New Card, Random Blackouts

[hoopslam]

I just installed a new 9800gt (replacing a 7600gt) and all was good for the first 10-15 minutes. I got to testing some games on it(Grid and CS:S) and everything runs great except the screen blacks out every once in a while. While the screen's blanked out, I can still hear the game sounds and everything is still running, just not displaying on the screen for a brief second. The blackouts are occurring every 1-5 minutes or so. It doesn't matter what I'm running/not running, I seem to be getting these at very random times.

The specific model I bought is the Gigabyte GV-N98TOC.

There are three things that I think might be causing these random screen blackouts:

#1 My power supply. It's an Antec SU 380W. It's got a 6 pin pcie connector built in, which lends me to believe it can handle 1 pcie card but Gigabyte recommends a 550W for this card so maybe it's this psu that's causing the problem?

#2 There's something wrong with the signal coming from the video video card to the monitor.

#3 Bad drivers/bad windows settings? Maybe I need to mess around with the refresh rates or something. I'm running @ 1440x900 with 60hz refresh on a 19inch lcd. Haven't had any problems with these settings with my old 7600gt.

I did have my system OC'd prior to installing this card and upon first using this card but I started getting these blackouts and thought that maybe if I reduce my CPU speeds, it'll reduce the power draw, thus reducing the random blackouts but changing my cpu speeds doesn't seem to be having any effect on this problem. Went from e5400 OC'd at 3.5 with 1.31 volts down to 2.7ghz(stock) with 1.26 volts. Will changing just the cpu effect my power usage much?

I'll most likely stop by Fry's tomorrow to pick up a beefier power supply to see if it is indeed my psu that's causing this problem. If it is, do you guys have any recommendations for a cheap (less than $30) psu for my build:

e5400 (usually oc'd to 3.5 with 1.31 volt)
9800gt
4 gb ddr2 800
7400rpm hard drive
dvd/burn rom
g31 mobo

Edit: Temps seem very stable, reading max of 41c, 39c idle from hwmonitor

 

[hundredislandsboy]

To make it easier to isolate the problem, lower the variables. The first suspect is a heat issue. You are already overclocking the CPU and probably the GPU core and memory too. Run everything on stock and see if the problem persists.

Unistall the videocard card in your hardware/device manager screen. When the system reboots, don't install the drivers from the Windows prompt.

I would think 380 watts is enough to run your system but it's also possible there's not enough amps on the 12v rail/s. Check the specs of your PSU. If your your PSU is already being maxed out or redlined, overclocking the components will push the PSU past your max limits.

Use the Nvidia setup file to install the latest driver.

 

[WR2]

Using the Antec NSK2400 HTPC case?
Antec SU380 PSU data label.

 

[hoopslam]

Just wanted to let you good people at Tom's know that my problem has been solved. Apparently, it was my power supply because I replaced the 380watt (yes, the Antec SU380 WR2) with an Antec EA 500 and everything is not only running flawlessly but also ten times quieter! Who knew that a PSU change could have such a big effect on a system...well, maybe you guys did but now I do too lol...