Alright, I'm getting tired of this

[Josh7289]

I've gone through most games (demos) I have on my computer and the same thing happens with all of them. I just want to get it out quick, so...

I play a game, and shortly after I start, the game locks up, wait 30 seconds...starts working again, but at about 10 - 20 fps... It keeps doing that as long as I play the game until my computer eventually just restarts by itself.

I'm getting tired of this, and I have tried everything I can think of. First of all, I'm using the PC in my sig, so look at that first, please. I used to have some drivers for the ATI motherboard and the IGP on it leftover on my PC, even with the Nvidia video card, so I uninstalled everything ATI-related with Driver Cleaner, and the problem persists.

I completely rewired everything in my computer and got all of the cords out of the way to help air flow, and I even have the side of my case off while in operation, and the problem persists. I even stuck a box fan next to it and tried playing games with the fan moving air in both directions, both from the PC and into the PC, but the problem persists.

Now, while playing games, if I touch the processor's heatsink, it does not feel any warmer than it should, so I don't think it's a problem with that. However, if I touch my video card, it is actually very hot while playing games. Almost uncomfortable to touch, so I'm guessing the problem is the video card overheating, even though it is not overclocked at all. The thing is, I have had this card since late last year, but it has rarely if ever given me problems like this. Perhaps it's the summer weather that could be causing this possible overheating?

I really don't know what to do now, since I can't play any games, but I would appreciate any help.

Oh yeah, and I'm using the newest non-beta Nvidia drivers, whichever those are.

EDIT: 84.21

 

[GalvanizedYankee]

How hot is the discharge air from the PSU? Is it's fan even spinning??

Tape a pocket digital thermometer over the grill and see what you got. Every nerd has
a pocket protector with digital thermometer inplace. Mines on the bench.
DO NOT stick it into the grill. Take the temp to a lock-up condition. At 40C I'd start to
worry if the fan has not ramped up. At 50C you have serious problems. If there is no
discharge air...You know the answer to that one...PM, AntecCSR


...Galvanized

 

[Josh7289]

The fans all throughout my computer are definitely spinning, but you want me to check the temperature of the from the little PSU fan? So you think it's a PSU problem?

 

[GalvanizedYankee]

If the discharge from the PSU is non-existant, yes that is an issue.
If the discharge air is over 40C, yes that is a concern. The PSU will shut down or output
will fall like a stone when hot.


...Galvanized

 

[Josh7289]

I just checked it with an old, possibly-not-functioning-correct indoor and outdoor digital thermometer, and throughout the entire game, while it was running fine until when it locked up until when it restarted, the temparture of the air coming from the small PSU fan was about 90 degrees Fahrenheit, which Google tells me is about 32 degrees Celsius. Also, the temperature reading I got during normal operation, like while typing this post, was just about the exact same thing.

Of possible note is that fact that the room is quite a bit cooler now that it is night, and the game took a bit longer than usual to lock up. Normally, it's within the first three or so minutes of play, but now it took about five minutes or so.

Remember, though, my vidoe card was running quite hot, even though I could not get an accurate reading on it with this thermometer.

 

[GalvanizedYankee]

Sure sounds heat related if shut down takes longer the cooler the ambients get.

If you have any spare PSU you could hook it up w/o removing the one inplace for a test run.

That's what I would do.


...Galvanized

 

[Josh7289]

It's too bad, but I dont' have any spare PSUs. I don't know what I else I could do to pinpoint this problem, but, as you said, the fact that it is heat-related is pretty apparent, I think.

 

[GalvanizedYankee]

Well, you could put a DMM on the 12V and watch while the comp is loaded.
The unit that comes with your case, is infamous. I'm not being funny. The 350, right.
A fair DMM is only$10~$20, buy one if you don't have one. Tip: Never loan tools.


...Galvanized

 

[Josh7289]

So, what's a DDM? ^^

And could you explain to me what the 12V is on the PSU? I know next to nothing about PSUs.

 

[GalvanizedYankee]

Sorry, I edited it. Must have been thinking of double D's

DMM=digital multi-meter. They are used for measuring many things but most use the for voltage readings and OHMs(resistance).

You can pick up the 12V by using the probes on the yellow & black leads of the 4 pin Molex or the 5V at the red & black on the 4 pin Molex.

Google checking PSU voltage and many tutorials will come up. Google digital multimeter. You should not have to spend over $10 for a Harbor Freight cheapy. I just ordered a $45 DMM that is a mini-lab. It should be here today. Anyone that builds a comp should have at least one.


