gpu crash with this error message

[David_384]

i was gonna do a 3dmark test but my drivers crashed and i got this error message: Workload work failed with error message: eva::d3d11::rendering::scene_renderer::render(): draw_depth_task for thread 0: File: device_context.cpp
Line: 515
Function: struct ID3D11CommandList *__cdecl eva::d3d11::deferred_device_context::do_finish_command_list(bool)

Expression: native()->FinishCommandList( restore_deferred_context_state, &result): DX11 call failed.

Ran out of memory
E_OUTOFMEMORY: ID3D11DeviceContext::FinishCommandList:

so i wounder what that means?

 

[junkeymonkey]

if you oc'ed your card its failing to hold the overclock is my first guess 2ed would be a failed or a failing card in some way if its still at stock ???

 

[combinebasic]

If your gpu card driver is crushing, you can try to uninstall that driver using the device manager. Then restart your computer, then install again your driver. If the same problem you will encounter, go on for the repair of your OS I'm sure there are damage or corrupt files. Because OS managing your driver registry. Be sure also that your gpu is normal temperature.

 

[junkeymonkey]

'' If your gpu card driver is crushing, you can try to uninstall that driver using the device manager ''

you do know that's not supported by NVidia , right ??

 

[combinebasic]

Yes but its better you need to try the procedure I told to you.

 

[junkeymonkey]

''Yes but its better you need to try the procedure I told to you.''

if its known to cause issues then why would you recommend he do it ??

''Uninstalling Drivers Using Device Manager is Not Supported
Issue
On all supported versions of Microsoft Windows, uninstalling the NVIDIA driver using the Windows Device Manager may not remove associated files or applications.''

from any NVidia driver footnotes


what he should do is google '' FinishCommandList( restore_deferred_context_state, &result): DX11 call failed.'' and see whats said

example
https://community.futuremark.com/forum/showthread.php?157468-What-is-this-suppose-to-mean


https://rog.asus.com/forum/archive/index.php/t-55213.html

https://forums.geforce.com/default/topic/528042/gtx-660-ti-crashing-windows-need-help/?offset=25

https://www.reddit.com/r/techsupport/comments/4fxvff/directx_errors_and_crashing_when_launching_games/

ends up most times the card

same with AMD as well seeing I thought he had EVGA above ?? but anyway

https://www.techpowerup.com/forums/threads/artifacts-on-brand-new-6950.154045/

http://www.techsupportforum.com/forums/f25/amd-display-driver-crashes-669595.html

 

[David_384]

my card is factory oc to 1126 mzh but do i need to make that lower

 

[junkeymonkey]

should not your guaranteed the card will do factory clocks as stated .. downclocking below that to me is a faulty card

so if your card sold as 1126mhz card it should do that with ease fully stable all day

just like from here on this error as well '' E_OUTOFMEMORY: ID3D11DeviceContext::FinishCommandList ''

http://forums.evga.com/Odd-Graphics-ProblemGlitch-with-GTX-560-Ti-SC-m1075701.aspx


to me sounds like a power issue weather it in the card or board or PSU ?? some way seems the card is not getting its full power to run correctly

with a overclock you did to it if so then you could call it a bad oc but you say it does it at stock out of the box clocks ??

all you can do is try the card in another capable computer and test it out to see if the card does the same if so call it a bad card if not then look at your psu or something else not powering the card stable at full loads

go ahead and try clean drivers installs of the same or older / different driver to see but ....?????

 

[David_384]

yea i tried my card in my friends computer and i worked fine but maybe the psu but i checked in bios and everything seems om the +12 was on 11.7 everything else was ok. i tried a different pci-e slot but i will try the last slot tomorow and see if it works or i will try an new psu

 

[junkeymonkey]

''+12 was on 11.7 ''

and that's at full load power draw ?? or is that just running at like desktop ?? if that's what it shows its doing sitting in the bios that's maybe too low cause when you run things up top full load power draw its going to drop that lower and maybe past the atx limit ???

best I recall '' According to ATX 2.3, spec of your PSU, the specs for +12V are between +11.40 V to +12.60 V.

so can the psu drop below that under a full load spike ??

use hwinfo64 and run it in the background it will give max-min- and average - on that 12v+ and see under something demanding how low it min dips to

my psu can drop to 11.7 v under full load power draw stress at desktop like right now its at 12.091v

http://www.hwinfo.com/

 

[David_384]

it was 11.7 in desktop to but in full load it was 12.004

 

[junkeymonkey]

maybe, just don't sound right seems backwards ?? your gaining voltage under a load ?? hmmmmm......

 

[David_384]

yeah idk is there any software u can recommedn so i can troubleshoot my motherboard

 

[junkeymonkey]

not really ? what is that psu your using ?? may as well give you whole specs

thing is I guess I never seen a psu gain voltage when under a load to me that's odd I would think doing nothing it should show 12.004 then as you put the poad to it it will drop down to 11.7

that's what I don't get that's why I gave you that hwinfo to use to check the bios readings your using

my card peaks at 275w full load and with that my psu drops to 11.60 v from the 12 .xx volts it started out with

don't know something there just don't seem right .

well I checked my bios readings on the 12v+ and its shows for the cpu voltage ?? now weather or not that's true to how the psu is supplying the whole computer or just to the cpu ??

mine shows the same bios and hwinfo ?

that's how mine dropped under a gpu stress test from a max of 12.091 down to a min of 11.722

 

[David_384]

ive got a corsair rm750x

 

[junkeymonkey]

well run that hwinfo in the background and run your card intill it crashes or what ever and look over the readings from hwinfo to see if anything looks to be out of whack ?

you may find or see a bad dip in something at the time of the issue occurs ? like maybe for that instant the 12v+ may drop below the threshold of safe voltage or what ever you may can find in it to help point to something to look at ?

need to like the rest of your system specs as well like motherboard and all

 

[David_384]

mu specs is: asrock fatal1ty z97x killer
intel core i7 4790k @ 4.00 ghz
16 gb 1866 mhz Ram
corsair rm750x psu
pny xlr8 gtx 970

 

[junkeymonkey]

looks like a nice build and should work well ? still only some things I can guess is [1] your card is bad if its at stock settings [2] you try to overclock the card and it just fails to hold the overclock you use [any] [3] power issue weather in the card or psu failing under stress / full load ???

my 900 card if I push the oc too more then the card likes or can do is when I can get errors as yours - like a good high oc may work all day looping heaven bench or some other things but may crash with a like error or something in a certain game or program/ application and needs a lesser or even on overclock at all ??

you said it worked fine with out issue in your friends computer ? so it still could be a power issue with your psu in yours ??? like when the card get to need full load the psu may fail to supply the min 12v needed to do so ?

tough call

 

[David_384]

yeah i tried to fet it on load and the 11v got down to 11.23 wich i guess is not good and the 5v got down too 0 att i 1 point but i think it was just the program who got that wrong 3.3v was abit under like 3.296 but thats ok i guess

 

[junkeymonkey]

well ? cant say if you mean the 12v got down to 11,23 that's bekow ATX spec. [

+12 VDC
Tolerance : ±5% (±0.60 V)
Range (min. to max.) : +11.40 V to +12.60 V

+5 VDC
Tolerance : ±5% (±0.25 V)
Range (min. to max.) : +4.75 V to +5.25 V

+3.3 VDC
Tolerance : ±5% (±0.165 V)
Range (min. to max.) : +3.135 V to +3.465 V

could be your monitoring program ? but find another to run as backup compare to what you show above don't look good especially if the 5v is really dropping to 0 ???