Kadorak

Hi folks,

Although I use computers for a heck of a lot - gaming, writing, coursework - I admit I don't know a darn thing about hardware. I was hoping some of you could help me with a problem I've been having, either by confirming my hunch or offering an alternative explanation.

A month ago, my video card fried, so I took my computer into the local store which I bought it from to have this replaced under warranty. They offered to upgrade the video card from the warranty funds, and I agreed, so they installed a new GeForce GTX 285. The tech support crew also told me that my current power supply would be insufficient, so they installed a new one.

Well, after taking the computer home, it started exhibiting problems while playing modern 3D games. The computer would randomly crash while playing many games; the frequency varied with the game, but it would be about 10-50 minutes. I can give you a list of games that have crashed if you want. Anyways, the monitor would go black and say "no signal detected". The sound would keep going for a few seconds, then crash too. The only solution was to turn off the computer by holding down the power button. I tried using a different monitor, updating my drivers, and updating DirectX, all to no avail.

I sent the computer back into the local store with a note explaining the problem, and they took it into their shop and ran a 24-hour hardware diagnostic program. They then reported back to me that there was no problem with the computer, and that I should try changing resolutions and re-installing any problematic software. Well, I've tried re-installing the games in question, and I've tried playing with lots of resolutions, and the crashing persists.

I looked around this forum, and I saw other people having similar problems who were diagnosed as having power supply issues. This seems plausible, since the shop installed a new power supply alongside the new graphics card, and I noticed that no program can really check on the power supply. I opened up the computer, and found the power supply was an Antec TP-750. I looked at the power requirements of the GeForce GTX 285 and that should be sufficient, and I didn't see anything saying Antec was a terrible brand, but this particular unit could just be a lemon.

Does this sound like the right explanation?

Here are the full specs for the computer, if they give you any hints toward other causes -

Power Supply - Antec TP 750

Rest from Sysspec:

Windows: Microsoft Windows XP Professional 5.1.2600 Service Pack 3
Internet Explorer: 6.0.2900.5512
Memory (RAM): 2046 MB
CPU Info: Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz
CPU Speed: 2397.1 MHz
Sound card: IDT Audio1
Display Adapters: NVIDIA GeForce GTX 285 | NetMeeting driver | RDPDD Chained DD
Monitors: 1
Screen Resolution: 1680 X 1050 - 32 bit
Network: Network Present
Network Adapters: Intel(R) PRO/1000 PL Network Connection - Packet Scheduler Miniport
CD / DVD Drives: D: TSSTcorpCD/DVDW SH-S182M | E: SONY DVD-ROM DDU1615
COM Ports: COM1
LPT Ports: LPT1
Mouse: 3 Button Wheel Mouse Present
Hard Disks: C: 463.0GB
Hard Disks - Free: C: 60.2GB
USB Controllers: 5 host controllers.
Firewire (1394): 1 host controllers.
PCMCIA (Laptops): Not Installed
Manufacturer: Intel Corp.
Product Make: ps9202B2
AC Power Status: OnLine
BIOS Info: AT/AT COMPATIBLE | 06/20/06 | INTEL - 518
Time Zone: Eastern Standard Time
Battery: No Battery
Motherboard: Intel Corporation D975XBX
Modem: Not detected

Thanks!

Phædrus2401

I would say it's either the power supply or overheating. Antec has been rather hit-and-miss lately, they used to have their PSUs made by SeaSonic (one of our recommended brands) but over the past few years they've been switching suppliers a lot and quality is somewhat variable--from decent to garbage.

Let's start with overheating. What type of case do you have? Do you know if it's mid tower, full tower, etc? Also, how many fans does it have, where are they, what size are they (most common is 120mm, about the size of a CD), and what direction are they blowing (in or out)?

Download GPU-Z: http://www.techpowerup.com/gpuz/ and record your GPU temperature while playing games. If it starts to creep higher than about 70C then exit the game immediately because you're going to overheat.

If gaming temperatures stay in the range of 50-65C then you're fine cooling-wise and it's a PSU issue. If that's the case I would suggest a 750-850W unit from Corsair or SeaSonic, the Corsair TX850W should do nicely: http://www.newegg.com/Product/Produc...=Corsair%20850

Kadorak

Thanks. I ran GPU-Z while gaming and the highest temperatures I got were 67 degrees, although it was only 65 degrees when it crashed. So that's sort of in between. If it might be overheating, is there anything relatively easy I can do to cool the computer down a bit, then see if it still crashes?

Phædrus2401

Well, you could add an aftermarket video card cooler, but that's a rather complex process. What about your case, what's the airflow like?

Kadorak

I think it may well be a cooling issue, after running a number more tests. It's a PowerSpec 9202; the case says "Chassis by Cooler Master". Things that look relevant on the old "computer summary" sheet from when I bought it (though the video card and maybe something else was upgraded immediately, and then again when I got the replacement card, I don't think it would've changed any of this):

MEI S775 Heatsink NBT-CMI7754BX
CSP C5 Powerspec Case-MEI
2 MPO Microcenter Mid-Tower Foam
1 CRO Microcenter Mid-Tower Outer Carton

I think I'm far too clueless to make any modifications myself; since they created the problem by installing the new video card [with no warning this might happen], I'll see if I can get a new case, fan, or something that'll keep it cooler. At worst, maybe just give me another one of the older video card.

Phædrus2401

All you need to do is tell me:
How many case fans your computer has
Where they are
What size they are
Which direction they blow

And I'll tell you if you need a larger/better cooled case.