Could video card be overheating?

[onepercent]

When the room temp goes above 27C my system starts hanging and requires a reboot, always during games. This system ran fine for about six months before this gradually became a problem. The system reports these temps when the problem occurs:
Case temp: 33C
CPU temp: 41C

The CPU reading is obviously low enough, but I am wondering if this is an acceptable case temperature. The case has reached 35C at the hottest, but hanging starts before that.

My case has all rounded cables and:
(2) 80mm rear exhaust 40CFM
(2) 80mm front intake 32CFM
(1) 80mm side intake 32CFM

I have also tried taking the side off the case, but hanging still occurs and the case temps remain the same. Should the case temperature be closer to the room temperature?

If the case temperature seems reasonable, then what is the most likely cause? The Leadtek GF4 video card has a pretty huge heatsink and fan that comes on it stock, but could it be overheating?

AMD XP 2000+
Leadtek WinFast Ti4200 ViVo
512MB Corsair XMS PC2700
Soyo Dragon KT333 Ultra Platinum
Enermax 365P-VE 350W PSU
Thermaltake Volcano7 w/Smart FanII
40GB Maxtor ATA133
52x24x52 Lite-On CD/RW
16x Lite-On DVD

[BipolarBill]

If you convert those temps to Farenheit, you will find that they are too LOW if anything. :p

It ain't heat unless the video card fan is dead.

Try completely uninstalling both the sound and video drivers and reinstalling updated versions. Everyone forgets about sound...

[onepercent]

The system never crashes when the case temp reads at 31C or below, but when it gets above 33C it crashes constantly. The video card fan is spinning at what appears to be the normal speed, so I can only assume that it is working correctly. I tried reseating the RAM and video card just for kicks too.

When I place my hand over the exhaust fans, the PSU fan feels like it is pushing out very warm air. Could the PSU be starting to fail when it gets to higher temps?

[BipolarBill]

I'm telling you it's not the temps. It's just coincidence.

If it was the temps, you would be sunk because you can't get them lower if you plan on using the PC.

[FrnchDp]

I agree with BpB (but when does anyone disagree with him!!) It doesn't sound like a temp problem, unless your readings are way off and your system is running much hotter than it indicates. But puters nowaday can run pretty darn hot.

I would start looking at other possible problems. you say this happens during gaming.. what kind? I'm assuming it's 3D gaming.

Maybe a problem with the psu (these seem to be common lately).. check this out -> http://www.jscustompcs.com/power_supply/

I dont know much about your vid card, but it may be at fault too. Update the drivers (both vid and sound, as BpB said).




Maybe the recent heat wave did it in... j/k but it was too hot for Oregon.

good luck

:t

[onepercent]

I agree that it doesn't make much sense, since the temp readings appear normal. Also, since I do not have any other equipment to check temps and must rely on the mobo readings, maybe my readings are way off! I'm pretty sure my CPU is getting proper cooling though, cuz I upgraded from a 50CFM to a 75 CFM fan and dropped the CPU temp about 5C. I also remounted the HSF w/ Arctic Silver just for fun. So even if the CPU reading isn't exactly accurate, I don't think the CPU has anything to do with it.

I am about 99.9999% sure that heat is the culprit. I believe this because once the room temperature rises (and along with it system temps) I get hangs about every 10-30 minutes. This will continue until the temperature drops the next morning, or until I turn on the AC and get the room nice and cool. Then the system runs absolutely spectacular as long as the temp stays low, even for days. This has been going on for a couple months now, and I am stumped.:confused:

[BipolarBill]

I have PCs in which the CPU runs 55°C and up and they are not unstable. It's possible that yours suffered enough heat damage to become somewhat sensitive to lower levels of heat, but I doubt it.

I''ve never seen an Athlon run less than 35°C. That's cold, really. Human body temp is 37°C.

I don't know what to tell you other than to try another video card or return the motherboard.

[Imperion1]

Case temps are fine. Cpu temp is fine.

To find out if its the video card, have the system on and run 3dmark 2001 or 2003. While 3dmark is running touch the back side of the video card where the gpu is, you should be able to hold your finger there for 10 secs. If not, then the vid card is getting too hot. Be careful when you do this, you don't want to accidentally knock something loose.
Another test is to take the side panel off and set a small box fan or 12" floor fan beside the case, with the fan blowing into it.

[co2]

Could be the power supply I had something alot like this happen to me last year. Power supply wasn't able to pump out enough juice when doing hi def games. If you have a spare power supply might be worth checking out

If you don’t have an extra power supply try unhooking some the extra stuff just for testing i.e. Surplus fans cdrom drives usb stuff. When the temp gets higher the electricity transfer isn’t as efficient

Last resort: Unhook everything you don’t need. Disable sound card/remove disable nic/remove or any other pci cards you might have in there. Start from Scratch, get the computer so it’s not crashing also try lower your fsb. Then add things back in till you find out what’s causing it.

