VGA crashing, PSU problems maybe?

waitforchris

Distinguished
May 28, 2011
23
0
18,510
GPU: ASUS R9 280X
PSU: Corsair VS 650

Hello tomshardware people,
My newest problem is that the VGA(or just the driver itself) started crashing under heavy load. Screen freezes while playing, black screen, and 1-2 seconds after its normal again. A few games which I've experienced the problem with are: MGS V, Mad Max, Guild Wars 2
If it's not under heavy load, then nothing happens ever.
The reason I started suspecting a faulty PSU, is because I've started having another problem, which is: when I start the PC under cold boot, it starts to spin up, it reaches the win10 logo, then it shuts down. No hanging, restarting, just shuts down like I didn't even start it. On second attempt, it boots just fine.
Could it be that the two problems are related, and this is a sign of a faulty PSU?
Thanks in advance :)
 
Solution
No, I do think the PSU is the problem. The temps on the GPU were a bit high, but that would not cause the problems you have had, and the CPU temp was fine. And fortunately, turning the fans up to keep the GPU properly cooled costs nothing.

Your initial description to me just screamed "Power Supply", but I always like to try to rule out overheating, as it can appear to be the same issues.

I know that finding a quality power supply in some countries can be hard at times. Use this Power Supply List as a guide on what to buy, and even more importantly, what NOT to buy. Not every power supply out there is on this list. But the ones that we see here in the USA which have also been reviewed by a respected reviewer are listed...
Hello... Yes it is worth a try if your Temperatures for the CPU and GPU are not the reason... the ASUS R9 280X will need over 250 watts 24 amps @12 Vdc when pushing the graphic settings.

As a "simple test" take a cover off your computer case to reduce the temperature to them.
 

waitforchris

Distinguished
May 28, 2011
23
0
18,510
I've been doing stress tests with the GPU, but it never goes above 82-83 °C, even after 10 mins of 99% load.
Same with CPU, with stress test it produces 58-59 °C, but no crashing if I just use the tester programs.
 
I would suspect the Power Supply. Corsair seems to have chosen to use cheap capacitors in their smaller units, and when those start failing, they will cause random reboots, unexpected shutdowns, and sometimes they will cause random errors that seem to follow no patterns.

The other thing that makes me think it might be the power supply is that none of this happens, when the machine is idle.

However, download HWinfo, and run it before you start playing games. Make sure your CPU is staying at or below 70c, and that the GPU is staying at 75c and below.

While you are playing your games, HWinfo will track the temps for you, and you can switch to it at any time to see what the current, lowest and highest temps have been. Then if you can rule out high temps, then it reinforces the suspicion on the power supply.

There is a Power Supply List kept here on Toms, that breaks down power supplies into Tiers. Tiers 1 and 2 are good power supplies. Tier 3 are good for office and/or very light gaming. Tier 4 is not recommended for any gaming system. And tier 5 is the junk/fireball category which we do not recommend that you ever power up.

Corsair VS, CX and CXM power supplies are listed on that list as Tier 4.
 

Jesus_In_Adidas

Reputable
Sep 9, 2015
272
0
4,810
im having a very similar problem. Very similar boot issues. The issue has solved itself, when the air temperature went down. The pc used to crash when playing simply Hearthstone - ridiculous right!? i also suspect a faulty PSU / maybe heating problem - cannot provide a stable current etc.

What are your system specs?

EDIT:

I took long to reply :D i see the specs and the comments are very detailed and clear. However! i believe the PSU has lost his efficiency/power as it also crashes on cold starts, there's no possibility that the psu temperature is high.

My PSU is OCZ 700 mortal combat super extreme crap box
 

waitforchris

Distinguished
May 28, 2011
23
0
18,510


MOBO: AsRock B85M Pro4
GPU: Asus R9 280X
CPU: i5 4440
PSU: Corsair vs 650
RAM: 2x Kingston HyperX 4GB-1600 MHz
Case: Zalman z3, with 3 stock coolers, 1 big at the back, and 2 small at the top
 

waitforchris

Distinguished
May 28, 2011
23
0
18,510


After an hour with playing Mad Max, HWInfo shows the GPU max temp at 87 degrees, and the CPU at 59 degrees. No crashes aswell, which is surprising, but will comment back later after more playing.
 
GPU is getting too warm.

Download MSI Afterburner, and setup a fan curve in it as follows:

40% at 45°C
50% at 60°C
65% at 70°C
70% at 75°C
80% at 80°C
100% at 85°C

The check the checkbox to allow Afterburner to control the fans at all times. This will cause afterburner to load up at boot time. It loads quietly and will keep your GPU cooler.

The CPU temp looks fine to me.

CPU Temp Chart
80C Hot (100% Load)
75C Warm
70C Warm (Heavy Load)
60C Norm
50C Norm (Medium Load)
40C Norm
30C Cool (Idle)

After this, I think I would start saving up for a quality power supply. The sooner you get a good one, the better.

If you are in the USA, this is a very good unit that is at a very low price due to the mail in rebate.

XFX Core Edition 850W 80+ Bronze Certified Non-Modular ATX Power Supply $84.99 - $35.00 mail in rebate = $49.99

Otherwise, these are units I like a lot:

SeaSonic S12II 620W 80+ Bronze Certified Non-Modular ATX Power Supply $64.99

SeaSonic M12II 620W SLI Ready 80+ Bronze Certified Full Modular Active PFC Power Supply $79.99

EVGA 110-B2-0750-VR 750W 80+ Bronze Certified Semi-Modular ATX Power Supply $89.99 - $10.00 mail in rebate = $79.99
 

waitforchris

Distinguished
May 28, 2011
23
0
18,510


Thanks for your reply! Unfortunately I'm not from the USA, but I'll try to search for an adequate replacement.
So you're implying that the PSU is not the problem here? What about the booting problem then?
Thanks in advance :)

 
No, I do think the PSU is the problem. The temps on the GPU were a bit high, but that would not cause the problems you have had, and the CPU temp was fine. And fortunately, turning the fans up to keep the GPU properly cooled costs nothing.

Your initial description to me just screamed "Power Supply", but I always like to try to rule out overheating, as it can appear to be the same issues.

I know that finding a quality power supply in some countries can be hard at times. Use this Power Supply List as a guide on what to buy, and even more importantly, what NOT to buy. Not every power supply out there is on this list. But the ones that we see here in the USA which have also been reviewed by a respected reviewer are listed. Seasonic and XFX can be assumed to be on the list even if a given model is not listed. Beware of some companies like EVGA where a B1 or G1 unit is not a good one, but the B2 or G2 is a very good unit. So when doing your search on that list, find (CTRL-F) the brand, and then search up or down (using arrows next to the search box) for the model number next to or right under the brand name. If you find both the brand name and the model number, then scroll back up to the section header to see what tier it is. We only recommend buying Tier 1 or Tier 2 power supplies.

If you have any questions on power supplies, you can ask in this thread or in the power supply list thread.
 
Solution