Question 2 new games, max settings, video turned off, video turns off at windows password screen ?

[gslonesoldier84]

I installed Arma 3 and Battlefield Bad Company 2. I had been playing Company of Heroes with no problems.

I played Arma 3 at super-high settings with long draw distances too.

Mid-game like 20 minutes into Arma 3 the video turned off but I could hear audio.

Since then the video switches off with the video card at the windows password screen like 98% of the time (except a few rare times when I let it sit for a long while and then it seemed to get past the windows password screen for like 2-3 minutes before switching video off again).

I turned the video to CPU graphics in the BIOS and I can be on the desktop for a long time now and it seems ok. But it will crash or not load the game even in company of heroes often and I haven't tried Arma 3 or Battlefield Bad Company 2 since.

If I can figure out the problem before then. I would be helped by having the video card working before Tuesday this week because I use Revit to make renders. With my workflow, I need the quick renders the video card was making possible. Otherwise the renders will take a long time to complete. It's still way faster without the GPU than it is from my laptop though!

I did a clean install of windows and reinstalled the video drivers for the GPU even though it is running video from the CPU right now.

So, that's the situation. What should I do? How can I troubleshoot further?

The system is:

PSU = Corsair CX500M
OS = Windows 10 Home
CPU = Intel Core i5-4460 CPU at 3.2 ghz 4cpus
RAM = 16GB (two RAM chips...one is Spektek and the other is Offtek)
GPU = GTX 780Ti 3GB
Motherboard = ASUS (says on it with directX 11 support ...I have DirectX 12 version is this a problem?

Further info that is relevant is that I worked out how to connect the power cable by figuring it out from guides/etc. It all worked great for a long while with many renders put out and games of Company of Heroes played. It went pear-shaped when I installed Arma 3 and Battlefield Bad Company 2 and hasn't been ok since. (been few days now)

+1 Also the fan goes really loud when it crashes in company of heroes now from the CPU with the GPU disabled (but when I switch off the advanced video settings it will be ok for company of heroes running on the CPU not the GPU)

Please help!

 

[Karadjgne]

Wierd the installer did that, I had to install an older copy of 9.0c to get a couple of older games to run. Because the directx files are different, literally you'd see dx3d9.dll and dx3d11.dll etc, they don't overwrite, but sit parallel with the newer files. Games etc only access whatever files ate needed, not all the files.

If the pcie slot was damaged, there'd be no consistency, the gpu would fail, not wait until software is started then crash. As far as I can tell, it's a software/firmware conflict, most likely with a recent windows version update, not a hardware issue.

 

[Groveling_Wyrm]

Another thing to check.....Go to

Settings - System (You should be on display at this time) - Graphics settings (At the bottom) - Hardware Based GPU scheduling

Ensure that this is set to ON.

 

[gslonesoldier84]

I will pull the sale advert and will try again to fix it. I'll post the results within a few days.

 

[MJS WARLORD]

Hi i see a lot of people are helping you but i am curious as to why you get stuck at the log in screen.
Just out of curiosity , the next time you get to the desktop type netplwiz into the "run" box and follow the prompts.
Now restart the pc and if you followed the prompts correctly you should get past the log in screen ..... if you did get to desktop without having to use the microsoft login screen did anything else odd happen.

 

[gslonesoldier84]

Hi i see a lot of people are helping you but i am curious as to why you get stuck at the log in screen.
Just out of curiosity , the next time you get to the desktop type netplwiz into the "run" box and follow the prompts.
Now restart the pc and if you followed the prompts correctly you should get past the log in screen ..... if you did get to desktop without having to use the microsoft login screen did anything else odd happen.

When I did this, there were no prompts. It just opened a User Account window with tabs and content with properties to change and add to. I have closed it without making any changes. What prompts may I have seen when typing that?

 

[gslonesoldier84]

Another thing to check.....Go to

Settings - System (You should be on display at this time) - Graphics settings (At the bottom) - Hardware Based GPU scheduling

Ensure that this is set to ON.

It wasn't listed. It only says that no preference is selected and there are no options to change. So I added NVIDIA Control Panel as "choose an app to set preference" and selected the 780ti as the option for high performance just to see if it does anything. I checked Google to find where it would have the Hardware Based GPU scheduling and it is this window. And it's not listed there as an option to change.

 

[helper800]

This may sound redundant but download and install DDU. Then go download the latest standalone drivers for the 780 ti. Next boot into safemode for Win10. You can do that by typing in advanced startup options in the windows search function. Go through the prompts to get into safemode. Then run DDU to get rid of the current Nvidia drivers. Restart the computer after that finishes and then install the new drivers you just downloaded.

