| Page 1 of 1 |
demde
Posts: 6538
Location: Lake Karachay
|
 Posted: Sat, 18th Jul 2015 08:33 Post subject: Strange Hardware? problem |
|
 |
Have strange problem lately and can't figure out what is causing it.
When i was playing Witcher 3 about 2 weeks ago in about 1.5h session my pc restarted itself.First video signal is lost, sound from game stops right away and 2-3 sec after that restarts and then everything runs fine. Problem is this started to show in other games too and it happens quite random. Happened twice even in loading screens. Sometimes i can play 20-30 min (GTA V, MK X) and it doesn't show. Sometimes it takes 10-15 min and it restarts. Did some ram testing tool, cpu stress for 30 min, gpu stress for 20-30 min. Doesn't show. Changed drivers and its still doing it in games.
So i wonder what could it be? Doubt is windows. I'm suspecting PSU or GPU. Maybe PSU can't support the high requirement for a long time if it overheated in Witcher 3 and its stops working. If its GPU why its restarting and not giving artifacts then comes in mind. I'm sure GPU or CPU didn't overheated as max CPU temp i ever reached is 73C and CPU 68C.
PSU is CX750W Corsair, CPU i5 4670k, Gigabyte 760 GTX Windforce OC
When i'm not playing games i have no problems at all.
Edit:
I do notice that when i do stress tests - AIDA64 shows that +V12 value goes to V10.8-11
Also for some reason GPU core clock in theory should be 1150mhz boosted , but ingame it reaches 1264mhz and i never tried to overclock myself besides the fabric OC of the card.
AMD Ryzen 7 7800X3D / MSI Gaming X Slim GeForce RTX 4080 SUPER / ASRock B650 PRO RS / Arctic Liquid Freezer III 360 / G.Skill Trident Z5 Neo 2x16GB DDR5-6000 CL30 / Samsung 980 Pro 2 TB / SeaSonic VERTEX GX-850 850W Gold / Lian Li LANCOOL 216
|
|
| Back to top |
|
|
JBeckman
VIP Member
Posts: 35043
Location: Sweden
|
| Posted: Sat, 18th Jul 2015 08:48 Post subject: |
|
|
I would probably suspect the PSU if I encountered such a problem, if you have access to a spare GPU you could always see if the same thing happens with that and then that would exclude the GPU completely.
As a suggestion it might be useful to check the S.M.A.R.T status of the HDD as well just to rule out any corrupt sectors or such damage but it's probably not the problem here.
(I don't have the technical knowledge to say for sure but it almost seems as the GPU doesn't get enough power so it resets and then the system shuts down completely a few seconds later.)
Another suggestion would be to perhaps shut down the system and disconnect the power cable and then open the PSU and check for dust buildup though as far as I know it's pretty varied on how easy they are to open, if there's visible signs of damage that would also confirm where the problem lies. (GPU could be checked as well I suppose, easier to clean via compressed air instead of having to dismantle the cooler.)
Is there something in the event log on what caused a reboot?
(Might not have had time to log anything though if it just shut down unexpectedly instead of a BSOD.)
|
|
| Back to top |
|
|
tonizito
VIP Member
Posts: 51456
Location: Portugal, the shithole of Europe.
|
| Posted: Sat, 18th Jul 2015 09:06 Post subject: |
|
|
My money's on the PSU.
No easy way to test it IMO. If you use the CPU's "GPU" and nothing happens, well... still might be from either the PSU (but now with you not pulling enough power from it to show) or from the GPU (that is now removed from your system).
Maybe borrow a GPU/PSU from a friend or buy one at a store just in order to test it, return it and get your money back when done.
| boundle (thoughts on cracking AITD) wrote: | | i guess thouth if without a legit key the installation was rolling back we are all fucking then |
|
|
| Back to top |
|
|
demde
Posts: 6538
Location: Lake Karachay
|
|
| Back to top |
|
|
ragnarus
Posts: 686
Location: Somewhere in Warsaw Pact
|
| Posted: Sat, 18th Jul 2015 23:07 Post subject: ... |
|
|
For the future, never believe in voltages shown in programs or bios, they are almost always off, multimeter is the only way to know actual voltages.
