Maniphest T96046

PSU shutdowning power of elemets on motherboard, causing power shutdown with system boot up
Closed, Archived

Assigned To
None
Authored By
Mark (Orgham)
Feb 26 2022, 8:26 PM
Tags
  • BF Blender
Subscribers
Mark (Orgham)
Pratik Borhade (PratikPB2123)
Ray Molenkamp (LazyDodo)

Description

System Information

Operating system:

release Windows 10 Pro для рабочих станций
version 21H2
date install 14.‎02.‎2022
assembly OS 19044.1566
interaction Windows Feature Experience Pack 120.2212.4170.0

()

Graphics card:





Blender Version



Worked: newest version not worked
problem appeared, look like 13.13.22

I have three versions of 2.80 plus blender, steam, upbge and this evening cleanly installed from the site in section D:

Short description of error

Power shutdowning, causing restart
Very same problem : https://blenderartists.org/t/why-blender-causing-my-pc-to-restart/1190036/208


About trying games i dont know, i have 800+ hours in mechwarrior online, 84 h dreadnought, 360 h planetside 2,
1300 h warframe (too max settings), city skylines (too max graphic setting).
In this year as i remember i no plaing at all

Long description of error
What I tried to solve the problem:

  1. reinstalled the operating system with the preservation of personal files
  2. several times completely removed all video card drivers and installed them from the official site
  3. manually reinstalled all drivers in the device manager, after reinstalling the operating system
  4. as shown above, I underlocked the video card, relative to its basic settings from the factory
  5. blocked the frequency of the processor in the BIOS, up to factory 3.20
  6. unsuccessfully, unsuccessfully but still updated the BIOS of the motherboard (more details below)
  7. I completely visually rechecked the entire system several times, every month I comb the dust everywhere with an ear stick, I often change the thermal paste of the video card and processor, respectively I check the video card with analysis, I check the power supply with lamp lighting and a large magnifying glass for the presence of oxide and or dust, I also often go through the vacuum cleaner the whole system, I have been carrying out all these procedures since the commissioning of my computer
  8. rechecked and reconnected all power cables including sata cables (swapped them); double-checked and rearranged and securely installed 4 ddr4 modules (the video card was propped up with a stick a long time ago so as not to squeeze out the port)
  9. tried to identify the sequence of the shutdown problem when working in the program, but it seems that it does not exist, but now the SHUTOFF HAPPENS IMMEDIATELY AT THE MOMENT OF NAVIGATION
  10. it’s my version of the video card that for some reason disables Google Chrome when hardware acceleration is enabled, it didn’t happen a year and a half earlier, I found an identical problem in a forum on some site; in general, I don’t use hardware acceleration and sit in edge (also disabled)
  11. I tried to check for a short circuit from the incoming signals of the rear connector of the board, but even so it turns off

I also tried to measure the values at different loads (below the factory ones) of the processor in the BIOS and the voltages in the same place, and also inside the operating system using the hwmonitor program, but it seems to me that even with a multimeter it is stupid to substitute as this is too exotic a problem, although everything can be
I don’t know, maybe I tried something else, but I don’t remember anymore

asrock b150 gamink k4
intel core i5 6500

I don’t know, maybe I tried something else, but I don’t remember anymore

Specifically when the shutdown occurs
as said earlier problem appears 13.13.22
Every version from it date of Blender and UPBGE (based on it), causing it.
The most interesting thing is that this began to happen, as it were, according to the knurled one, that is, in the first days I could use it for a while, then less and less and or in random order, now it’s just when navigating, that is, I can’t manage to do anything
In all the months of this trouble, only this program is causing the problem, not version 2.79b

important information
at some point, I decided to update the BIOS of the motherboard in order to exclude this as a last resort, and I really didn’t want to do this because there are no guarantees for success, and, accordingly, I damaged the BIOS and now I don’t have it after turning off the power for about 26 minutes provides power to the entire system, after it turns on. Despite this, I still tried to re-update all versions from p2.10 (factory) to the latest and vice versa and then back (with shutdowns and waits for 26 minutes), update updated but this did not solve the problem with the program. I will never update the bios again, it is too hardware-independent of programs.

