Crashes for the last 6 months...

[-Mew-]

At least your os is intact. It would be pointless to troubleshoot a system with a corrupt os.

Download and run DDU, and grab the latest drivers for your card if you don't already have them.

Go to the bottom of the page for the latest version.

http://www.guru3d.com/files-details/display-driver-uninstaller-download.html

Did that 7 hours ago.

EDIT :

It's been an uneventful hour. PC is stable around 75°c both GPU and CPU with Prime and Heaven running.

EDIT 2 : Two hours now...

Could it be VRAM or RAM in question? Anyone know any working VRAM test software? I'm noticing Haven runs at like 1500 MB while Xcom 2 runs at like 4.5 GB of VRAM.

Could it be the VRAM? Or the normal RAM?

EDIT 3 : Four hours stable 75 nothings happening I give up for now...

 

[trents]

How about a RAM problem? I would run Memtest86+ overnight to check that.

 

[rayraysfunhouse]

I had similar issues not two weeks prior exact same. Would crash and bsod on witcher 3 and could not simulate with p95 and heaven. Run memtest if it passes most likely power supply. Swapped my psu and no more issues.

 

[-Mew-]

How about a RAM problem? I would run Memtest86+ overnight to check that.

Did for a good 6 hours while I was asleep. I need to find time to like leave the PC alone for 24 hours or so, so it finishes. No problem for the 6 hours.

I had similar issues not two weeks prior exact same. Would crash and bsod on witcher 3 and could not simulate with p95 and heaven. Run memtest if it passes most likely power supply. Swapped my psu and no more issues.

I wish I had bought the PSU locally ;_;

 

[Johan45]

A couple things you could try with the cables is make sure they're all plugged in tight and feed the GFX card with 2 separate cables instead of the piggy back

 

[Nebulous]

I'm thinking the PSU could be the culprit. That CM750 has bad /weak caps according to Jonny Guru here . That unit is most likely not up to the task of feeding that 390 the juice it needs, hence the reboots/crashes.

I'll bet with a quality unit like a supernova G2, the OP won't have that problem.

 

[trents]

A couple things you could try with the cables is make sure they're all plugged in tight and feed the GFX card with 2 separate cables instead of the piggy back

This. If you are using a splitter on one line from the PSU to power the video card switch to two lines. It may mean buying an adapter if you currently have only one line with the correct connector for the video card.

 

[Dlaw]

Could just be the PSU is giving up. Definitely reconnect everything, but if that doesn't work and it passed 6hrs of Memtest plus the P95/Heaven test, I'd bet on the PSU. I had a very similar problem with my 960T/550Ti rig when I used a CM 550w PSU on it. Worked fine for over a year, then suddenly I had random lockups and crashes. Swapped for a XFX Core 550w (rebranded SeaSonic unit), and 3 years later the guy I sold the rig to still uses it every day.

 

[EarthDog]

If it passed P95 and Heaven at the same time... its NOT the PSU (at least its output)... those two tests at the same time use more power than a game ever will.

 

[rayraysfunhouse]

If it passed P95 and Heaven at the same time... its NOT the PSU (at least its output)... those two tests at the same time use more power than a game ever will.

That's the same thing I thought when I was having the same issues ED. I was using the exact same psu with my setup that the op has and used it for a year and a half once I upgraded had all these issues. Passed p95 and heaven running at the same time for 4 hrs but still random crashes during gaming. Swapped psu and runs perfect now

 

[Dlaw]

If it passed P95 and Heaven at the same time... its NOT the PSU (at least its output)... those two tests at the same time use more power than a game ever will.

It probably isn't output, but that doesn't rule out the PSU entirely. There are a few other reasons a PSU can cause issues like this.

Bottom line is, regardless if it is the PSU, that CM only got a rating of 7 on jonnyguru, and voltage output and ripple were both not great. Being that the unit is aging, it should be considered for replacement. If it doesn't fix the problem, at least it won't cause more. My bet is that it will fix the problem, though.

 

[trents]

Yeah, I'm thinking something might be wrong with the voltage regulation of the PSU and that it drops out of tolerance now and then.

 

[Johan45]

Well something about what the OP said about the amount of Vram makes me think either GFX power (separate the power bad cable) or driver/OS conflict

 

[mimart7]

I'm surprised that the OP's os is still intact, given the fact that there have been so many system crashes. If I were the OP, I would run sfc/scannow. I'm leaning towards power issues, but I'm wondering about the other entries in Event Viewer. You can't get too hung up on all the errors listed, but I'm curious about Error and Warning entries. Kernel-Power entries at this point are meaningless. They just indicate that the system has improperly shutdown.

I'd also look into getting a UPS as well.

 

[RollingThunder]

I'm surprised that the OP's os is still intact, given the fact that there have been so many system crashes. If I were the OP, I would run sfc/scannow. I'm leaning towards power issues, but I'm wondering about the other entries in Event Viewer. You can't get too hung up on all the errors listed, but I'm curious about Error and Warning entries. Kernel-Power entries at this point are meaningless. They just indicate that the system has improperly shutdown.

