Question Pc hangs, freezes only on gaming

jamesmak

Reputable
Jan 19, 2018
23
0
4,510
0
Hello, I have problems with my pc which started a couple days ago whilist gaming. The pc runs fine the first 1 or 2 games and then freezes on black screen or restarts itself, without a BSOD just right back to post. I tried replicating the problem with memtest, prime95, furmark and occt (for monitoring) to see if hardware is involved in this problem. memtest prime and furmark all run together and the pc didnt hang or freeze one time. I uninstalled with ddu my gpu's driver and reinstalled it, windows are fully updated, and I don't know what the problem is. Looking at the event viewer after the crash I see event id: 41 The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly. The psu seems to be running fine, I did install a new AIO and thinking that maybe the wiring caused the problem I doubled checked them and reseated them, I run both chkdsk and sfc /sacnnow (returned with corrupted files and fixed them), but the problem continues. Any ideas?

System Specifications:

Mobo: GIGABYTE H110M ds2 rev1
GPU: NVIDIA GeForce RTX 2070 SUPER
PSU: CORSAIR rx 1000
CPU: INTEL i5 6400 2.7 Ghz
RAM: TRIDENT Z 1066mhz 8gb
SATA1: KINGSTON ssd
SATA2: TOSHIBA hdd
OS: Windows 10 Pro 64bit
 
Last edited:

Colif

Win 10 Master
Moderator
Cleaned PC recently? what temps are you getting?

its shouldn't be a power issue with that psu, slight overkill

you have run all the tests I would suggest.

if the only change you made was AIO and it worked fine before, we have to assume its part of problem. What make/model is it? what temps CPU get to in game?

you could run this on CPU - https://downloadcenter.intel.com/download/19792/Intel-Processor-Diagnostic-Tool
 

jamesmak

Reputable
Jan 19, 2018
23
0
4,510
0
Cleaned PC recently? what temps are you getting?

its shouldn't be a power issue with that psu, slight overkill

you have run all the tests I would suggest.

if the only change you made was AIO and it worked fine before, we have to assume its part of problem. What make/model is it? what temps CPU get to in game?

you could run this on CPU - https://downloadcenter.intel.com/download/19792/Intel-Processor-Diagnostic-Tool
So first things first. The cange I made to the AIO was refilling the tank with new liquid, my AIO is from EKWB i just removed all the wiring so I can jumstart the psu with a piece of wire in the 24pin connector so I could get the pump going, to flush the old liquid from the system, but as I said I reseated every wire in my pc properly. Psu is a bit of overkill cause I was thinking upgrades back then. The temps Im getting are 45 - 47 for cpu and 65-70 for GPU (but I am assuming that is normal being in a hot climate and stress testing didnt shut down the pc). Intel Diagnostic Tool got a pass on everything on my cpu.
 

jamesmak

Reputable
Jan 19, 2018
23
0
4,510
0
alas, nothing overly useful in there. was worth a try. Nothing that indicates any particular hardware

is that 1 8gb stick? how many times did you run memtest?
Yes my mobo runs 2 ram slots and one of them is broken (it was restarting the pc before post) so I removed 1 ram stick from my pc but that was 2 years ago and it was running fine afterwards. It's a 2 dimm x8gb each and now it runs only 1 dimm
Edit: I run memtest twice for 30 minutes.
Edit2: it crashed again right now 4:33pm and I was not using it it was idle in this forum and it turned into a black screen, it's like the gpu disconnects because my monitor gives me a signal not found
 
Last edited:

Colif

Win 10 Master
Moderator
Maybe we need to track sensors on board and see if anything obvious happens when it crashes
  1. Download https://www.hwinfo.com/download/
  2. when you start it, click the box next to sensors only and click run
  3. in next screen, along bottom row are a series of buttons
  4. Click on the button to the left of the clock, it will show "Logging start" when you hover mouse over it
  5. This opens file explorer, use this to create a log file in a location you can find again
  6. run logging when you play the games that cause problem.
  7. Makes sense to start a new log each time, makes it shorter.
  8. Results are in CSV format, can be read in excel or Google docs,
you can upload results and I can ask someone to look at them, as I am not really great at working out cause from them
 

jamesmak

Reputable
Jan 19, 2018
23
0
4,510
0
So I got crashed again I made the log file as you said, took some time though, here it is. I also included the last 2 logs from the event viewer Application and System as well for any reason you might find it usefull.
 

Colif

Win 10 Master
Moderator
i have asked others to help, you seem to have done everything I can think of.

Someone else might look through the reports.
Event viewer isn't showing anything really. Appears to me Windows isn't the cause here.
 

jamesmak

Reputable
Jan 19, 2018
23
0
4,510
0
i have asked others to help, you seem to have done everything I can think of.