i5 6600 + Silentiumpc Spartan Pro, Asus H170-plus mobo, HyperX 16 GB DDR4 2133MHz, Asus GTX 970 Strix OC, 1TB Seagate Barracuda, BeQuiet L8 500W, Zalman Z9 case, crap monitor 
|
|
| Back to top |
|
|
3E74
Posts: 2559
Location: feels wrong
|
| Posted: Sat, 18th Jul 2015 23:30 Post subject: |
|
|
hmmm,
im not shure, but i had a similar problem with witcher 3..
it frooze my whole system and i had to hard reboot or it crashed my whole PC..
times where random, it happened after 10, 20 or more minutes.
first i thought its the temps, but nope, they where all green..
i tried alot, and read on an NV forum that the new drivers did something to the voltage, or i dont know..
and i set teh gpu via afterburner to +20 voltage, no OC nothing else, and i put the fan on 80% so it stays under 70C°.
since then i never had an issue again..no more system freezes where i had to use teh hard reboot or bsod´s etc..
This also fixed my BF4 issue with teh market map..
so i made a game profile with afterburner with +10 - +20 overvoltage..
not shure if this helps you, but you have kinda the same card.. i have an gtx670, we have both kepler and im shure you are using teh same driver, or at least teh witcher 3 driver..
also, im on an 550w (Corsair) PSU, with an q9550 Oced to allmost 4ghz, the ram banks are full. so i would think that your PSU is ok.. IF it isnt an cheapo one..
EDIT: i just read ur post again, you also have an corsair and its an 750W. naah, its not teh PSU i think.. you need less wattage then me i think. you have newer nm and such and should be more effizient.
hope that helps 
greetz
edit:
i also cleaned my whole PC at the same time and dusted teh PSU completly.. but i dont think that was related^^
as i think of it, before i upped my gpu voltage, this Freeze/syscrash only happened in scenarios/games where my GPU was hitting 99% usage, so in games/scenarios where teh gpu was hitting its limits..
wich could explain why it worked when i upped the volatge a bit..
well, id say its worth a try. its within the sve limits and when your cool enough it shoudnt be a prob at all..
..:: Life - A sexually transmitted disease which always ends in death. There is currently no known cure::.. 
|
|
| Back to top |
|
|
demde
Posts: 6538
Location: Lake Karachay
|
| Posted: Sun, 19th Jul 2015 06:38 Post subject: |
|
|
Thanks will try that if the problem shows again.
But sessions with the test PSU i took look good. Since i put it i played 1h sessions for W3, GTAV, MKX and my PC didn't restart.
Will do another session later today, but looks like is my PSU, causing the problems.
If problem doesn't show before i return the test PSU i will surely give mine another chance as maybe cables re-plug or something might be the case, but if it returns then surely PSU fails when it have to give a bit more juice for a long time.
IMO if it was GPU it would've failed in W3 1 hour torture.
AMD Ryzen 7 7800X3D / MSI Gaming X Slim GeForce RTX 4080 SUPER / ASRock B650 PRO RS / Arctic Liquid Freezer III 360 / G.Skill Trident Z5 Neo 2x16GB DDR5-6000 CL30 / Samsung 980 Pro 2 TB / SeaSonic VERTEX GX-850 850W Gold / Lian Li LANCOOL 216
|
|
| Back to top |
|
|
|
|
| Posted: Sun, 19th Jul 2015 21:29 Post subject: |
|
|
Demde it's a PSU problem. Check your Windows event log. I had the exact same problem some months ago and I replaced the PSU. You don't have to look any further. Buy a new PSU.
Q6600 OC 3.0ghz, MSI R9 280x Gaming Edition, 7gb ram, Asus P5KC
|
|
| Back to top |
|
|
demde
Posts: 6538
Location: Lake Karachay
|
| Posted: Mon, 20th Jul 2015 06:24 Post subject: |
|
|
Yea looks like it is. Another 1.5h W3 session with the test PSU didn't gave any problem.
I really hope i won't have to buy PSU though. They have to accept warranty.
AMD Ryzen 7 7800X3D / MSI Gaming X Slim GeForce RTX 4080 SUPER / ASRock B650 PRO RS / Arctic Liquid Freezer III 360 / G.Skill Trident Z5 Neo 2x16GB DDR5-6000 CL30 / Samsung 980 Pro 2 TB / SeaSonic VERTEX GX-850 850W Gold / Lian Li LANCOOL 216
|
|
| Back to top |
|
|
|
|
| Posted: Mon, 20th Jul 2015 10:57 Post subject: |
|
|
| demde wrote: | Yea looks like it is. Another 1.5h W3 session with the test PSU didn't gave any problem.