If this does not fix your issue related to not being able to get into the normal windows desktop with your GPU as the display device then I would suspect you have a hardware issue somewhere. This conclusion I have come from the fact that you reinstalled windows and this step would eliminate driver issues as being the cause of not being able to get into windows.

What are the operating temperatures of your GPU and CPU? Use this program to check if you don't have a means too. Anything over 85C is suspect as potential issue causing problems. If it starts going into the 90s for either the GPU or CPU the computer may shut itself down to protect itself.

 

[gslonesoldier84]

This may sound redundant but download and install DDU. Then go download the latest standalone drivers for the 780 ti. Next boot into safemode for Win10. You can do that by typing in advanced startup options in the windows search function. Go through the prompts to get into safemode. Then run DDU to get rid of the current Nvidia drivers. Restart the computer after that finishes and then install the new drivers you just downloaded.

If this does not fix your issue related to not being able to get into the normal windows desktop with your GPU as the display device then I would suspect you have a hardware issue somewhere. This conclusion I have come from the fact that you reinstalled windows and this step would eliminate driver issues as being the cause of not being able to get into windows.

What are the operating temperatures of your GPU and CPU? Use this program to check if you don't have a means too. Anything over 85C is suspect as potential issue causing problems. If it starts going into the 90s for either the GPU or CPU the computer may shut itself down to protect itself.

As far as installed the linked drivers. They were a newer version than the ones I downloaded a little while back.

While they were installing the fans went loud. I am now trying to set the video to run from the GPU in the bios now.

...Third restart and it's not doing anything for video at all now. I can't even get into the bios now to change the video processor back to CPU from PCIe. Or at least I can't see anything on the screen to that effect.

... sorry I had the video cable in the VGA slot. It's at the login screen now.

Pressing escape...
Typing pin...

And....video switched off again.

...Changed it to Auto instead of PCIe and it switched off the video as soon as I pressed escape and then the fan went loud.

....changed back to CPU graphics and it is ag the desktop fine..

....the HWiNFO program is running but I don't see a temperature listed. I will check in more detail soon and try to look it up it might be turned off in the settings maybe? the GPU isn't running right now anyway though.

 

[gslonesoldier84]

General information:

I just updated/installed these drivers via windows update:

Trying with it running from GPU again
....now it only shows the windows blue spinning dots screen for a flash of a second before it switches off video. And fan went loud right after too two times in a row now

Got furthest with gpu as video output at 20:42. It spun the white dots after typing the password for 2 seconds.
Then video went off but I heard a windows noise after even though there was no video.
Then at 20:45 it went only as far as flashing the login screen then going black and turning on fans loudly.


Some event viewer entries from around 20:42:
------------------------
The following boot-start or system-start driver(s) did not load:
dam
EhStorClass
Gemma
------------------------
The application-specific permission settings do not grant Local Activation permission for the COM Server application with CLSID
{2593F8B9-4EAF-457C-B68A-50F6B8EA6B54}
and APPID
{15C20B67-12E7-4BB6-92BB-7AFF07997402}
to the user DESKTOP-UGTD3G3\gurpr SID (S-1-5-21-2375795573-99323447-2641846060-1001) from address LocalHost (Using LRPC) running in the application container Unavailable SID (Unavailable). This security permission can be modified using the Component Services administrative tool.
-----------------------------
The winlogon notification subscriber <SessionEnv> was unavailable to handle a critical notification event.
-----------------------------
The winlogon notification subscriber <SessionEnv> was unavailable to handle a notification event.
----------------------------
The Desktop Window Manager has registered the session port.
------------------------
ERROR
Faulting application name: dwm.exe, version: 10.0.19041.746, time stamp: 0x6be51595
Faulting module name: KERNELBASE.dll, version: 10.0.19041.1202, time stamp: 0xc9db1934
Exception code: 0xe0464645
Fault offset: 0x000000000010be3e
Faulting process ID: 0x23d4
Faulting application start time: 0x01d7b17c47d52a6e
Faulting application path: C:\Windows\system32\dwm.exe
Faulting module path: C:\Windows\System32\KERNELBASE.dll
Report ID: 64f00ea0-3446-410a-9256-3beff8591e5a
Faulting package full name:
Faulting package-relative application ID:
------------------------------
Fault bucket , type 0
Event Name: APPCRASH
Response: Not available
Cab Id: 0