What can't I do

  1. I can’t give it to a service center, it’s very expensive and I will never entrust my equipment to anyone, I don’t have really smart guys in my acquaintances, but just with qualified specialists, then I myself have a secondary education in PC maintenance
  2. to purchase equipment of similar performance, now the amount of my PC has multiplied by 5 just like a car, it is excessively expensive
  3. maybe something else

Also additional information

At the beginning of using the PC, I had an uninterruptible power supply, fixed an interruption 2 times over the years of use, then its power supply died
Since that time, the computer's power supply has been operating from the network as two phases.
There have been power outages many times and for about a year and a half they are gone.
Power surges, as for me, were only once at the moment of a strong thunderstorm that began abruptly (it seems that on the same evening, lightning struck the transformer in the Olympic Park in the cross wires, maybe I heard it through the computer speakers while sitting in silence)
As I wrote above, I check everything in the system from time to time, and none of the capacitors and fuses are swollen everywhere and does not look like damaged
power supply corsair 750 80 plus gold

Exact steps for others to reproduce the error

I don't know what else to write my head is already empty

Event Timeline

Mark (Orgham) created this task.Feb 26 2022, 8:26 PM
Pratik Borhade (PratikPB2123) added a subscriber: Pratik Borhade (PratikPB2123).Feb 27 2022, 11:38 AM
Pratik Borhade (PratikPB2123) changed the task status from Needs Triage to Needs Information from User.Feb 27 2022, 12:27 PM

Hi, thanks for the report.
Issue is vey likely related to your graphics driver.
Do you have the latest GPU driver as shown on their website: https://www.amd.com/en/support/graphics/amd-radeon-r9-series/amd-radeon-r9-300-series/amd-radeon-r9-380
Have you got this issue with other software or just Blender?
When you say it restarts, is that a normal restart or Blue screen error (BSOD) appear?
Are you able to run blender with your Integrated graphics?
I don't understand the screenshot you've attached (I'm not known to that language)
I don't think if PSU can really cause your PC to restart when running Blender.

Mark (Orgham) added a comment.Feb 27 2022, 1:43 PM
  1. Issue is vey likely related to your graphics driver.=


yes, on their site there is an installer for the latest and most suitable software for the video card, I have it installed

  1. Have you got this issue with other software or just Blender? =

as described in the problem, yes, I have probably not had anything for 2 years besides the program and a couple of non-resource-demanding applications; I did not arrange a stress test, since this is the same problematic action as with the BIOS

  1. When you say it restarts, is that a normal restart or Blue screen error (BSOD) appear?=

well, this is exactly as described in the title of this problem, how much I measured and checked it is the motherboard power off (I don’t know completely or not completely, my keyboard’s backlight power does not blink when the power is turned off), that is, as I understand it, a partial blackout occurs looks like a system restart

  1. Are you able to run blender with your Integrated graphics?=

I understand what you are getting at, no, I haven’t tried it, it’s built into the processor and these chips are certainly not adapted for the latest releases of a resource-intensive blender, I tried to install Intel software, but it’s raw and it’s not clear at what level it works, I’m very worried about the program burning this chip, especially after I saw posts on the Internet with damaged video cards when working with the program, of course, due to overload and not concisely installed load blocking, and the drivers, as I wrote from Intel, are still not clear, raw; in general, I can try and also can stay without a processor.

As for the power supply, I don’t want to be a fortune teller and I’ll say this, there are no visible damages, as well as oxides and dust.

Mark (Orgham) added a comment.Feb 27 2022, 1:48 PM

regarding the software of the video card, and based on the factory sticker, the program took the correct data, this is specifically about factory identifiers, sapphire nitro r9 380 4 GB

Pratik Borhade (PratikPB2123) added a comment.Feb 27 2022, 7:22 PM

Hi, thank you for the information.
I was considering GPU driver as a culprit.
Considering PSU, reboot probably occurs because hardware is not getting powered or drawing too much power?
I've to say I don't have enough knowledge about PSU, so unsure if I can be really helpful.
But from present information, issue may not be related to Blender.
I'll ask someone to have a look at this report.
Before that, can you upload output of minidump if it's generated after the reboot? (you can find it at C:\Windows\Minidump)

Mark (Orgham) added a comment.Feb 27 2022, 7:36 PM

Mark (Orgham) added a comment.EditedFeb 27 2022, 7:41 PM

also if you interest

in a critical error, it means that this is exactly the shutdown, as if I held down the power button, well, in principle, this is clear

Mark (Orgham) added a comment.Feb 27 2022, 7:56 PM

in order to find out, you need to fix the voltage of the power line of the motherboard at the moment of shutdown, it sounds like shooting yourself, although this is more than logical, I think program like hwmonitor be bad with it only in that dump.dat i sended, trouble

Pratik Borhade (PratikPB2123) added a comment.Feb 28 2022, 4:02 PM

Hi, please prefer to upload screenshot showing English language. Otherwise I can't understand the information.
Those minidump files points to different error codes: kernel-security-check-failure, memory-management, page-fault-in-nonpaged-area
Were those files generated after the crash due to Blender?
Just to be sure, did you install Blender from https://www.blender.org/download/?
The scenario described is too time consuming for us to track down, we require the bug reporter to narrow down the problem.