Someone else might look through the reports.
Event viewer isn't showing anything really. Appears to me Windows isn't the cause here.
Thank you. I expect to hear from you if you can find anything. Thank you again for your time
 

Colif

Win 10 Master
Moderator
I don't see anything that stands out in the HWINFO report
well, its not power as its the same throughout
it's like the gpu disconnects because my monitor gives me a signal not found
does anything show in reliability history at the time the screen goes off? its likely to show a hardware error as it sees the restart as a hardware error.

wonder if it is gpu, but then you ran furmark and got no errors.

sysnative shows explorer crashes a lot, I am not sure if its connected as its just an event in windows error reporting
 

jamesmak

Reputable
Jan 19, 2018
23
0
4,510
0
Hello again, I run reliability history and searched the last couple of days when I started running the tests and it came up with some bugchecks and hardware error of LiveKernelEvent here is the imgur with the technical details:
and here is the XML file:
 

Colif

Win 10 Master
Moderator
SATA1: KINGSTON ssd
SATA2: TOSHIBA hdd
I wonder if its the drives. would explain the explorer crashes maybe. And perhaps freezes
i don't know if this has any diagnostics - https://www.kingston.com/en/support/technical/ssdmanager
toshiba don't seem to have a tool for their drives

you can use crystaldiskinfo to look at its SMART scores. you don't have to run any tests, it should be auto scan
it should look something like this


only showed photo as there are 2 apps on this page - you don't need crystaldiskmark, its more a benchmark - https://crystalmark.info/en/software/crystaldiskinfo/
click one of the blue icons on top of page.
 

jamesmak

Reputable
Jan 19, 2018
23
0
4,510
0
So, a bit of an update, I searched through the nVidia forums and it seems that the last update 471.11 has a problem on many pcs, so I decided to roll back on the previous driver. The game that was giving me crashes on 1 or 2 games worked until I played 4 matches. A strange thing in the reliability monitor is that each time the pc crashes is that razer synapse moments before the black screen stops working. I uninstalled it as well. A little more google search showed me that windows has a tdr mechanism? (im not sure how it's called properly) and I think that this might be the issue because it seems that the gpu hangs and windows tries to unfreeze itself so therefore closes because it can't. What shall I do with this information, I'm kinda stuck. Also I run crystaldiskinfo and it says that both my drives are good but the ssd (windows partition) is on 70% health.
 

Colif

Win 10 Master
Moderator
tdr = Timeout Detection & Recovery -
TDR stands for Timeout Detection and Recovery. This is a feature of the Windows operating system which detects response problems from a graphics card, and recovers to a functional desktop by resetting the card. If the operating system does not receive a response from a graphics card within a certain amount of time (default is 2 seconds), the operating system resets the graphics card.
http://developer.download.nvidia.com/NsightVisualStudio/2.2/Documentation/UserGuide/HTML/Content/Timeout_Detection_Recovery.htm

I don't see that many tdr errors with NVidia cards, mostly AMD.


I don't see any crashes in Event Viewer that mention GPU at all. At least, none of the crashes on the 4th are GPU.

Might want to think about a new ssd, 70% isn't bad but ssd don't really give a warning when they stop working, they just die. or can. So maybe replace it at some stage and retire it to less stressing duties like storage.

i can't do anything with the reliability history thing, it seems there is no viewer for it. You can buy software to do it but well, I hardly ever ask for it so that would be a waste.

looking at your events, I can see problems on the ssd just before 1 BSOD

Some of your crashes aren't creating dump files.

can you search for command prompt
right click the app result and run as admin
type diskpart and press enter
type list vol and press enter

take a screenshot and upload to image sharing website and show link.

Windows talks about volume numbers, I want to see what drive it is. Volume C is healthy according to event viewer
 

Colif

Win 10 Master
Moderator
judging from that, D is a 1tb hdd - 931gb is about right for a formatted drive
C is the ssd
vol 2 & 3 are likely on the ssd, given their purposes: Recovery & Boot partitions

its odd. you have 4 volumes but some of the events same volume 5 is okay... unless sometimes you have a USB drive plugged in.

120gb isn't very big anymore and I can't tell how much free space you have, but I would think about replacing it with at least a 250gb and try to keep 10% free space for both error correction and to maintain its speed.

BSOD seem to be pointing at Nvidia if thats what it says. I am just side tracked looking at ssd :)
which Nvidia drivers are you on? from nvidia or did you run windows update after running ddu and use the older drivers.
 

jamesmak

Reputable
Jan 19, 2018
23
0
4,510
0
Sorry for the late response
BSOD seem to be pointing at Nvidia if thats what it says. I am just side tracked looking at ssd :)
which Nvidia drivers are you on? from nvidia or did you run windows update after running ddu and use the older drivers.
Currently I rolled back to 461.92, the oldest one I found and remember it was fine. No I did not run windows update after I rolled back, because I didn't think it would have much of an impact. I will have the ssd replaced at some point which one do you suggest, cause I'm not good at ssds as much as other pc parts :p

Edit: So i run windows update and it says the system is updated even with the old drivers. is there anything I should do? :unsure::unsure:
 

Colif

Win 10 Master
Moderator
running windows update after removing drivers in safe mode just lets Windows find GPU and use older Nvidia drivers. I don't know how old they are as I use the newest from Nvidia.

you don't need to do it now, wait and see how the current ones go.

I would get a Samsung ssd, but I might be biased. So far both I have bought still work. Western Digital and Seagate are pretty good as well.
 

jamesmak

Reputable
Jan 19, 2018
23
0
4,510
0
Pc works fine for now, I opened it I reseated the cables again I topped up the liquid in the tank and it seems to be working fine now without closing, I don't know what happened :unsure:. I will replace the ssd as you said but for now I want to see how it runs.
 

ASK THE COMMUNITY

TRENDING THREADS