Computer restarts when playing graphic intensive games

[carwiz]

Yeah, I saw that too but the OP said he took care of it. Silly me.

 

[carwiz]

How old is the build? You might try reseating the CPU cooler. Make sure both are clean and use just daub of heatsink compound. Spread it well with a credit card or similar. None is better than too much. Some of the temperature monitors don't report AMD temps very accurately. Until recent models, the AMD CPU would go up in smoke before it would attempt to throttle or save itself.

Rerun the Diagnostic Utility again. Be sure to allow time for it to gather all the files. It can take five minutes or more to complete.

 

[antilegacy]

I recently added more thermal paste and reseated the heatsink. I may have used more grease than needed though. I'll have to redo it tomorrow I guess. But here's the logs.

 

[carwiz]

Well, nothing new there.

Try the heatsink.

 

[Golden]

The latest BSOD on 24 March still points to the Etron driver. There are no .dmp files that point to the CPU. What am I missing here?

 

[carwiz]

The latest BSOD on 24 March still points to the Etron driver. There are no .dmp files that point to the CPU. What am I missing here?

Yes, you are correct and I considered that but review the thread and the OP's claim. It doesn't explain the other sudden stops with no dumps. And the application faults are all over the place. It's as if the CPU is just coming to a hard halt. The sys-log ends in January now for some reason or it might shed some light on the errors.

Added:
Perhaps just disabling the USB-3 hardware to stop the driver load would help isolate the problem.

 

[Golden]

Mmmm.........perhaps the PSU is really shaky at high output?

OCCT does a PSU test where it maxes both CPU and GPU - the idea is to stress the PSU (heat wise). It may be worth a crack?

 

[antilegacy]

I'm at work right now, so I haven't had a chance to mess with the heatsink yet to re-apply the thermal paste more sparingly. But the PSU is brand new(as in I just installed it 2 days ago when it arrived). The only time there's really a problem with the PC is when it's under a heavy load(stress test, playing Dragon age: inquisition). I can play all other games that I own without any problems what so ever. I've also moved all USB cables to the slots that aren't USB 3.0.

I had read else ware that when someone had disabled the 3.0 ports it disabled all of their USB ports.

 

[carwiz]

What's the make and exact model of PSU you installed? You should size the PS so that 50-60% of its rating is enough to supply the systems TDP (Total Design Power). Even the best PSUs that are rated 80+Gold or Platinum range in the 83-87%. Some may hit the low 90s but that is PEAK output efficiency. Few, if any, will sustain that output continuously.

That said, the Etron driver is still questionable. There are many problems with it documented here and on the net. The most recent version seems to be from 2012 and that's directly from Etron. The dump is showing that the driver is out of date or doesn't match. One option would be to try and force Windows to do an update and see what it finds. There must be a more recent (or different) WHQL driver available or WinDbg wouldn't have flagged the driver in the dump(s).

 

[antilegacy]

Rosewill LIGHTNING-1000 Continuous 1000W & 50 Degree C Power Supply - Newegg.com <-This is the PSU that I bought in the last week and installed. The Amps and Watts are a bit higher than I actually need. According to voltage calculators I need a minimum of 650 watts.

I will check on the Etron driver again, but I got that from the gigabyte website for my MOBO. Maybe I'll try the etron website directly for a more up to date version? I have told windows to search for updates.

When I get home the first thing I'm going to do is re-seat the heatsink and apply new thermal paste(since I know I added too much).

 

[carwiz]

I'll be back this evening. Make sure no compound got on the motherboard or CPU pins. That will definitely cause a problem.

 

[antilegacy]

So I re-applied the thermal paste(sparingly), and put a new fan into the case. Ran the OCCT Linpack test and 5 minutes later it shut down again. I then re-installed my graphics drivers with a clean install(only installing the graphics driver), and downloaded the most recent version of Etron from their website.

Running another test now....

 

[antilegacy]

Okay so I ran the CPU large data set test and it almost immediately quacked at me and posted this. Should I assume this means my CPU is going bad?

 

[carwiz]

This is the first time I've seen those charts so someone may need to look at them for a second option. What I see; I don't think it's the CPU as much as the motherboard regulation. The +5vcch is around 3.57v and it should never drop below about 4.75v or 4.5v ABSOLUTE MINIMUM. And the -12v is all over the place which might indicate capacitor failure. Visually check the power regulation capacitors. These are the 14 tubular components in a row next to the CPU on your board. Look for swollen or split cases and/or for discoloration in that area of the board and other components around them. Make sure there is ZERO heatsink compound on the board in that area. Even a fingerprint across bare traces in that area can cause a failure.

 

[antilegacy]

I've been watching the -12v rail for a while going all over the place and thought it was something bad. At least this gives me an answer to which piece is going bad. I didn't know that the +5vcch was that low.

I may grab a cpu/mobo bundle deal and just be done with it if it is the mobo.

 

[carwiz]

I hate to see folks buy parts for diagnostics so it might be worth taking a look anyway. Also check to see if the MB has a warranty. The image below shows the area of the motherboard to check. This is the power regulation components.

 

[carwiz]

Well, I got a little loose on the left side but you can see that the USB-3 components are there too. The vcch powers the integrated circuits. Purely speculation but it could have something to do with the USB-3 driver failure if the chip is under powered or inoperable.

 

[antilegacy]

When looking at new motherboards just now I got a BSOD with ntoskrnl.exe to blame. That's a new development. I do believe it's time to purchase a new board.

 

[carwiz]

Yeah, that can be power related too. Disk drives rely on having good power. Are you sure you didn't bump a connector loose?

Sorry we couldn't find it sooner. Keep us posted.

Spurs are back on so ..... Gotta go!

 

[antilegacy]

Well my case has been open a lot lately trying to fix this problem(new CPU, taking ram out and testing individual sticks, putting in a new fan, adjusting the heatsink). I try to be careful, but it's entirely possible it's my fault.