RMA or not to RMA

[hotchip]

Hi guys

Pretty much on the day of release i bought a sapphire 7950 OC edition, its been a fantastic card up until now but i think its turned faulty.

The problem is that under stress (either BF3 on ultra or furmark) the screen just goes black and windows crashes after about 30sec to 5 mins. This can happen ever time i try and then not happen for a few days.

I have tried running memtest, and prime 95 to check cpu and memory and all is fine. I have no way of checking the psu to see if thats the problem, but after the crashes the fans are still spinning so its not getting a complete power cut. Ive removed all overclocks and the problem still persists.

I have also tried drivers 12.1 12.2 12.3 and 12.4

My question is shall i just RMA the card or am i missing something?

i7-950 @ stock

Asus rampage 2 extreme

6GB Mushkin radio active 2005mhz

850w Enermax galaxy psu

ocz 120GB ssd

1TB storage

sapphire 7950 oc edition

Many thanks

 

[cerbrus2]

Im not sure just how true it is, But i have read lots and lots online about Furmark damaging cards. I cant see it being true myself. but that's the buzz on the net.

Run afterburner in the background on BF3. and make sure to display the GPU usage and temprature. and make sure it isn't getting to hot. if it is you can use afterburner to change the fan profile. and make a steaper curv on the fan profile. to keep it a bit cooler. if all that fails. then RMA it.

 

[hotchip]

Im not sure just how true it is, But i have read lots and lots online about Furmark damaging cards. I cant see it being true myself. but that's the buzz on the net.

Run afterburner in the background on BF3. and make sure to display the GPU usage and temprature. and make sure it isn't getting to hot. if it is you can use afterburner to change the fan profile. and make a steaper curv on the fan profile. to keep it a bit cooler. if all that fails. then RMA it.

No its not overheating when stress testing or gaming, max temps are about 70-72c when stress testing and about mid 60`s in BF3.

This couldnt be a motherboard fault then?, is it deffinatly looking like a faulty GPU? I have no idea how to test the motherboard without swapping it and i only have 1

 

[jamesriley94]

Furmark damaged my card! Locked the fan to 20 percent without thinking, ran furmark then went to make a cup of tea. Came back to the card at 120 degrees C and then constant crashes and stuff since then! Cant really blame furmark for that though

The fan is not really an indicator since that can draw it's power straight from the motherboard or power cable without being controlled by the gpu if it is having problems.

I'd personally RMA as it isnt worth the hassle of trying to fix it, but it's up to you. You sure it isnt overclocked too much?

100th post!!

 

[airdeano]

what is the crashlog code? this should put you in the area of issue.

event log can be very useful for this issue.

airdeano

 

[hotchip]

Furmark damaged my card! Locked the fan to 20 percent without thinking, ran furmark then went to make a cup of tea. Came back to the card at 120 degrees C and then constant crashes and stuff since then! Cant really blame furmark for that though

The fan is not really an indicator since that can draw it's power straight from the motherboard or power cable without being controlled by the gpu if it is having problems.

I'd personally RMA as it isnt worth the hassle of trying to fix it, but it's up to you. You sure it isnt overclocked too much?

100th post!!

No it has no overclock on it atm, I Did have it running at 1150/1450 @ 1.81v for quite a few months but it was never overvolted or overheated. Always under 1.25v and 75c

what is the crashlog code? this should put you in the area of issue.

event log can be very useful for this issue.

airdeano

The only thing i get in event viewer is -The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

 

[Firat]

I know this may not relate, but what is you ram usage ? I've been having a similar screen blackouts and it's due to windows f*****g up!

 

[jamesriley94]

Any chance it could be a power supply issue?

My mate is having the same problems as you and he has a 7870 OC with an OC'ed 2700k and he's adamant his Coolermaster GX 650watt isnt enough, although I disagree

Just a though

 

[S_I_N]

Furmark damaged my card! Locked the fan to 20 percent without thinking, ran furmark then went to make a cup of tea. Came back to the card at 120 degrees C and then constant crashes and stuff since then! Cant really blame furmark for that though

The fan is not really an indicator since that can draw it's power straight from the motherboard or power cable without being controlled by the gpu if it is having problems.

I'd personally RMA as it isnt worth the hassle of trying to fix it, but it's up to you. You sure it isnt overclocked too much?

100th post!!

no sir that was not furmark fault your card was damaged. Cant blame it on Furmark. I know ya said it but needed repeating lol

On to topic have you tried the card in another system???

 

[Liarma]

I'd RMA it. better to be safe than sorry.

 

[hotchip]

I know this may not relate, but what is you ram usage ? I've been having a similar screen blackouts and it's due to windows f*****g up!

I`ll have to check that m8 i havent monitored it.

Any chance it could be a power supply issue?

My mate is having the same problems as you and he has a 7870 OC with an OC'ed 2700k and he's adamant his Coolermaster GX 650watt isnt enough, although I disagree

Just a thought

850W psu should be loads for my rig, i did used to crossfire with this set up. Although i guess it could be voltage fluctuations or something the psu is about 5 years old.

no sir that was not furmark fault your card was damaged. Cant blame it on Furmark. I know ya said it but needed repeating lol

On to topic have you tried the card in another system???

No unfortunately i dont have access to another pc

 

[nepas]

Have you monitored the voltage on the card?could it be trying to run at the downclocked volts whilst trying to run at 3D speeds?

 

[KING_OF_SAND]

Furmark damaged my card! Locked the fan to 20 percent without thinking, ran furmark then went to make a cup of tea. Came back to the card at 120 degrees C and then constant crashes and stuff since then! Cant really blame furmark for that though

The fan is not really an indicator since that can draw it's power straight from the motherboard or power cable without being controlled by the gpu if it is having problems.

I'd personally RMA as it isnt worth the hassle of trying to fix it, but it's up to you. You sure it isnt overclocked too much?

100th post!!

Your card should have throttled down after 90c - 100c.

 

[jamesriley94]

umm, apparently not? it was a radeon 5770. It still works now, just constant crashes

 

[MaXiMiZe]

RMA it dude. Easiest solution.

 

[kamoulox]

The fan is not really an indicator since that can draw it's power straight from the motherboard or power cable without being controlled by the gpu if it is having problems.

 

[jonnyfgm]

No it has no overclock on it atm, I Did have it running at 1150/1450 @ 1.81v for quite a few months but it was never overvolted or overheated. Always under 1.25v and 75c

That must be a typo?

Even with the AMD overclocking BIOS you can only go up to 1.3V, if its not a typo then i'd say that might be your problem

 

[jamesriley94]

The fan is not really an indicator since that can draw it's power straight from the motherboard or power cable without being controlled by the gpu if it is having problems.

Did you just copy and paste my previous post word for word...?

 

[Liarma]

Did you just copy and paste my previous post word for word...?

LOOOOOOOL it would appear so. I had someone do that to me once. Why, i do not know

 

[hotchip]

I RMA`d the card and there gonna give me a full refund! wooo! So now i have £380 to spend again, I can probably get the 7970 version now for that money.. result