I really hope i won't have to buy PSU though. They have to accept warranty. |
Ofc they have to accept it but it depends on the store too. The PSU is faulty but they have to test it in a gaming environment or else it's very difficult to figure out the problem with it. Testing with Prime or anything like will not reboot the pc.
Run some GTA V sessions also or even better some PCars ones. If there's a problem PCars will reboot your system in a couple of minutes.
Q6600 OC 3.0ghz, MSI R9 280x Gaming Edition, 7gb ram, Asus P5KC
|
|
| Back to top |
|
|
Slizza
Posts: 2345
Location: Bulgaria
|
| Posted: Mon, 20th Jul 2015 12:30 Post subject: |
|
|
If running system stress tests doesn't make it happen then you would question the PSU conclusion as these tests should work the PSU harder than any games.
Your GPU clock is normal. That's Nvidia GPU boost.
Corsair 750D :: 750W DPS-G:: Asus x370 PRO :: R7 1800X ::16gb DDR4 :: GTX 1070::525gb SSD::Coolermaster 240MM AIO::
|
|
| Back to top |
|
|
demde
Posts: 6538
Location: Lake Karachay
|
| Posted: Mon, 20th Jul 2015 12:41 Post subject: |
|
|
It might be placebo or something or re-plug of the cables solved the problem. Maybe it had bad connection or something also changed the GPU additional power supply ones and used another cable for HDD, SSD. I did small sessions after i returned the test PSU and connected mine again and PC didn't restart yet.
Will do some more later as now ambient temps are quite high and i can't stand in the room when the PC adds to the temp. Hopefully problem is gone i won't need to replace anything.
Installing PCars game to test it out.
AMD Ryzen 7 7800X3D / MSI Gaming X Slim GeForce RTX 4080 SUPER / ASRock B650 PRO RS / Arctic Liquid Freezer III 360 / G.Skill Trident Z5 Neo 2x16GB DDR5-6000 CL30 / Samsung 980 Pro 2 TB / SeaSonic VERTEX GX-850 850W Gold / Lian Li LANCOOL 216
|
|
| Back to top |
|
|
Slizza
Posts: 2345
Location: Bulgaria
|
|
| Back to top |
|
|
|
|
| Posted: Mon, 20th Jul 2015 14:05 Post subject: |
|
|
| Slizza wrote: | If running system stress tests doesn't make it happen then you would question the PSU conclusion as these tests should work the PSU harder than any games.
Your GPU clock is normal. That's Nvidia GPU boost. |
Nope. When I installed the R9 280x on my old PSU I had system restarts in gaming as Demde. I ran a stress test for 24h and the system passed without a problem.
Q6600 OC 3.0ghz, MSI R9 280x Gaming Edition, 7gb ram, Asus P5KC
|
|
| Back to top |
|
|
|
|
|
| Back to top |
|
|
demde
Posts: 6538
Location: Lake Karachay
|
| Posted: Mon, 20th Jul 2015 14:47 Post subject: |
|
|
Actually i suspect exactly the power cable that is connected to the GPU. One of them had some weird white stuff over it so i used another one this time.
AMD Ryzen 7 7800X3D / MSI Gaming X Slim GeForce RTX 4080 SUPER / ASRock B650 PRO RS / Arctic Liquid Freezer III 360 / G.Skill Trident Z5 Neo 2x16GB DDR5-6000 CL30 / Samsung 980 Pro 2 TB / SeaSonic VERTEX GX-850 850W Gold / Lian Li LANCOOL 216
|
|
| Back to top |
|
|
|
|
| Posted: Mon, 20th Jul 2015 15:46 Post subject: |
|
|
White stuff on wire is usually corrosion.
|
|
| Back to top |
|
|
demde
Posts: 6538
Location: Lake Karachay
|
| Posted: Mon, 20th Jul 2015 16:12 Post subject: |
|
|
Not on the wire, but on the black plastic that goes into the socket on the GPU. Not sure if it was the one that i was using before, but still i avoided using it now.