Problem signature:
P1: dwm.exe
P2: 10.0.19041.746
P3: 6be51595
P4: KERNELBASE.dll
P5: 10.0.19041.1202
P6: c9db1934
P7: e0464645
P8: 000000000010be3e
P9:
P10:

Attached files:

These files may be available here:
\\?\C:\ProgramData\Microsoft\Windows\WER\ReportQueue\AppCrash_dwm.exe_e1b212a7af91fba76b3eb4ebe214acef5b75dfef_c2f10bab_e4ff8f44-d207-41de-bd84-83b758c999e5

Analysis symbol:
Rechecking for solution: 0
Report Id: 64f00ea0-3446-410a-9256-3beff8591e5a
Report Status: 6
Hashed bucket:
Cab Guid: 0
------------------------------
warning
The Desktop Window Manager process has exited. (Process exit code: 0x000000ff, Restart count: 1, Primary display device ID: NVIDIA GeForce GTX 780 Ti)
------------------------------------
error
Faulting application name: dwm.exe, version: 10.0.19041.746, time stamp: 0x6be51595
Faulting module name: KERNELBASE.dll, version: 10.0.19041.1202, time stamp: 0xc9db1934
Exception code: 0xe0464645
Fault offset: 0x000000000010be3e
Faulting process ID: 0x261c
Faulting application start time: 0x01d7b17c499cfb95
Faulting application path: C:\Windows\system32\dwm.exe
Faulting module path: C:\Windows\System32\KERNELBASE.dll
Report ID: 35acc11e-a1e8-4120-b195-f39f923a54fd
Faulting package full name:
Faulting package-relative application ID:
-------------------------------
Fault bucket , type 0
Event Name: APPCRASH
Response: Not available
Cab Id: 0

Problem signature:
P1: dwm.exe
P2: 10.0.19041.746
P3: 6be51595
P4: KERNELBASE.dll
P5: 10.0.19041.1202
P6: c9db1934
P7: e0464645
P8: 000000000010be3e
P9:
P10:

Attached files:

These files may be available here:
\\?\C:\ProgramData\Microsoft\Windows\WER\ReportQueue\AppCrash_dwm.exe_e1b212a7af91fba76b3eb4ebe214acef5b75dfef_c2f10bab_98bb5f3f-d231-4ae9-b939-eba53a873c84

Analysis symbol:
Rechecking for solution: 0
Report Id: 35acc11e-a1e8-4120-b195-f39f923a54fd
Report Status: 6
Hashed bucket:
Cab Guid: 0
----------------------------
warning
The Desktop Window Manager process has exited. (Process exit code: 0x000000ff, Restart count: 2, Primary display device ID: NVIDIA GeForce GTX 780 Ti)
------------------------
error
Faulting application name: dwm.exe, version: 10.0.19041.746, time stamp: 0x6be51595
Faulting module name: KERNELBASE.dll, version: 10.0.19041.1202, time stamp: 0xc9db1934
Exception code: 0xe0464645
Fault offset: 0x000000000010be3e
Faulting process ID: 0x24e4
Faulting application start time: 0x01d7b17c4b0c7008
Faulting application path: C:\Windows\system32\dwm.exe
Faulting module path: C:\Windows\System32\KERNELBASE.dll
Report ID: 3dce28d4-8050-4599-b43b-eb69e5ad275c
Faulting package full name:
Faulting package-relative application ID:
----------------------



then more along those lines. I can check in more detail again but those seem the most relevant to me....

 

[gslonesoldier84]

Connected to a different and bigger PSU, the video is now running on through past the windows login screen consistently.

It still has a couple problems, but that was a big one. The GPU seems operational but the PSU does not based on that. The stupendously full-on run on ARMA II may have damaged some of the PSU. I will confirm all by tomorrow.

 

[Groveling_Wyrm]

General information:

I just updated/installed these drivers via windows update:

Trying with it running from GPU again
....now it only shows the windows blue spinning dots screen for a flash of a second before it switches off video. And fan went loud right after too two times in a row now

Got furthest with gpu as video output at 20:42. It spun the white dots after typing the password for 2 seconds.
Then video went off but I heard a windows noise after even though there was no video.
Then at 20:45 it went only as far as flashing the login screen then going black and turning on fans loudly.