...Galvanized

 

[aboothman]

bump the volts

 

[Oyeve]

Like others have stated it may be a heat issue, but, recently since my new build, I have noticed a lot of recent games locking up and crashing. It turned out to be Norton Anti-virus corporate edition 10.x that was crashing my games. So most of the time when I play games that I know will crash with NAV I disble the autoprotect and everything works great. See whats running in the background, it could be some rouge app causing the crash.

 

[skooma]

I play a game, and shortly after I start, the game locks up, wait 30 seconds...starts working again, but at about 10 - 20 fps... It keeps doing that as long as I play the game until my computer eventually just restarts by itself.

That happened to my buddies rig with his 6600gt although he never played through the pain to where his rig would actually crash.

Running rthdribl while watching the temp we saw that around 90c his card would lock up and then throttle back. The temp would drop along with 20FPS

Put on a zalman vf700 (kind of a pita but it fits) and he's fine now.

 

[pkme2]

Antec SLK2650-BQE case with included Antec 350 Watt SmartPower2.0 power supply.

Your PSU looks as suspect. Replace the PSU with a Forton 450 @ $51

 

[ForgetCassettes]

Looking at your motherboard, it looks as though you have on-board video. Have you tried playing games using the onboard video to see if it still locks up?

 

[ronach]

Pull out your video brd and give it a real good look see, especially look in the area of the heatsink, if you don't see anything obvious like lots of dust build up, then carefully see if you can wiggle the heatsink, if there is any movement that would take the H/S off of the GPU, even a little, then you have to do the deed on the GPU and H/S. Clean the goop off of the GPU, re-goop it with your fav heat transfer paste, and do a good job of re-attaching your H/S firmly to the surface of your GPU. Go for it.

 

[Josh7289]

Alright, I want to reply to each post here individually:

Originally posted by: GalvanizedYankee
Google checking PSU voltage and many tutorials will come up. Google digital multimeter. You should not have to spend over $10 for a Harbor Freight cheapy. I just ordered a $45 DMM that is a mini-lab. It should be here today. Anyone that builds a comp should have at least one.

My dad might already have one, so I'll check on that.

Originally posted by: aboothman
bump the volts

What does this mean?

Originally posted by: Oyeve
Like others have stated it may be a heat issue, but, recently since my new build, I have noticed a lot of recent games locking up and crashing. It turned out to be Norton Anti-virus corporate edition 10.x that was crashing my games. So most of the time when I play games that I know will crash with NAV I disble the autoprotect and everything works great. See whats running in the background, it could be some rouge app causing the crash.

Well, I use AntiVir myself, and I have never had any problems with it before, so I don't think it's that, and I haven't installed anything new or different recently, so I can't imagine what rouge app could be causing the problem. Besides, I wouldn't even know which of my running apps is "rouge", anyway.

Originally posted by: skooma
Running rthdribl while watching the temp we saw that around 90c his card would lock up and then throttle back. The temp would drop along with 20FPS

Put on a zalman vf700 (kind of a pita but it fits) and he's fine now.

First of all, what is "rthdribl" and of what were you measuring the temperature? The video card?

Finally, what is a "pita", besides the food? ><

Originally posted by: pkme2
Antec SLK2650-BQE case with included Antec 350 Watt SmartPower2.0 power supply.

Your PSU looks as suspect. Replace the PSU with a Forton 450 @ $51

Well, if it turns out that the PSU is indeed the problem, then I'll have to look into replacements in another thread or something. :roll:

Originally posted by: ForgetCassettes
Looking at your motherboard, it looks as though you have on-board video. Have you tried playing games using the onboard video to see if it still locks up?

Well, last year, during the sumemr, I actually did play games with the onboard video until I was able to get a video card, which did not happen until later that year, sometime after summer, and I don't believe I ever had any of these lock up problems, and if I did, it was probably for some other reason. I'm thinking that the fact that running games with onboard video does not generate much heat and it never crashed using onboard video combined with the fact that playing with a discrete video card now, which generates considerably more heat, crashes my PC means that the problem is almost assuredly a heat-related one.

Originally posted by: ronach
Pull out your video brd and give it a real good look see, especially look in the area of the heatsink, if you don't see anything obvious like lots of dust build up, then carefully see if you can wiggle the heatsink, if there is any movement that would take the H/S off of the GPU, even a little, then you have to do the deed on the GPU and H/S. Clean the goop off of the GPU, re-goop it with your fav heat transfer paste, and do a good job of re-attaching your H/S firmly to the surface of your GPU. Go for it.

I've already blown through my computer with an air compressor a few times to see if the problem was dust, and apparently it wasn't, since I'm still having problems. Without removing the card from my PC, the heatsink seems to be firmly in place. or ? I don't know...^^

 

[zagood]

Same thing started happening to my wife's 6600GT AGP even with a Zalman vf700 heatsink, ended up having to run her machine with the side panel off since it's an HP with only 1 exhaust fan. She hasn't let me tear apart her case to add an intake fan.