One other idea…. Having your heatsink incorrectly mounted. Uneven pressure can toss the temp readings off, and cause other problems. Remove the ram put it back in make sure it’s properly placed. If you have spare ram change it out, and test it. Amd CPU’s are fragile if you remove the heatsink be careful take your take have enough light to see what you’re doing. Good luck dude

[onepercent]

Imperion1-
Thanks for the tips. I tried taking off the side of the case and blowing a fan on the video card, but still got freezes. I also checked the temp on the back of the card after running 3Dmark and it wasn't hot feeling at all.

co2-
I'm leaning towards the power supply as the problem too. I have tried starting from scratch, and the system runs fine until placed under load when warm. Maybe at higher temps the power supply is placed under just enough extra stress to cause it to fail. Got to love those intermittent power supply problems!

[FrnchDp]

I'm leaning towards the power supply as the problem too.

um.. isn't that what I said?!?! Maybe a problem with the psu (these seem to be common lately).. check this out -> http://www.jscustompcs.com/power_supply/ j/k it's cool.. but I have noticed more and more people saying that their vid cards freeze in 3d and I've found that in most cases, the psu is to blame. Just not enough juice to go around. Best thing would be to get your hands on a better psu and try it out. never know...

[BipolarBill]

I've found that bad sound drivers cause more 3D freezes than anything. If 3DMark doesn't freeze the PC and games do, it's probably bad sound drivers. Uninstall and reinstall both video and sound drivers and the problem will likely subside.

Games make noise too. ;)

[ukulele]

When I place my hand over the exhaust fans, the PSU fan feels like it is pushing out very warm air. Could the PSU be starting to fail when it gets to higher temps?

Have you ever dusted out the inside of the psu? Just a thought but most people clean out the dust bunnies but are lazy to open up the psu and clean it out too. Mine was disgusting after a year of use with no cleaning. Make sure that the huge wad of wires coming out from it or ribon cables are not blocking the air flow.

[naf360]

i agree with BpB's non "surgical" approach. try reinstalling your vid and sound drivers first.

if nothing good happens, try resetting your CPU/RAM clock in safe mode. i had a similar problem with an Athlon 1700+ unit, had the system running without any problems for about six months before the system started freezing. i've replaced almost everything (including the power supply) except the CPU, vid card and RAM but the problem persisted.

i also thought the vid card and the whole setup was overheating so i added 50cfm auxillary fans on the case and replaced the stock fan on the vid card. but the freezing continued.

to make the long, ironic story short - the problem was a failing RAM. i only found out because when i reset my cpu/ram clock from 266/266 to 200/200, the system works without a problem even for hours. consequently, i have new RAM sticks installed.

it's worth a try. it will save you time and money. goodluck:(

[Imperion1]

Now, we know its not a heat issue.

Try updating the drivers as previously said.

This isn't a new build is it? Or rebuild?

[onepercent]

Imperion1 -
Now, we know its not a heat issue.

OK, check this. The weather here in Southern Oregon has been in the high 90s and 100s for the last month. During this time, I have had constant crashes except when I use the AC to cool the room to about 75F, then it's fine. This week, a series of storms blew in, and the temperature has been in the 80s for 8 days now. Guess what? I've ran 3DMark, played games, even OC'd my video card. Not one crash.

Now I know that the CPU temps are fine, and the case temps are reasonable, so I don't know why the change in room temp is affecting the system's stability. What I do know is that this would probably be one of the most stable W2K systems in the world if you ran it inside your refridgerator.

If there's one thing I've learned in PC troubleshooting, it is to not discount the obvious just because it is not typical.

BTW, I have already tried reinstalling sound, video, chipset and many other drivers. I have tried drivers which I know are good on my system (because I used them successfully) and newer, updated drivers. I have updated my BIOS, remounted my HSF (twice), reseated RAM, even removed the motherboard and started from scratch. I also used ghost to restore backups of the system from before the problem began. The problem can only be replicated when running 3D applications.

I guess it's time for replace and pray! Now where to start . . .

[causticVapor]

Hold it...

Several questions:
What are your +5v and +12v readings like? Somtimes low readings are not caused by the PSU, but components on the motherboard. And very hot air coming out of the PSU can be a cue to imminent PSU failiure.

Check the temperatures of every single chip on the motherboard, as well as the ATX connector, and especially the voltage regulator MOSFETs. Do this while the system is under CPUBurn load at high priority. I have a feeling this is voltage regulator related and not video card related.

Do you run SETI? UD? F@H? Any other distributed program? If not, then I wouldn't necessarily blame the video card as a problem when you run games. Games tax the CPU and memory busses quite a bit.

If it ran perfectly OK a few months before, and is now experiencing problems, then failing components are possible culprits.