Ray Molenkamp (LazyDodo) closed this task as Archived.Feb 28 2022, 4:19 PM
Ray Molenkamp (LazyDodo) added a subscriber: Ray Molenkamp (LazyDodo).

Sorry to hear you are having issues running blender, the root cause appear to be faulty or overclocked hardware not something we can address as a bug in blender, so for those reasons i'll have to close your report.

Mark (Orgham) added a comment.Feb 28 2022, 4:35 PM
In T96046#1314853, @Ray Molenkamp (LazyDodo) wrote:

Sorry to hear you are having issues running blender, the root cause appear to be faulty or overclocked hardware not something we can address as a bug in blender, so for those reasons i'll have to close your report.

how did you determine that I overlocked the equipment?
and how did you determine that my equipment was damaged?

Mark (Orgham) added a comment.Feb 28 2022, 4:41 PM

I tried to find the answer here because it only happens in the blender program from your site, and I can’t understand how you write to me in English and don’t understand English, everything is clearly set on the screenshot of the blender version

Mark (Orgham) added a comment.Feb 28 2022, 4:45 PM

Ray Molenkamp (LazyDodo) added a comment.Feb 28 2022, 4:49 PM

from the dumps you posted

022122-6828-01.dmp PROCESS_NAME:  WmiPrvSE.exe ERROR_CODE: (NTSTATUS) 0xc0000409 - The system detected an overrun of a stack-based buffer in this application. This overrun could potentially allow a malicious user to gain control of this application.
021422-6937-01.dmp PROCESS_NAME:  wermgr.exe   FAILURE_BUCKET_ID:  MEMORY_CORRUPTION_ONE_BIT
021422-6343-01.dmp PROCESS_NAME:  svchost.exe  failure unknown appears to be memory corruption as well.

none of them are blender, and all of them indicate some form of memory corruption going on which is usually a clear indication of hardware or overclocking related problems, not something we can investigate from the blender side.

Mark (Orgham) added a comment.Feb 28 2022, 4:59 PM
In T96046#1314872, @Ray Molenkamp (LazyDodo) wrote:

from the dumps you posted

022122-6828-01.dmp PROCESS_NAME:  WmiPrvSE.exe ERROR_CODE: (NTSTATUS) 0xc0000409 - The system detected an overrun of a stack-based buffer in this application. This overrun could potentially allow a malicious user to gain control of this application.
021422-6937-01.dmp PROCESS_NAME:  wermgr.exe   FAILURE_BUCKET_ID:  MEMORY_CORRUPTION_ONE_BIT
021422-6343-01.dmp PROCESS_NAME:  svchost.exe  failure unknown appears to be memory corruption as well.

none of them are blender, and all of them indicate some form of memory corruption going on which is usually a clear indication of hardware or overclocking related problems, not something we can investigate from the blender side.

I never overclocked video cards, I don’t need it, based on my wounded screenshot, I lowered them by a revolution

and these are the basic settings taken by the AMD program from the BIOS identifier of the video card manufacturer

Mark (Orgham) added a comment.Feb 28 2022, 5:01 PM

and why did you close the post without advising me to do at least something? And thus I now have to write to you in a closed post?

Mark (Orgham) added a comment.Feb 28 2022, 5:11 PM
In T96046#1314872, @Ray Molenkamp (LazyDodo) wrote:

from the dumps you posted

022122-6828-01.dmp PROCESS_NAME:  WmiPrvSE.exe ERROR_CODE: (NTSTATUS) 0xc0000409 - The system detected an overrun of a stack-based buffer in this application. This overrun could potentially allow a malicious user to gain control of this application.
021422-6937-01.dmp PROCESS_NAME:  wermgr.exe   FAILURE_BUCKET_ID:  MEMORY_CORRUPTION_ONE_BIT
021422-6343-01.dmp PROCESS_NAME:  svchost.exe  failure unknown appears to be memory corruption as well.

none of them are blender, and all of them indicate some form of memory corruption going on which is usually a clear indication of hardware or overclocking related problems, not something we can investigate from the blender side.

so what should i do? All I see is that the memory is damaged, and for what reason was it damaged? That screenshot with the date 13/13/22 is my diary and I clearly know what and when happens, I have been working only with the program for several months and updated it 2 times using auto-update in steam which doesn’t really ask me, the current version means it’s necessary, as I will find out if this will damage the memory or not?

Ray Molenkamp (LazyDodo) added a comment.Feb 28 2022, 5:11 PM

We close a ticket when we determine there is nothing we can do about it from the blender side, do note that the problems reported in your dumps all appear to be RAM issues, not GPU issues.

Mark (Orgham) added a comment.Feb 28 2022, 5:17 PM

this is how i would write it first, thanks