R9 295X2 stabillity issues

[VonSchakelhoff]

GPU: R9 295X2

CPU: FX 9590

MoBo: M5A99FX pro R2

PSU: Antec High Current Pro 1300W

RAM: 2*8GB Corsair dominator platinum

 

On the above build, the Graphics Card powers off under gaming load (no issues on windows desktop or web browsing). I've already updated the motherboard BIOS to resolve a CPU based issue, along with updating the Graphics card drivers, yet the issue persists. I'm out of ideas, and Google searching the problem has been fruitless. Any of y'all know how to fix this?

[Tedster]

Welcome to the Forum.

 

I'm not sure how far you should trust the PSU - it looks like it might be the problem.

[Godlygamer23]

Welcome to the Forum.

 

I'm not sure how far you should trust the PSU - it looks like it might be the problem.

The power supply is a great one.

http://www.techpowerup.com/reviews/Antec/HCP-1300/1.html

[Guest]

Welcome to the Forum.

 

I'm not sure how far you should trust the PSU - it looks like it might be the problem.

Antec is a fairly well known brand with a good rep from what I can see. However 1300W is definatly enough power which leads me to believe that the power supply could be faulty. 

Try to troubleshoot to find the root problem. Swap out the PSU and use another (if you have one). If the problem remains then you know it's not the PSU that's causing the problem.

 

The odd twist is that the power-out does not affect less demanding tasks. This could mean that the issue lies within your GPU, so to troubleshoot further swap the GPU and keep the original PSU to see if the issue still occurs. 

If the problem is the GPU, monitor it using the GPU-Z 'log to file' tool. This will allow you to see if the maximum voltage is being reached and how that effects other aspects of the card. Keep an eye on the temperature, as the card is water cooled it could be an issue with the pump, causing the card to heat up due to bad flow.

[Tedster]

The power supply is a great one.

http://www.techpowerup.com/reviews/Antec/HCP-1300/1.html

I'd been told that Antec was horrible and decided to believe it. My bad.

 

Antec is a fairly well known brand with a good rep from what I can see. However 1300W is definatly enough power which leads me to believe that the power supply could be faulty. 

snip

 

But yeah, the PSU could be faulty.

[VonSchakelhoff]

Swapping out the PSU didn't help the issue, but the log file suggests that the gpu is reaching it's shutoff temperature (~75 degrees C). 

[OceaniaAE-1]

The 295x2 requires separate 12v rails for a total of 24v to power it. Make sure you are running separate 8 pins to it instead of a splitter. http://www.guru3d.com/articles-pages/amd-radeon-r9-295x2-review,12.html

[OceaniaAE-1]

Single Rail vs. Multi Rail PSU - http://www.overclock.net/t/761202/single-rail-vs-multi-rail-explained

[tmlhalo]

Swapping out the PSU didn't help the issue, but the log file suggests that the gpu is reaching it's shutoff temperature (~75 degrees C). 

Any kinks in the hose, pump sounds like it is working, fan spinning, or any leaks?

[Godlygamer23]

Swapping out the PSU didn't help the issue, but the log file suggests that the gpu is reaching it's shutoff temperature (~75 degrees C). 

Are you monitoring temperatures as you're using it?

[OceaniaAE-1]

Courtesy of Sam66er over at Overclock.net. Pretty known to just shut off during load. 

 

You need to have two eight pin PSU cable for one R9 295x2 each eight pin cable must capable of produce at least 28 A . Cannot use combine cable with 2 x eight pin to one card. It mean you need two different cable producing 28A each

[VonSchakelhoff]

I'll check that I have it connected to the correct rails, but as for kinks in the pipe, that's not the issue (custom loop cooling it, water flow, airflow through the rads is apparent, no kinks, no leaks etc.)

[Godlygamer23]

Courtesy of Sam66er over at Overclock.net. Pretty known to just shut off during load. 

 

You need to have two eight pin PSU cable for one R9 295x2 each eight pin cable must capable of produce at least 28 A . Cannot use combine cable with 2 x eight pin to one card. It mean you need two different cable producing 28A each

So, does that mean you need to use two separate power supply connectors for the card? What I mean is that on my Enermax power supply, I have a single connector(at the power supply end), and it gives two eight pins. Do I need to use a single connector(at the power supply end) for each eight pin plug?

[OceaniaAE-1]

So, does that mean you need to use two separate power supply connectors for the card? What I mean is that on my Enermax power supply, I have a single connector(at the power supply end), and it gives two eight pins. Do I need to use a single connector(at the power supply end) for each eight pin plug?

Yeah you have to use two separate cables straight from the PSU. It looks like your PSU has 4 rails so you just gotta get all 16 pins off of your single connection.

[Godlygamer23]

Yeah you have to use two separate cables straight from the PSU. It looks like your PSU has 4 rails so you just gotta get all 16 pins off of your single connection.

I don't actually intend to buy the card. I was just curious. I knew that the card was a power hog, but I didn't know that was required. Good little piece of information to keep in mind.

[OceaniaAE-1]

I'll check that I have it connected to the correct rails, but as for kinks in the pipe, that's not the issue (custom loop cooling it, water flow, airflow through the rads is apparent, no kinks, no leaks etc.)

Yeah let us know how it goes. 

[OceaniaAE-1]

I don't actually intend to buy the card. I was just curious. I knew that the card was a power hog, but I didn't know that was required. Good little piece of information to keep in mind.

Right on yeah a lot of manufacturers won't even give you that info. Can suck balls to buy everything and realize you are short on power.

[Godlygamer23]

Right on yeah a lot of manufacturers won't even give you that info. Can suck balls to buy everything and realize you are short on power.

I also remember from Anandtech that AMD had to go outside of the PCIe spec to release the card in its current form.

[VonSchakelhoff]

The card is plugged into the third and fourth rails (which are otherwise unoccupied) but the problem is persisting

[OceaniaAE-1]

how are temps looking like @Godlygamer23 mentioned?

[VonSchakelhoff]

45 C at idle, then hovering around 55 C during loading, then in the space of about 5 seconds the temp jumps to 73, from there it takes 11 seconds to reach 75 and cause the GPU to shut off. I could run it again to see if I can get that sort of thing consistently, but this matches up with all the previous instances quite well

[OceaniaAE-1]

45 C at idle, then hovering around 55 C during loading, then in the space of about 5 seconds the temp jumps to 73, from there it takes 11 seconds to reach 75 and cause the GPU to shut off. I could run it again to see if I can get that sort of thing consistently, but this matches up with all the previous instances quite well

I believe the 295x2 has a Thermal Throttle of 75c. 

[OceaniaAE-1]

Is it causing a windows restart/shut off or a game crash?

[VonSchakelhoff]

All I can really see is that the GPU turns off, but everything else in the case (pump, fans, ram, motherboard) stays lit/operational. I assume based on that that windows is still running, but in order to revive video output, I have to power off the whole system, then power it back on after a few seconds. 

[OceaniaAE-1]

Yeah im thinking its throttling and killing the card/ graphics driver crash or its power related. Seeing how it runs until you hit a thermal wall im going with the 1st option. Maybe try MSI afterburner and set up a more aggressive fan profile for the Radiator and card if you can. 75 does seem high.