on end without anything going wrong. But I am not able to play Burnout Paradise or Mirror's edge without having the computer crash on me after every few minutes. Description of the crash - It never goes to a BSOD, just shuts off. Doesn't restart on its own, requires me to restart it. I turned off the automatic restart option in the system menu hoping to take a look at the BSOD but to no avail. There was also one day when this was happening while playing Bioshock as well, but when I tried playing the game the next day, all was well, and i played for several hours at a stretch and finished the game. My specs - intel pentium D 3.4 Ghz intel g41 motherboard Kingston 3.25gb Ram (32bit system) Western Digital 150Gb ati 4670 hd, 1 gb LG DVD drive My psu - Mercury 450W +3.3v +5v +12v1 +12v2 -12v 24A 15A 17A 16A 0.3A I saved a log of GPU Z during one of the crashes. I'm not able to attach the txt file. Here is a link to it at easy-share.com. http://www.easy-share.com/1911488367/GPU-Z Sensor Log.txt when i try to get the exact amperage of the 12v rails, the HWMonitor doesn't give any individual voltage ratings. All the insides of the cabinet are clean and well dusted. I know mine is not really a great gaming pc, but it is very dear to me as a gaming pc is to every owner. Plz help me. If you could guide me in the right direction, i will be very grateful. Reply soon. Thank you.
I see that you don't have very big HDD, I there enough space on it for all your application to run? I think that those games that you talk about uses physix, try to download new drivers to you ATI (as it does not support Physixs technology).
Could also be a problem with DirectX. See if Microsoft has any updates. Reset your VGA to have default settings. Find a driver update for it. If all fails, it could be a faulty memory stick.
I haven't heard of Mercury brand PSU's, but the amperage looks good enough. I wouldn't mark it off the radar yet, but before that, I would look at two things: temps for the CPU and GPU and if your RAM is a problem. Most motherboards come with some sort of application to monitor temperatures, but if not, something like [google]Speed Fan[/google] will suit your needs. For memory testing, [google]Memtest 86+[/google] is your friend. For the temps, something under 60*C is fine. For memtest, you want zero errors after running it for at least an hour. If those check out, then you can move along to the power supply. Don't rely on software monitoring as the sensor for this on the motherboard isn't checking it at the the source. Use a multimeter on a spare molex connector with the system on. Molex pin-out guide. Up to a 5% variable over or under the values is acceptable, but more than that is a problem. I would be mostly concerned with the +12V rail as that's where much of your power is coming. I do think the power supply is a suspect, but check the above first before you start swapping out parts.
What's the PSU rated at? Could it be when the GPU is engaged it tried to draw more power and the PSU provides inadequate power and crashes the computer?
