PC reboots while gaming

[amitej27]

Hi Guys,

I have recently assembled a new PC and it has been rebooting while playing games. Tested with Assassin's Creed Unity and Dragon Age Inquisition.
It runs for a couple of hours of gaming before rebooting. I have disabled Automatic Restart under Startup and Recovery but haven't seen a Blue Screen. The system simply reboots.
Attached are the dumps per BSOD posting instructions.

Hardware:
Processor: Intel Core i5 4440
Motherboard: MSI Gaming 3
PSU: Corsair VX450W
RAM: Corsair Vengeance 8GB
HDD: Seagate Barracuda 200GB
SSD: Samsung 840 Evo 120GB
GPU: Zotac GeForce GTX 750Ti


Initially I was set up with two 8GB sticks. I ran memtest and saw errors on one of the sticks so I'm currently running a single 8Gb stick in the DIMM 2 slot
I've tried running FurMark and the GPU temp goes upto 75C and stays constant

If someone can take a look and help me figure out what could be causing the reboot I would really appreciate it.
Do let me know if there are any other logs I can provide to help with troubleshooting.

Thank you

 

[AstaLaVista]

OK, let me start by saying that this is a UBI game, or in other words... the worse console ports ever to be launched to the public and as such, full of bugs and glitches, unoptimized code, etc, etc, etc.

Being said that... are you running their latest patch?

Have you tried updating your Video and Audio Drivers?

Sorry, I don't look at zip files, not so much for potential viruses but, more importantly, because I am not going to understand them so why bother LOL.

About the best tip I can give you is... Go to UBI Dedicate forum for that game and see what the others are saying about it and perhaps there will be an easy solution to your problem.

Good Luck

 

[amitej27]

Thanks for the tip, agreed on the UBI title but I'm facing the same problem with Dragon Age Inquisition
Both games run very smoothly but after a couple hours the system reboots

Audio, video, bios, chip set, almost all drivers have been updated.

I hope someone looks into the zip file. I'm just following thread rules

Please ignore any typos, I'm on my phone.

 

[Wishmaster]

Usually with a hardware issue, such as RAM, you'll get a BSOD. Or errors in a test.

If it just shuts down without any error or warning at all, it could be a heat issue.
But most likely a power supply issue.

It looks like you may have a bad stick of RAM if its failing a test. But that shouldn't cause a shut down. Or at least I don't think so.
It would cause errors, a game to crash or a shutdown via BSOD though.

 

[Solarstarshines]

I agree the PSU looks a Bit weak for what your running i would go 550 watt 450 looks like you are really taxing it

 

[amitej27]

Thanks guys. Is there a way to narrow down that my PSU is the issue? Maybe some diagnostic tools.

Also, I'm not using the bad RAM stick. Only the one that passed memtest

My max temps from speedfan are below
GPU : 72
Core 1: 58-60
Core 2: 58-60
Core 3: 58-59
Core 4: 57-59


I see the thread was moved from BSOD subforum to gaming so I guess the zip file I provided won't be checked.

Let me know if there's anything else that could help

 

[Wishmaster]

In regards to testing, without needing special equipment, you are quite limited.

Technically I think you have enough power. As A 750ti needs a 300W system but .... and a PSu calculator says a 300W is enough with the equipment listed in the first post but... those numbers are kinda generic.
It all depends on how good of power the PSU puts out. And though Corsair is typically good, I don't really know anything about the one you have. I would think you're probably pretty close though.

But those symptoms scream PSU failure to me.

You have a couple options:

1) Try a bit beefier PSU as a test. Do know anyone who may have an extra laying around? a 500-550W one perhaps? Swap it out and test.


2) Run a CPU/RAM stress test to make sure everything is stable there. It seems you ran a memory test, but thats a bit different from a stress test.
Next a GPU stress test to know its stable

When these components get under stress could be the cause of the shut down.

I would try the stress tests first, and make sure everything is stable, and set properly. if its shutting down mid test with no errors, dowble check all connections then move to testing a different PSU.

 

[amitej27]

Appreciate the response Wishmaster!

I will try a prime95 run to see if it restarts. Any idea how long I should run this? I'm thinking overnight might be overkill
I have run FurMark for an hour with no reboots but I'm not sure if that is long enough to stress test the GPU

The PSU is old, 5 years I think, not sure how much the power output might have degraded. I do not have a higher Watt PSU available, might go ahead and buy one anyway.

 

[Wishmaster]

They do degrade overtime.

I would say if you survive 45minutes - 1hour of Prime you should be fine. Sure, there are instances when some issues wont show up until a few hours in (Ive seen them show at 8hrs in),
but if its a major stability issue or a insufficient power issue, I would think that would show up in an hour or less. Most major issues show themselves pretty fast.

Remember too, keep a close eye on the temps. Heat may be a cause as well and just overlooked so far.

 

[Fantail]

This might sound strange but is the north bridge properly ventilated?
If you don,t have another psu maybe disconnect an unused optical. Ihad a psu with a bad molex connector while back it caused a short of sorts.

 

[sygnus21]

Not to take anything away from Wishmaster as she is more than capable of diagnosing game/PC issues but...

Honestly I wouldn't run Prime 95 on that system as we all know Prime 95 is a very heavy handed stress test program. Heck a lot of desktop PC's with better cooling couldn't do 45 minutes with Prime without pushing temps too high. That said, If you do decide to run Prime 95, keep a darn good eye on your temps.