AMD Ryzen 7 7800X3D / MSI Gaming X Slim GeForce RTX 4080 SUPER / ASRock B650 PRO RS / Arctic Liquid Freezer III 360 / G.Skill Trident Z5 Neo 2x16GB DDR5-6000 CL30 / Samsung 980 Pro 2 TB / SeaSonic VERTEX GX-850 850W Gold / Lian Li LANCOOL 216
|
|
| Back to top |
|
|
tonizito
VIP Member
Posts: 51456
Location: Portugal, the shithole of Europe.
|
| Posted: Mon, 20th Jul 2015 16:27 Post subject: |
|
|
| demde wrote: | | Not on the wire, but on the black plastic that goes into the socket in the GPU. Not sure if it was the one that i was using before, but still i avoided using it now. |
Should had tested that before sticking the new one in. Now you might have to put the old PSU back in, redo the internal connections to check if it wasn't from one of those too, etc etc.
| boundle (thoughts on cracking AITD) wrote: | | i guess thouth if without a legit key the installation was rolling back we are all fucking then |
|
|
| Back to top |
|
|
demde
Posts: 6538
Location: Lake Karachay
|
| Posted: Mon, 20th Jul 2015 17:54 Post subject: |
|
|
What do you mean by old one Tony? Mine or the test PSU i took from the store?
Currently testing mine again. Did Pcars test and indeed is killing the gpu like no other game i tried. Reached 76C even.(Even in 2h W3 sessions i never had more than 73C before). Was afraid to play too much, but still 15 min on track and another 3-4 min in intro video (which was also quite demanding) and settings tweaks with no restarts. Guess this is good news for now.
AMD Ryzen 7 7800X3D / MSI Gaming X Slim GeForce RTX 4080 SUPER / ASRock B650 PRO RS / Arctic Liquid Freezer III 360 / G.Skill Trident Z5 Neo 2x16GB DDR5-6000 CL30 / Samsung 980 Pro 2 TB / SeaSonic VERTEX GX-850 850W Gold / Lian Li LANCOOL 216
|
|
| Back to top |
|
|
|
|
| Posted: Mon, 20th Jul 2015 18:14 Post subject: |
|
|
| demde wrote: | What do you mean by old one Tony? Mine or the test PSU i took from the store?
Currently testing mine again. Did Pcars test and indeed is killing the gpu like no other game i tried. Reached 76C even.(Even in 2h W3 sessions i never had more than 73C before). Was afraid to play too much, but still 15 min on track and another 3-4 min in intro video (which was also quite demanding) and settings tweaks with no restarts. Guess this is good news for now. |
Do you have a custom fan profile or automatic? 73C for W3 isn't too high? The maximum I get is 61C with the automatic fan setting.
Also run a 1h session in PCars to be sure everything is fine eventually.
Q6600 OC 3.0ghz, MSI R9 280x Gaming Edition, 7gb ram, Asus P5KC
|
|
| Back to top |
|
|
demde
Posts: 6538
Location: Lake Karachay
|
| Posted: Mon, 20th Jul 2015 18:47 Post subject: |
|
|
Automatic setting and i'm getting over 70C only in warm days. In may early june i was 65-67 avg.
But IMO 70-72C shouldn't be problem for kepler.
Anyways i think even with tests i did so far i can exclude GPU. If its really fucked up i doubt it would've survived W3 and Pcars sessions.
AMD Ryzen 7 7800X3D / MSI Gaming X Slim GeForce RTX 4080 SUPER / ASRock B650 PRO RS / Arctic Liquid Freezer III 360 / G.Skill Trident Z5 Neo 2x16GB DDR5-6000 CL30 / Samsung 980 Pro 2 TB / SeaSonic VERTEX GX-850 850W Gold / Lian Li LANCOOL 216
|
|
| Back to top |
|
|
tonizito
VIP Member
Posts: 51456
Location: Portugal, the shithole of Europe.
|
| Posted: Mon, 20th Jul 2015 18:50 Post subject: |
|
|
| demde wrote: | What do you mean by old one Tony? Mine or the test PSU i took from the store?
Currently testing mine again. Did Pcars test and indeed is killing the gpu like no other game i tried. Reached 76C even.(Even in 2h W3 sessions i never had more than 73C before). Was afraid to play too much, but still 15 min on track and another 3-4 min in intro video (which was also quite demanding) and settings tweaks with no restarts. Guess this is good news for now. |
Your CX750.
So it seems like it might have been related to the power cable after all? If so good news, might spare you more hassle beyond simply having to return the Fortron.
Also had some issues with one of those, but it was even worse, it made the rectifier go off 
Butchered that power cable with a knife and threw it out  
| boundle (thoughts on cracking AITD) wrote: | | i guess thouth if without a legit key the installation was rolling back we are all fucking then |
|
|
| Back to top |
|
|
| Page 1 of 1 |
All times are GMT + 1 Hour |
