Gfx card overheats

[Zanu]

Evenin'. I've had problems with my graphics card for the past few weeks. Every time I play a graphics-intensive game for longer than, say, 30min, the computer shuts down as if the power had been cut off.
I found reports about the gfx card overheating in event viewer, so I made sure the fan was spinning and tied my cables together as best I could. Just got 3 case fans.

 

[Zanu]

I downloaded hwmonitor and the GPU is at around 85C while idling and around 110 while playing CoD4. CPU temp is 30-50. I used ATI CCC to manually crank the GPU fan to 100% to lower the temp temporarily.

Here are my specs:
MB: Asus M4A785TD-V EVO
GPU: ATI Radeon HD7700
PSU: Cougar 750W(IIRC)
CPU: AMD Phenom II Quad 3.4 gHz
One 200mm fan, 4 120mm Nexus fans

 

[Zanu]

Oh, and another thing, when I restart the computer after it has shut down, the fans keep spinning at top speed until some minutes after Windows has booted. If I use BIOS for longer than a few minutes and sometimes while Windows is booting up, the screen loses signal but the computer keeps running. I'm using Windows 7 64-bit.

 

[Chainzsaw]

Hello,

Wow there is something seriously wrong with your GPU. It shouldn't even be reaching that temp (110C is not good).

Did you buy your video card recently?

Is there anything physically wrong with you GPU? Is the cooler even on properly?

Have you tried re-seating your video card (and while re-seating it make sure to check out the entire card itself, make sure the heatsink and fan assembly are on properly).

With such a weak GPU, it shouldn't be getting that hot.

What case are you using currently?

 

[Zanu]

I got the entire computer custom made in January of 2010 IIRC. The case is an Antec 600. I'll check the cooler and seating now.

 

[Chainzsaw]

I got the entire computer custom made in January of 2010 IIRC. The case is an Antec 600. I'll check the cooler and seating now.

Are you sure you have the Radeon 7700 then? AFAIK it just got released like last month.

 

[Zanu]

Aw hell, forgot which it was. It's a 5770. Dem numbers be confusin'

 

[Chainzsaw]

Well since your GPU is so old - and i'm not sure if you clean your case regularly...but it could be dust build up on the GPU/FAN.

Your GPU might be going out the door if your GPU is completely clean/dust free.

 

[Zanu]

I cleaned my case with a vacuum cleaner some days ago. Don't have any compressed air available to thoroughly clean the GPU. What's weird is that even with the temps that high, the fan barely reaches 80% speed automatically.

Actually, the temps seem to be steadily lowering right now. It's at 64 degrees C now with the fan at 45% speed.

 

[Chainzsaw]

Maybe your fan is dieing. It happens. They don't last forever.

You could possibly buy an aftermarkeget cooler for your GPU, however with it being so weak - it would be more worthwhile to buy a brand new GPU.

 

[Zanu]

I can still get 100% speed manually with CCC. If I do indeed need a new GPU, would you kindly recommend some decent ones for under 150 euros? http://multitronic.fi/index.php?setlang=en
Edit: Is it plausible that the heat sensor (or whatever is used) might be broken?

 

[Chainzsaw]

This is a decent card:
Asus GTX 550 TI 125 Euros

http://multitronic.fi/showprod.php?prod_id=ENGTX550+TI+DI%2F1GD5&b=1

As well as this:

Asus 6770 1GB 121 Euros

http://multitronic.fi/showprod.php?prod_id=EAH6770%2FDI%2F1GD5&b=1

Both are similar in terms of performance.

 

[Zanu]

Is there a way to remove unused cables from my PSU? The entire bottom of my case is clogged with them, can't hurt to try.

 

[Zanu]

I can't afford a new GPU right now, I'll likely get one of those bendy fans that you plug in your motherboard. Should it draw air away from or toward the GPU?

 

[Chainzsaw]

You will definately want air drawn away from your GPU. However close as possible you can get it to the GPU the better.

 

[Zanu]

Right, thanks. I've found a bunch of fatal Northbridge errors in eventviewer, what of them?

 

[Chainzsaw]

Northbridge errors usually mean there's either something wrong with your chipset drivers, or possibly even a physical problem.

Your Asus M4A785TD EVO uses the 785G chipset. Your northbridge is a 785G and the southbridge is a AMD SB710.

Check this review out for a little more information on the NB:

http://www.bit-tech.net/hardware/motherboards/2009/08/04/amd-785g-integrated-graphics-chipset/1

If I were you, I would go to the Asus website and download your latest drivers for your chipset.

 

[Zanu]

Okay, I updated the chipset drivers but I'm still getting NB errors. No fatals, but I do find a fixed NB error in event viewer after every crash. Now when I'm playing games and the computer crashes, I hear a weird, monotone sound from the speakers for a few seconds after the screen goes black but before the computer shuts down.

 

[Zanu]

Apparently I'm not allowed to edit my messages... Anyway, the sound keeps playing until I turn the computer off with the power button. I seem to have about 50% RAM usage with CCC and Chrome open, could that be a/the problem?

 

[Zanu]

On Mon 26.3.2012 14:24:05 GMT your computer crashed
crash dump file: C:\Windows\Minidump\032612-16614-01.dmp
This was probably caused by the following module: ntoskrnl.exe (nt+0x4A4F0C)
Bugcheck code: 0x124 (0x0, 0xFFFFFA80049238F8, 0x0, 0x0)
Error: WHEA_UNCORRECTABLE_ERROR
file path: C:\Windows\system32\ntoskrnl.exe
product: Microsoft® Windows® Operating System
company: Microsoft Corporation
description: NT Kernel & System
Bug check description: This bug check indicates that a fatal hardware error has occurred. This bug check uses the error data that is provided by the Windows Hardware Error Architecture (WHEA).
This is likely to be caused by a hardware problem problem. This problem might be caused by a thermal issue.
The crash took place in the Windows kernel. Possibly this problem is caused by another driver which cannot be identified at this time.

I updated my BIOS today, still getting crashes. Found this with WhoCrashed.

 

[Chainzsaw]

How much ram do you have, as well as what kind?

Have you tried running memtest 86? If you have more than 1 stick of ram - try running 1 at a time.

 

[Zanu]

I have 2 2-gig sticks, DDR3 if I remember correctly. I've tried booting memtest from both a USB stick and a CD and with both the .bin and .iso formats. I downloaded a .exe memtest to use in Windows and will leave it to check my memory tomorrow when I go to school.

 

[Zanu]

Another one. This time it probably crashed while running the Windows .exe memtest, I was away when it happened.

On Tue 3.4.2012 6:11:57 GMT your computer crashed
crash dump file: C:\Windows\Minidump\040312-36535-01.dmp
This was probably caused by the following module: ntoskrnl.exe (nt+0x7CD40)
Bugcheck code: 0xA0 (0xB, 0xBFEA2000, 0x3, 0x151BC000)
Error: INTERNAL_POWER_ERROR
file path: C:\Windows\system32\ntoskrnl.exe
product: Microsoft® Windows® Operating System
company: Microsoft Corporation
description: NT Kernel & System
Bug check description: This bug check indicates that the power policy manager experienced a fatal error.
This problem might be caused by a thermal issue.
The crash took place in the Windows kernel. Possibly this problem is caused by another driver which cannot be identified at this time.

 

[Zanu]

I got the Antec Spot Cool and now the GFX runs at around 90 bloody degrees under very light load.