Brand New custom build; Display Glitching and Crashing Whilst Playing Games.

Toggle sidebar Toggle sidebar
MrHakisak

MrHakisak

Distinguished
Jul 3, 2013
43
0
18,540
Hello,
I Built a new PC for a mate of mine.
It was my first DDR4 system (iv built loads of PC's with DDR3 platform).

  • I got him some cheap'o Corsair Vengeance RAM (cuz who needs high speed RAM?).
    An Asrock motherboard,
    An i7 (because of light CAD work)
    And a GTX 970 (for games).
Full PC Specs Below.

The Problem:
The PC keeps crashing every 10 to 15min into a game (tested on skyrim and tf2).
After a BIOS update the computer does not blue-screen anymore! except now the graphics driver seems to be glitching out really bad then the display crashes (which you then need to restart the PC in order to get the display back), see last picture in album below.

Collection of Crash Images and Display Glitching



A (note) problem:
 Checking the motherboards RAM support list, the Corsair Vengeance RAM that the PC has; is not listed -_- That would be my fault for not checking. but i hope that is not the problem, and I don't have any other RAM that i can test it with.

PC specs as follow:
CPU: i7 6700
MOBO: Asrock B150M-Pro4
RAM: Corsair Vengeance LPX CMK8GX4M2A2133C13 8GB (2x4GB) DDR4
GPU: EVGA GTX 970 ACX 2.0
PSU: Silverstone Strider 600W (80+ Bronze)
Storage: WD Black 1TB
CPU Cooler: CM Hyper 212x
Monitor: Benq RL2455HM
OS: Legit Win 10 64

I have run some tests and changed some settings around, here is what i have already done.
HD tune quick scan; HDD is fine.
Windows Memory Diagnostics are fine.
All windows updates downloaded and installed.
The newest Nvidia drivers are installed (also tried a fresh uninstall then install).
standard mobo drivers from given disk.
UPDATED BIOS to the latest version (1.6).

Any help Is Greatly appreciated! $
 
Solution
D
Yep pretty much has to be a faulty card if it crashes with 2 different systems and 2 different power supplies. EVGA has the best warranty service in the business so an RMA should be painless.
Sort by date Sort by votes
D

Deleted member 217926

Guest
IRQL errors are very often memory related. