Graphic card is dying? How to check graphic card health?

ninoaoe

Honorable
Mar 3, 2014
63
0
10,640
I recently tried to overcklock my GTX 570. After playing Thief ,the screen blacked out and i had to do a restart.This happens everytime with Thief and now i see it happens after playing Dead Space as well for a few minutes.
I think i have damaged the graphic card and i wanted to check that. Is there any way? Except benchmarks wich my card seems to do just fine

Thanks a lot!
 
Solution
well im not 100% sure. for me the signs suggests a power failure. most common gpu failure signs are artifacts and dropped performance not sudden black screens.570 really hit really high temps generaly. i would like to always have it below 90c though. thats why i also suggested to try the gpu on another system if possibly can. or another gpu on yours
http://www.hardwaresecrets.com/article/ADATA-BN-550-Power-Supply-Review/1268
your psu can out put 480/12=40A on the 12v rail more than the 35A recommended for your 570 but having a black screen suddenly while playing it does sounds like a psu issue. how much have you o/ced your gpu? have you o/ced the cpu also.?maybe all this o/c is giving your psu a hard time. also if you o/ced too much maybe you need more voltage for your gpu to get it stable. monitor temps of cpu-gpu on idle and under load and also the voltages to see if are within range

also you should also state your mobos specific model.
 

ninoaoe

Honorable
Mar 3, 2014
63
0
10,640
You might be right about the psu although I could play all the games just fine not so long ago..
About the graphic card I have put it back to default since the first crash. The cpu yeah is overclocked at 3,3 so I tried with stock speed to play thief again. The only thing that happened was that it took a little more time for it to crash. Might be the psu as you say but why suddenly? P55M-UD2 is my motherboard
 
first of all return to stock speeds on everything.try your games again and see if you get a crush again.if it crushes again it maybe a psu issue.also check if its ventilated properly or overheating. if you dont get a crush you dont have a stable cpu o/c.
you have to start benchmarking. o/c the cpu and try a prime 95 to determine you cpu o/c stability at least half an hour. maybe as hardware wears out you will need additional voltage to get it stable while you check your psu also.
 

ekagori

Honorable
Feb 9, 2013
407
2
10,960
Well power supply capacitors wear out over time and it means that the psu can no longer provide it's rated output, especially for prolonged periods or under heavy stress. If adata didn't use high quality capacitors and/or other components, then the wear will be more premature hence creating psu related problems. For longevity stick to seasonic, corsair's hx/ax/tx series or xfx (made by seasonic), nzxt's hale psu's were also good.
 

ninoaoe

Honorable
Mar 3, 2014
63
0
10,640
With everything stock it crashes again so you say tha it's the psu.. the GTX 570 has minimum requirement a 550W psu. I am just wondering why now? I have this system more than a year and i don't remember crashes like this before
 
examine it a bit further, noises like coil whine, overheating and sudden black screens under load is a sign of a dying psu.
consider these as great options and better than your current one that can easily run your setup or a possible future upgrade with cards like 280x-770 plus o/c
http://pcpartpicker.com/part/rosewill-power-supply-capstone550
http://pcpartpicker.com/part/xfx-power-supply-p1550sxxb9

make sure anything isnt overheating
 

ninoaoe

Honorable
Mar 3, 2014
63
0
10,640
Ok new things happened. Sometimes the card's fan would start working like crazy and nor gpuz neither evga precision x could show the temperature. I installed the previous drivers of Nvidia as well and it seems to be working a little bit more before the crash
 

ninoaoe

Honorable
Mar 3, 2014
63
0
10,640
So you still insist that it's the psu! :lol:
I was thinking of buying a new one but i just don't want to spend money on something that i am not sure of!

GPU TEMPS (via Kombustor MSI)
Idle : 46 celcius (but i remember less usually)
Stress: up to 80 and maybe more

And i saw a GPU throttling from 732 to 366 Mhz
 
well im not 100% sure. for me the signs suggests a power failure. most common gpu failure signs are artifacts and dropped performance not sudden black screens.570 really hit really high temps generaly. i would like to always have it below 90c though. thats why i also suggested to try the gpu on another system if possibly can. or another gpu on yours
 
Solution
well my domestic marker didnt (dont) have the psu i would like to have.i got it at a really nice price. its not top quality. its ctw(channel well tech) a new product at the time which has thankfully proven good.i have measured my system through plug power meter and never passed 270w playing crysis 3, far cry 3 -metro 2033 and othe demanding games. only at times have a peak close to 300w. quality is what you are looking for not quantity. my card has a 30A recommendation on the 12v rail witch means 12*30=360w.even a 550 would suffice.mine has 52A. it also depends on the temp the psu can output its rated power. low end are rated for 25-30c mine is 40c top ones are at 50c or even more. the higher temp the lower the ability of the psu to output power. the psu 's i have previously posted are better than mine in terms of quality and can continiously output their rated power at 50c. the xfx is seasonic made and top quality for its price.:)
 

ninoaoe

Honorable
Mar 3, 2014
63
0
10,640
ok here's what i did:
I bought a CX750 crosair cause it was the only one that i could get ASAP.
Thief worked like a charm but when i tried Dead Space 3 it crashed again at the same point. So i had to restart once more.
Thief crashed again using some different settings (exclusive fullscreen).
I don't think that it's because of the games. Something else must be
 

TRENDING THREADS