As for you PS, it's a little low, but should be fine for what you're running. And I agree with Wishmaser in that it could be a PS issue, but without proper tools to test, it's just guess work.

What you might do is take you dump files and post them in the BSOD Help and Support forum and have those guys look at them, then go from there. In fact I would do that first before purchasing any new hardware or making any drastic changes.

The other thing that can cause your issue is an overheating or failing GPU. Especially is the issue only happens during gaming sessions. If you're a handy person you could take the GPU apart and re-apply some new thermo past. If not, leave well enough alone and just apply a can of compressed air to it, and the system.

However, if you notice video artifacts and other strange video anomalies, it's probably time to replace the GPU.

Good luck and let us know.

 

[Layback Bear]

I agree with sygnus21.
Not knowing what is used for cpu cooling I would be very careful with Prime/95. It can cause temps to shoot high in a few minutes.

With a system with lots of cpu cooling, running Prime 95 over 30 minutes is a wast of time. If it doesn't overheat in 30 minutes it's not going to overheat.

Seldom if ever will one work a cpu as hard as Prime 95.

A quality higher wattage power supply has it's advantages.
To do the needed work the higher wattage power supply doesn't have to work as hard, therefore it creates less heat and will last longer. Normally it will also put out cleaner volts and amps.

 

[amitej27]

Thank you syngus21 and Layback Bear

I am using the stock heatsink and cooler. I have the corsair n300 mid tower case with 4 fans, 1 front 1 right side front, 1 rear, 1 upper rear. 3 of these fans are connected to the 3 SysFan ports on the motherboard and one is connected to a molex of the PSU, I might try changing that to check what Fantail mentioned.

Here are my temps from SpeedFan, they hit these numbers and remain consistent while gaming
GPU : 71-72
Core 1: 58-60
Core 2: 58-60
Core 3: 58-59
Core 4: 57-59

No artifacts or video anomalies yet.
And I actually posted this in the BSOD forum but I guess it got moved here, not sure if they would accept the thread again

I'll run Prime95 over the weekend and report back.

Your help is much appreciated!

 

[Wishmaster]

Sygnus21 and Layback Bear are correct. I mentioned Keep a close eye on the temps but didn't really go into detail and failed to ask about cooling. I should have. Sorry.


I would say that those temps look OK for a gaming load on a stock heat sink.
Being that its a stock heat sink though may make a Prime stress a bit risky. So if you do it, stay with it and watch closely.

 

[sygnus21]

Thank you syngus21 and Layback Bear

And I actually posted this in the BSOD forum but I guess it got moved here, not sure if they would accept the thread again

More than likely because you listed it as a game issue. If you're getting BSOD's, you need to state that and post the dump file there. If your issue is strictly game related than stay here.

As for running Prime95, a very close eye on those temps and be prepared to quit the program if things get too heated. No need in frying your CPU.

As for those SpeedFan temps, even though you say those are what's shown during gaming, bear in mind that just because you're gaming doesn't mean all the cores are under full load. In contrast a stress test program like Prime95 will run those cores under full load, and being that you have a stock cooler, I'm willing to bet you'll see temps in the high 60's if not 70's after 10 minutes. However if you stay below 70 on that stock cooler, not only is that impressive, but heat ain't your problem (with regard to the CPU).

And did you say you ran FurMark? This i'll test that your GPU isn't faulting.

Personally unless you're overclocking I see no need for these stress test programs; but on the other hand, if your system is faulty, they will fail/error out quickly... indicating there is an issue with your system.

Good luck.

 

[amitej27]

Ran Prime95 for around 30mins and temps slowly climbed to a maximum of 72C. No reboots during the run.
Strangely, my PC has not rebooted since running Prime.

No idea if this has solved the problem but no reboots yet after hours of gaming.

 

[sygnus21]

Well glad to hear that. Hopefully things has sorted themselves out.

Good luck.

Peace.

 

[Layback Bear]

I don't recommend doing any more Prime 95 with stock cooling.

Your cpu specs.
TCASE 72.72°C




ARK | Intel® Coreâ

The only thing I can think of and it is a stretch is Prime 95 and the high temps did what they use to call a burn in test. This allow your thermal past to heat up and then cool to allow better contact. It can be worth a couple of degrees better cooling sometimes.

 

[amitej27]

Reboots are back and I'm not sure where to go from here other than a new PSU.

What's strange is that it sometimes runs for hours with no reboots and sometimes only minutes. (while gaming)
Anyone has an idea what could cause this variation?

 

[Wishmaster]

At this point I would double check for any hardware/software issues.

First, Double check all connection on the Motherboard are connected well.
Personally, I would go into the motherboard bios and reset them to "Optimized Default" too.

next, Update all drivers to the latest. GPU etc. Allow windows update to find drivers as well. make sure all is up to date.

Run MEM test for a good 8 passes on your memory.


Have you run a spyware/virus scan? Just to be sure. I do doubt this but may be a good idea to rule it out. Malwarebytes would be a good free option.

Just try to rule out everything you can. if all that comes out OK, Im not sure what else to suggest. Id be looking at the PSU myself at that point.


I do agree, I wouldn't run Prime again with a stock cooler. Core temps typically run a bit hotter than the Tcase temp, so you should be safe. But your pushing it quite close since the difference between the two will be less with less efficient cooling (stock). Better cooling will widen the Gap between Core and Tcase a little bit. But this isnt needed unless you are overclocking.

But 72c under Prime isnt bad.
We can gather from that test that the CPU doesn't have a heat issue under normal conditions and doesn't have a major stability issue. So thats a good thing.