Some event viewer entries from around 20:42:
------------------------
The following boot-start or system-start driver(s) did not load:
dam
EhStorClass
Gemma
------------------------
The application-specific permission settings do not grant Local Activation permission for the COM Server application with CLSID
{2593F8B9-4EAF-457C-B68A-50F6B8EA6B54}
and APPID
{15C20B67-12E7-4BB6-92BB-7AFF07997402}
to the user DESKTOP-UGTD3G3\gurpr SID (S-1-5-21-2375795573-99323447-2641846060-1001) from address LocalHost (Using LRPC) running in the application container Unavailable SID (Unavailable). This security permission can be modified using the Component Services administrative tool.
-----------------------------
The winlogon notification subscriber <SessionEnv> was unavailable to handle a critical notification event.
-----------------------------
The winlogon notification subscriber <SessionEnv> was unavailable to handle a notification event.
----------------------------
The Desktop Window Manager has registered the session port.
------------------------
ERROR
Faulting application name: dwm.exe, version: 10.0.19041.746, time stamp: 0x6be51595
Faulting module name: KERNELBASE.dll, version: 10.0.19041.1202, time stamp: 0xc9db1934
Exception code: 0xe0464645
Fault offset: 0x000000000010be3e
Faulting process ID: 0x23d4
Faulting application start time: 0x01d7b17c47d52a6e
Faulting application path: C:\Windows\system32\dwm.exe
Faulting module path: C:\Windows\System32\KERNELBASE.dll
Report ID: 64f00ea0-3446-410a-9256-3beff8591e5a
Faulting package full name:
Faulting package-relative application ID:
------------------------------
Fault bucket , type 0
Event Name: APPCRASH
Response: Not available
Cab Id: 0

Problem signature:
P1: dwm.exe
P2: 10.0.19041.746
P3: 6be51595
P4: KERNELBASE.dll
P5: 10.0.19041.1202
P6: c9db1934
P7: e0464645
P8: 000000000010be3e
P9:
P10:

Attached files:

These files may be available here:
\\?\C:\ProgramData\Microsoft\Windows\WER\ReportQueue\AppCrash_dwm.exe_e1b212a7af91fba76b3eb4ebe214acef5b75dfef_c2f10bab_e4ff8f44-d207-41de-bd84-83b758c999e5

Analysis symbol:
Rechecking for solution: 0
Report Id: 64f00ea0-3446-410a-9256-3beff8591e5a
Report Status: 6
Hashed bucket:
Cab Guid: 0
------------------------------
warning
The Desktop Window Manager process has exited. (Process exit code: 0x000000ff, Restart count: 1, Primary display device ID: NVIDIA GeForce GTX 780 Ti)
------------------------------------
error
Faulting application name: dwm.exe, version: 10.0.19041.746, time stamp: 0x6be51595
Faulting module name: KERNELBASE.dll, version: 10.0.19041.1202, time stamp: 0xc9db1934
Exception code: 0xe0464645
Fault offset: 0x000000000010be3e
Faulting process ID: 0x261c
Faulting application start time: 0x01d7b17c499cfb95
Faulting application path: C:\Windows\system32\dwm.exe
Faulting module path: C:\Windows\System32\KERNELBASE.dll
Report ID: 35acc11e-a1e8-4120-b195-f39f923a54fd
Faulting package full name:
Faulting package-relative application ID:
-------------------------------
Fault bucket , type 0
Event Name: APPCRASH
Response: Not available
Cab Id: 0

Problem signature:
P1: dwm.exe
P2: 10.0.19041.746
P3: 6be51595
P4: KERNELBASE.dll
P5: 10.0.19041.1202
P6: c9db1934
P7: e0464645
P8: 000000000010be3e
P9:
P10:

Attached files:

These files may be available here:
\\?\C:\ProgramData\Microsoft\Windows\WER\ReportQueue\AppCrash_dwm.exe_e1b212a7af91fba76b3eb4ebe214acef5b75dfef_c2f10bab_98bb5f3f-d231-4ae9-b939-eba53a873c84

Analysis symbol:
Rechecking for solution: 0
Report Id: 35acc11e-a1e8-4120-b195-f39f923a54fd
Report Status: 6
Hashed bucket:
Cab Guid: 0
----------------------------
warning
The Desktop Window Manager process has exited. (Process exit code: 0x000000ff, Restart count: 2, Primary display device ID: NVIDIA GeForce GTX 780 Ti)
------------------------
error
Faulting application name: dwm.exe, version: 10.0.19041.746, time stamp: 0x6be51595
Faulting module name: KERNELBASE.dll, version: 10.0.19041.1202, time stamp: 0xc9db1934
Exception code: 0xe0464645
Fault offset: 0x000000000010be3e
Faulting process ID: 0x24e4
Faulting application start time: 0x01d7b17c4b0c7008
Faulting application path: C:\Windows\system32\dwm.exe
Faulting module path: C:\Windows\System32\KERNELBASE.dll
Report ID: 3dce28d4-8050-4599-b43b-eb69e5ad275c
Faulting package full name:
Faulting package-relative application ID:
----------------------



then more along those lines. I can check in more detail again but those seem the most relevant to me....

