PC instant reboots when games are launched

[deeledee]

Hey guys,

 

at this point i'm out of sorts already, no matter what games i launched, my rig would instant reboot itself, no errors no blue screen etc. the only time when it doesn't reboot itself is when i surf net/stream videos and also i occasionally leave my PC on for 3-5 days and yea, it doesn't reboot itself either. so at the point when it reboots itself, package temps on the cpu is 60c (launching pubg). not sure what is wrong at this point.

 

no OC to any other parts other than a manual voltage on the cpu, rest of the spec as per sig

Win10 pro, version 1803

cpu - 1.25v stock clocks

 

[amusedschrodinger]

Are you using on board graphics or gpu?

[deeledee]

Just now, amusedschrodinger said:

Are you using on board graphics or gpu?

my main 144hz monitor on my gpu, on-board for my secondary 60hz panel

[Shoe_Eater]

you need more POWAH... literally

try underclocking it and seeing if it still happens, or maybe make sure you have plugged in all the power cables since i would say the most likely cause is just insufficient power to the cpu or gpu.

[Jurrunio]

install a new copy of the OS and see if that helps

[ahuckphin]

week power supply or got thermal solution problem 

[Shoe_Eater]

1 minute ago, Jurrunio said:

install a new copy of the OS and see if that helps

it probably wouldnt be os since if it is a driver issue or something, that would affect things like using chrome and other browsers which use hardware accelleration, which would *lightly* use the gpu, meaning it wouldnt be a driver issue or an os issue, and if it was one of those, it would most likely cause a blue screen

[GoldenLag]

have you locked the voltage at 1.25 volts? have you locked the clockspeed aswell? otherwise me might have allready found your issue.

 

leave the motherboard to control the voltage if you arent overclocking. no need to touch it. 

[Shoe_Eater]

2 minutes ago, bindydad123 said:

week power supply or got thermal solution problem 

wouldnt be thermal since he said while launching a game before it dies, it peaks at 60c, which is in the bounds of perfectly fine, so power issues like i said earlier would have to be the cause

[deeledee]

3 minutes ago, GoldenLag said:

have you locked the voltage at 1.25 volts? have you locked the clockspeed aswell? otherwise me might have allready found your issue.

 

leave the motherboard to control the voltage if you arent overclocking. no need to touch it. 

i locked it at 1.25, if i just let it on auto, it's pretty insane how skylake works, it fluctuates the volts up to 1.4v which really isn't necessary  

[GoldenLag]

1 minute ago, deeledee said:

i locked it at 1.25, if i just let it on auto, it's pretty insane how skylake works, it fluctuates the volts up to 1.4v which really isn't necessary  

that behavior is normal. and as far as the manufacturer is concerned it is neccesary to obtain the boost speeds. 1,4 volts all core continous voltage is something else than a high boost voltage.

 

i hope you have locked the multiplier and validated the overclock or the voltage is what is creating your issues.

[Jurrunio]

10 minutes ago, Shoe_Eater said:

it probably wouldnt be os since if it is a driver issue or something, that would affect things like using chrome and other browsers which use hardware accelleration, which would *lightly* use the gpu, meaning it wouldnt be a driver issue or an os issue, and if it was one of those, it would most likely cause a blue screen

new OS kills problem from all software entirely, much more time efficient.

 

4 minutes ago, deeledee said:

i locked it at 1.25, if i just let it on auto, it's pretty insane how skylake works, it fluctuates the volts up to 1.4v which really isn't necessary  

Try start prime95 and run small FFT test, if it crashes then that's not a stable voltage

[deeledee]

Just now, GoldenLag said:

that behavior is normal. and as far as the manufacturer is concerned it is neccesary to obtain the boost speeds. 1,4 volts all core continous voltage is something else than a high boost voltage.

 

i hope you have locked the multiplier and validated the overclock or the voltage is what is creating your issues.

i'll give it a shot and load defaults, i'm gonna disable my secondary display and launch a game and see how it is.

[deeledee]

attached the graphs right after I load the defaults, i certainly feel uneasy with the stock volts though.

 

 

[NelizMastr]

1 minute ago, deeledee said:

attached the graphs right after I load the defaults, i certainly feel uneasy with the stock volts though.

Those temps are way too high for an AIO and at stock clocks. I'd start looking for why the temps are so high. Remove the cooler, replace thermal paste, make sure it's nice and tightly connected to the CPU.

 

To answer your reboot problem, yes, it's voltage. You need to disable turbo to maintain stability at 1.25 because boosting is unstable.

[deeledee]

so i launched pubg at all stock bios, reboot me once again.

 

2 minutes ago, NelizMastr said:

Those temps are way too high for an AIO and at stock clocks. I'd start looking for why the temps are so high. Remove the cooler, replace thermal paste, make sure it's nice and tightly connected to the CPU.

 

To answer your reboot problem, yes, it's voltage. You need to disable turbo to maintain stability at 1.25 because boosting is unstable.

i'll re-do the cooler and thermal compound. at this point, i'm wondering if it's the P2 rated psu which is the main problem...

[NelizMastr]

Just now, deeledee said:

so i launched pubg at all stock bios, reboot me once again.

 

i'll re-do the cooler and thermal compound. at this point, i'm wondering if it's the P2 rated psu which is the main problem...

Could be, but again, for a stock CPU, those temps are almost dangerous, so you'll need to fix both.

[deeledee]

4 hours ago, NelizMastr said:

Could be, but again, for a stock CPU, those temps are almost dangerous, so you'll need to fix both.

re-seated cooler with new thermal compound, while doing that, all 4 thumbscrews were not tight fit, basically they were so loose which it doesn't take any effort to unscrew them, pretty sure the loose block was what causing all these for me, will run some games/benchmarks and see if i still get the reboot

 

[deeledee]

@NelizMastr tried to launch pubg twice and it rebooted my PC again, so i ran benchmark instead

 

everything is in stock when i ran the benchmarks, i did prime95, aida64(10mins), no crashes on both though, below are the graph

only thing left is to re-install the game or windows entirely which i hope not to do so.

 

[NelizMastr]

I doubt it's software. The GPU pulls a lot more power than the CPU, so that it survives a CPU stress test but not a GPU one says it all. Replace the PSU.

[deeledee]

1 minute ago, NelizMastr said:

I doubt it's software. The GPU pulls a lot more power than the CPU, so that it survives a CPU stress test but not a GPU one says it all. Replace the PSU.

alright i'll buy one tomorrow and test, if it really is my psu, i would never imagine a platinum rated one would give me problems just after 3 years 

[NelizMastr]

Just now, deeledee said:

alright i'll buy one tomorrow and test, if it really is my psu, i would never imagine a platinum rated one would give me problems just after 3 years 

Reminds me of my dad when he bought a Ford Escort in the 1980's, brand new off the lot. Engine lived for about 30 minutes to never start again. Moral of the story is, anything can break even if it's new or high quality. Bad luck and all.