1. http://www.daionet.gr.jp/~masa/rthdribl/
2. pita = Pain In The *butt*
3. You've used compressed air on your case/system, but make sure you check your video card heatsink. Look for dust buildup behind the fan which you may need to scrape to get out. Consider re-mounting the heatsink and applying new thermal compound. Arctic Silver Ceramique is great for video cards.
4. It is possible that it's the power supply, but doubtful. Those PSUs have a combined 25a on the +12v, and they're not separate rails. (edit: as far as specs go)

-z

 

[skooma]

Originally posted by: Josh7289

Originally posted by: skooma
Running rthdribl while watching the temp we saw that around 90c his card would lock up and then throttle back. The temp would drop along with 20FPS

Put on a zalman vf700 (kind of a pita but it fits) and he's fine now.

First of all, what is "rthdribl" and of what were you measuring the temperature? The video card?

Finally, what is a "pita", besides the food? ><

rthdrbl.

Yes, we were watchinbg the temp of the card.

pita= pain in the ass

 

[Josh7289]

So, is rthdrbl just some demaning-on-the-video-card app? How were you checking the temperature of the card?

Also, assuming my PSU only has only one +12V rail (even though it also seems to have a -12V rail?), I just read it before, after, and during a lock up, and it stayed almost exactly at +12V constant, so I think it may be safe to conclude that my problem is not a PSU problem, unless there are some other tests I can do.

So, that leaves the video card, and with a very small possibility, the processor.

Now, you want me to remove the heatsink/fan from my video card and see if it's clean under there? I've never removed it before, so are there any helpful points you guys can give me to prevent me from damaging this card? Also, does eVGA allow any kind of tinkering with the HSFs on their cards?

Thanks for all the help so far, everyone. <i>I've got to get to the bottom of this!</i> ^^

Oh yeah, and since today is quite a bit cooler than previous days in the week, the lock up took about six or seven minutes after I started playing. Also, of note is that some games do not lock up the system, and they seem to be the slightly less demanding games, so, again, it definitely seems like a heat issue.

Again, thanks.

EDIT: I just went back into the game to test the +5V rail (there also seems to be a -5V rail?), and that also stayed constant at almost exactly +5V before, during, and after lock up. Also, the lock up happened after only about 30 seconds this time, probably because I just locked up the game a few minutes ago...Or something. Both times just now I Alt-F4'd the game before it crashed.

 

[skooma]

Well, on the PSU, don't worry about the (-) rails. But I wouldn't be worried about your psu at all anyway as I think its your card overheating.

I forget which card, the pci-e or the agp, that doesn't have the temp sensor enabled. You don't see your temp in your Nvidia control panel?

As for cleaning out the heatsink, try running RTHDRBL first. It is really simple yet VERY demanding on the graphic card. You'll see the FPS listed so you can see if it drops off suddenly after a lockup.

IIRC, the heatsink on those cards are only held on by thoise two little push pins though which allows it to rock a bit on the core. I'd make sure its nice and flush by gently pushing on each side of the heatsink.

 

[Josh7289]

My Nvidia Control Panel says, right now, my 6600GT PCIe is running at 55 C. I don't know whether that is good or not. Can someone tell me?

Also, while running RTHDRBL or anything else, how can I monitor the temperature of my video card?

Thanks again.

EDIT: It also says my core slowdown threshold is 145 C. Perhaps that's what's going on, since I don't have it notifying me? I'll check the box to notify me, then go through a game once more.

 

[skooma]

55c idle is bad. 55c load is very good.

rthdrbl runs in a window so just start rthdrbl and open the window as large as you can but so that you can still read the little temp monitor.

 

[Josh7289]

Alright, I'll download and try it now.

EDIT: I couldn't get a temperature reading while playing a game, only after, which didn't matter since the card had already cooled down. However, during RTHDRBL, I brought it up to the maximum resolution I could before it closed for lack of video memory, which was right before 1024 x 768, and my card got up to and ran on average at 86 C. Apparently, this is a very bad temperature. However, RTHDRBL never locked up and the framerate never dropped down suddenly, but it ran pretty constant at 20 fps - 30 fps at the above specified resolution. So, even though RTHDRBL, do you think my games could be locking up because of my video card temperature, for sure? Do you think cleaning or replacing the video card heatsink/fan would fix the problem, for sure?

Anyway, the graphics were pretty good in RTHDRBL.

EDIT EDIT: Oh yeah, and my card now, at idle, is running at 54 C to 55 C, again.

Thanks.

 

[skooma]

How long did you let it run for?

I don't think any game, save maybe Oblivion, jacks the temps of a card higher than rthdrbl.