I'd also look into getting a UPS as well.

I have never seen a faulty power supply and unexpected shutdowns show anything in Event Viewer.
Been down that road a few times with earlier Antecs with bad caps and those shutdowns never left
anything in any version of Windows Event Viewer.

 

[satrow]

There might be something in the WER reports in MSInfo32 but they take some reading, one bugcheck/crash might lead to many entries each time WER tries to check online for a fix (if it's allowed to, many users turn this off).

You could try attaching some zipped dumps, or uploading them to a public site that you have control over, OneDrive? Also zip the MSInfo32.nfo, wait ~10 minutes for it to fully populate before saving it.

 

[-Mew-]

The full memtest ran and RAM is okay. I didn't really expect anything but hey it's done.

A couple things you could try with the cables is make sure they're all plugged in tight and feed the GFX card with 2 separate cables instead of the piggy back

Well at this point I'll try anything.

I'm thinking the PSU could be the culprit. That CM750 has bad /weak caps according to Jonny Guru here . That unit is most likely not up to the task of feeding that 390 the juice it needs, hence the reboots/crashes.

I'll bet with a quality unit like a supernova G2, the OP won't have that problem.

Well it crashed like 3 times on Xcom, but interestingly I got 2 driver crashes on the GPU so I'm now concerned. I was playing MPC with hardware acceleration on one monitor and Dota 2 on the other.

This. If you are using a splitter on one line from the PSU to power the video card switch to two lines. It may mean buying an adapter if you currently have only one line with the correct connector for the video card.

Yeah I'll do it.

That's the same thing I thought when I was having the same issues ED. I was using the exact same psu with my setup that the op has and used it for a year and a half once I upgraded had all these issues. Passed p95 and heaven running at the same time for 4 hrs but still random crashes during gaming. Swapped psu and runs perfect now

So any recommendations? Should I go for more then 750W?

It probably isn't output, but that doesn't rule out the PSU entirely. There are a few other reasons a PSU can cause issues like this.

Bottom line is, regardless if it is the PSU, that CM only got a rating of 7 on jonnyguru, and voltage output and ripple were both not great. Being that the unit is aging, it should be considered for replacement. If it doesn't fix the problem, at least it won't cause more. My bet is that it will fix the problem, though.

Well mine is brand new like 6 months old.

Yeah, I'm thinking something might be wrong with the voltage regulation of the PSU and that it drops out of tolerance now and then.

Now there an unexpected thing that happens. When the washing machine motor turns at low speeds, the PSU buzzes. My bathroom isn't even near my room... So I'm starting to think it's the PSU definitely..

Well something about what the OP said about the amount of Vram makes me think either GFX power (separate the power bad cable) or driver/OS conflict

I'm surprised that the OP's os is still intact, given the fact that there have been so many system crashes. If I were the OP, I would run sfc/scannow. I'm leaning towards power issues, but I'm wondering about the other entries in Event Viewer. You can't get too hung up on all the errors listed, but I'm curious about Error and Warning entries. Kernel-Power entries at this point are meaningless. They just indicate that the system has improperly shutdown.

I'd also look into getting a UPS as well.

Well I do have 2 SSDs and from what I know they're a lot more resistant to sudden shutdowns compared to mechanical HDDs.

There might be something in the WER reports in MSInfo32 but they take some reading, one bugcheck/crash might lead to many entries each time WER tries to check online for a fix (if it's allowed to, many users turn this off).

You could try attaching some zipped dumps, or uploading them to a public site that you have control over, OneDrive? Also zip the MSInfo32.nfo, wait ~10 minutes for it to fully populate before saving it.

I'll try it. Once it happens again.

Sorry for the late reply. Yesterday I ran memtest pretty much the whole day...

EDIT : Just did it PSU now runs of 2 separate cables.

 

[-Mew-]

http://www.ebay.com/itm/Pro-LCD-PC-...675304?hash=item56777a0ea8:g:31YAAOSwMmBVlmAf

Would this help with a diagnosis?

 

[trents]

http://www.ebay.com/itm/Pro-LCD-PC-...675304?hash=item56777a0ea8:g:31YAAOSwMmBVlmAf

Would this help with a diagnosis?

The problem with those PSU testers is they don't test under load. On the other hand, you aren't having the problem when under load necessarily. That tester would tell you if the PSU is in spec when idle. A better way is to use a multimeter and test the unit under load. It's not that hard to do and there is a nice sticky note with directions for doing that in the PSU section of the forum.

 

[RJARRRPCGP]

never left
anything in any version of Windows Event Viewer.

Windows Vista and later, IIRC, has a shut down tracker and it detects shut downs without logging out first. You will get a "Critical" entry in the event log for your PC losing power for any reason, having to force it off for any reason and forcing a hard reboot for any reason.

7 and 8x definitely do. 10 probably has that too.