Evga Gtx 770, Screen goes Black and gpu fans max out.

[RubyTuesday]

Hi so I got a faulty gtx 770 that I'm unsure if the card is dead or not. 

Problems: 

- Screen goes black - prompting a reset

- Fans go to max speed when the screen goes black

 

What it crashes on:

- Crashes on heaven benchmark - at start it crashes after a couple seconds, but after a couple crash cycles it eventually crashes instantly on startup of heaven.

- Crashes on CS:GO and League of Legends after a couple some time.

- Crashed after trying to run halo infinite and getting direct x 12 error message - unsure why it did this

 

What it can run:

- It seems to be able to run Furmark, and is able to heat up the gpu to atleast 80+degrees

 

What I've tried(none successful):

- Cleaned out dust w/ air duster

- Replaced Thermal Paste

- Lowered Core clock/Mem clock/Power limit in afterburner

- changing to aggressive fan curve

- Driver DDU

- Tried on two different PSU's

- Tried on two different motherboards

 

So after all that it still crashes and crashes even worse, wondering if anyone else got has any idea's before I consider the gpu dead.

 

Thanks!

 

Attached will be a sensor log from gpu-z

 

GPU-Z Sensor Log.txt

[Dundersam]

15 minutes ago, RubyTuesday said:

Hi so I got a faulty gtx 770 that I'm unsure if the card is dead or not. 

Problems: 

- Screen goes black - prompting a reset

- Fans go to max speed when the screen goes black

 

What it crashes on:

- Crashes on heaven benchmark - at start it crashes after a couple seconds, but after a couple crash cycles it eventually crashes instantly on startup of heaven.

- Crashes on CS:GO and League of Legends after a couple some time.

- Crashed after trying to run halo infinite and getting direct x 12 error message - unsure why it did this

 

What it can run:

- It seems to be able to run Furmark, and is able to heat up the gpu to atleast 80+degrees

 

What I've tried(none successful):

- Cleaned out dust w/ air duster

- Replaced Thermal Paste

- Lowered Core clock/Mem clock/Power limit in afterburner

- changing to aggressive fan curve

- Driver DDU

- Tried on two different PSU's

- Tried on two different motherboards

 

So after all that it still crashes and crashes even worse, wondering if anyone else got has any idea's before I consider the gpu dead.

 

Thanks!

 

Attached will be a sensor log from gpu-z

 

GPU-Z Sensor Log.txt 201.04 kB · 0 downloads

Have you tested different GPUS or even integrated graphics on those rigs?

[RubyTuesday]

1 minute ago, Dundersam said:

Have you tested different GPUS or even integrated graphics on those rigs?

Yeah both rigs run fine on integrated and on a 3080ti

[WhitetailAni]

That sounds like a dead card to me. GPUs have 2 modes - 2D mode and 3D mode.

2D mode isn't too hard to do and faulty cards can usually manage this. 3D mode is where things start to get wack.

 

[RubyTuesday]

6 minutes ago, FakeKGB said:

That sounds like a dead card to me. GPUs have 2 modes - 2D mode and 3D mode.

2D mode isn't too hard to do and faulty cards can usually manage this. 3D mode is where things start to get wack.

 

Does Furmark only test in 2d mode? For some reason that was one of the only bench marks that the card was able to run consistently. 

[WhitetailAni]

1 minute ago, RubyTuesday said:

Does Furmark only test in 2d mode? For some reason that was one of the only bench marks that the card was able to run consistently. 

Furmark definitely uses 3D, so it doesn't fully make sense.

Hmmm...

Do you have it connected over one PCIe cable with pigtails or two dedicated cables?

[RubyTuesday]

2 minutes ago, FakeKGB said:

Furmark definitely uses 3D, so it doesn't fully make sense.

Hmmm...

Do you have it connected over one PCIe cable with pigtails or two dedicated cables?

I've tried both pigtailed connection and seperate pcie connectors on three different psus rated atleast 650w, all had the same behaviors. This gpu is kind off a mind bender I can't really understand why it does what it does haha.