PC Crashes When On Heavy Load.

[france86]

Lately my PC has been crashing when it's on a heavy load, specifically when gaming. The weird thing is when it crash, while gaming, the screen shuts off and the fans goes berserk (it goes max speed). I notice that when I'm using the computer just to surf the problem doesn't occur, only when gaming. I also monitored the CPU temp, which it floats around 45C when gaming. On the other note the system doesn't have any sort of OC. Now the system that I have built is not really old but components have been added/upgraded all throughout. The oldest component in there would be CD Drive.

SPECS
Case: Cooler Master CM 690
Mobo: Asus M5A97 EVO (Not the 2.0)
CPU: AMD FX8350
GPU: Nvidia GeForce GTX660
RAM: 4x Corsair Vengeance 4gb DDR3
PSU: Antec 650W Earthwatts
Storage: 1x Seagate 1TB SATA III HDD (OS Drive), 1x Crucial M4 128gb SSD
CPU Cooling: Zalman CNPS20LQ (Liquid Cooling)
OS: Windows 7 Ultimate

The CPU and GPU are fairly new, and when they were installed no problems were occurring. The Mobo and RAM are less than a year old, and the PSU is close to 3 years old, and both drives are very new. I have never had this problem before, and I'm thinking the problem could be the PSU and/or Mobo. Leaning more towards the PSU since new components were added, CPU and GPU, are a bit high end. Now, I don't have extra Mobo or PSU lying around to test which one is the problem, but before I go out there and buy a replacement which component is more likely be the culprit?

 

[kitsunestarwind]

I'd be leaning towards PSU as well, see if you can borrow one to test it out with

 

[quilciri]

When you say "shuts off" do you mean the machine restarts?

 

[france86]

When you say "shuts off" do you mean the machine restarts?

It doesn't restart. It's just blank screen (black), and no functionality. I have to hold the power button down to shut it down.

 

[scout_03]

since this board got 2 pci-e slot i would try the gtx 660 in the second one and do gamiing it hte issueappear then the gtx need to be check on another system .

 

[france86]

since this board got 2 pci-e slot i would try the gtx 660 in the second one and do gamiing it hte issueappear then the gtx need to be check on another system .

Will do that.

On another note, it's not the PSU. I swapped the PSU with a working one and problem still persist.

 

[ohiou_grad_06]

Was it the same type though? I used to use cooler master power supplies that had dual 12v+ rails. But when trying to power some of the phenom 2 6 cores, the machine would shut off, even when I swapped out for a new one of the same model. The fix was I changed the power supply to one with a single rail design, and the machine acted normal again.

 

[france86]

Was it the same type though? I used to use cooler master power supplies that had dual 12v+ rails. But when trying to power some of the phenom 2 6 cores, the machine would shut off, even when I swapped out for a new one of the same model. The fix was I changed the power supply to one with a single rail design, and the machine acted normal again.

Yes, it's the same type of crash. Was going to buy a new PSU anyways for SLI, and I bought a Corsair AX850 Modular. Replaced it and the exact same problem persist.

Now a new problem occurs, sometimes the machine won't even start up properly. It doesn't go past the Asus logo screen, and it doesn't have that 'beep' sound when booting up normally, it's just stuck on the Asus logo screen. Looking at the Mobo's light indicator if the components load up properly, whenever the machine gets stuck on the Asus logo screen sometimes red light comes up on HDD_boot, sometimes it doesn't. Now I know it's not the HDD because I tried booting it up another machine and it loads up fine. So now, I'm thinking it could be the Motherboard.

 

[quilciri]

Ok, finish the isolation of the issue. Take out the CMOS battery, and leave it out for several minutes. Take out the video card (and any other attached cards), unplug all drives from the motherboard. Put the battery back in, turn the machine on and see if it makes it through POST (that beep you normally hear indicates the end of POST).

If not, then it's an issue either with the motherboard or the CPU itself. Maybe someone wiser than me knows a way to confirm which is bad in this case...They're both still under warranty, so I would send them both back to be on the safe side.

RAM errors usually show up in a different way, but it wouldn't hurt to confirm your RAM is good. if the system can make it through POST, I reccommend getting memtest86, sticking the bootable binary on a usb stick and running that (good RAM should have no errors show up during the test). If the system doesn't POST, then try the RAm in another system if you have one available.

good luck!

 

[france86]

Okay, I did some numerous BurnInTest, and results were all components passed with the exception of the RAM. I have 4 DIMM for 4gb DDR3, all are the exact same but 2 sets of 2 were bought at different times. Each sets indicates that they have different version, so I did numerous tests with the RAM testing the 2 sets separately and testing on each DIMM. All testing failed, so I went on and grabbed numerous different RAM (all are compatible with the Mobo) and all of them failed.

So my conclusion would be the DIMM slot on the motherbaord are faulty, because the chances of the all the RAM I tested are faulty is slim to none. Although I didn't test my 4 current RAM on another Mobo. Regardless of the numerous tests I did, I'm still not sure if my conclusion are definite. Is it common for Mobo's DIMM slot to be faulty, and is this BurnInTest reliable?

 

[scout_03]

i would try these ram stick on another system with memtest.

 

[ohiou_grad_06]

Another way to test the board would be test one ram set in one set of slots, if it passes move to the other slots, if it tests good then the board is likely ok. Maybe at that point memtest 1 stick at a time, if all tests good you may have a configuration issue.