It looks like you may have no/corrupted/wrong chipset drivers. I suggest you go to your motherboard homepage, download all new drivers from them, as well as getting the latest gpu/sound/accessory drivers. Install your chipset drivers first, then video, sound, accessories.

 

[gslonesoldier84]

The big problem is fixed!

Thank you all for your help. I still have some of your input to implement to get the machine more functional. But, for now, an "after" picture showing the culprit and the big bahoonga that is replacing it:

To give a bit more information about the fix now:

In addition to the tweaks, additions, and changes that I was able to implement already from the input here prior to a few days ago, a BIOS setting which was causing the machine to think there was a surge was disabled as an exploratory measure.

The machine ran completely stable all the way through everything with an alternate PSU used to test the machine even under stress tests for the GPU.

The machine would run for a few minutes up to 15 minutes when connected back to the corsair CX500M PSU several times.

Once it ran for longer but then failed again when the stress test was applied.

The machine would run as normal with the CPU as the graphics output.

From LINK:

"For a system using a single GeForce GTX 780 Ti graphics card NVIDIA specifies a minimum of a 600 Watt or greater system power supply that has a maximum combined +12 Volt continuous current rating of 42 Amps or greater"

From LINK:

"The CX500 only has 38 amps on the 12v rail and isn't sufficient to power a 780."

The replacement PSU outputs 62 Amps on the 12 Volt rail.

But this problem was occurring on the previous GPU (an R9-280) which had been running the machine with no problems for years even with gaming usage. The PSU was sufficient for that video card and I had thought that this GPU would also be ok. The machine was actually ok for a good while until I did the heavy load on ARMA 2 that day after which point it didn't work anymore as described.

A possible explanation, I think, is that the PSU had developed a fault in supplying power to the GPU with the R9-280 still installed before or after I bought it. I would like to give the seller the benefit of the doubt as he assured me it had worked fine before the sale with no problems at all. He also continued to speak with me after I asked if he had any problems with it. He also gave me a partial refund for taking back the GPU which I had decided was most likely to be the cause of the problem at that time before buying a replacement 780ti replacement (also second hand). It's possible the first GPU was faulty then the new GPU damaged the PSU under loads.

Regardless, with video problems persisting after an initial period of successful operation, I was confident the new GPU wasn't the problem. But I couldn't be sure even then obviously.

Applying all the input in this thread which I was able to, the problems still persisted. The PSU hadn't been ruled out completely as I couldn't do the testing with a voltmeter/replacement PSU/etc.

Then I decided that I could do no more with the time, equipment, and budget I had. I asked what a repair cost would be from a local repair shop-owner. The cost was a significant percentage of the purchase price of the machine.

The replacement GPU cost £100. The refunded portion for the original GPU was £30 (so the seller got the R9-280 back for the £30 refund even though it worked fine it seems). The total cost of the machine including for all of this was £290. The repair/testing charge was £40 (roughly a quarter of the machine cost without the replacement GPU).

The replacement PSU was roughly £93. Altogether, the machine is operational now with a total financial cost of £423 along with months of problems finally resolved.

So, I did ask to have the machine repaired/tested at a repair shop. I couldn't really take it any further without a much bigger investment of time and equipment which I didn't have. So, I decided to go for it so that whatever the problem was it could be worked out once and for all in one big burst of test-troubleshooty-fixit.

Thank you all again for your time and help in trying to get this machine functional. If I had a bit more time to work out what was happening and to post in more detail what was happening, and/or if I had spare parts or machines to test parts, and/or if I could use a voltmeter/etc or have the time to learn, we could have solved the problem here. As it happened, after months of trying to fiddle through it with small pockets of time spent trying to work it out and one big replacement part that didn't get the machine working fully, one big burst of test-troubleshooty-fixit at a repair shop for a quarter of the cost of the machine minus the replacement GPU has finally yielded what seems to be a mostly functional machine. Still pending is the application of some input here that hasn't been applied yet and a good thorough clean and arrangement of everything now.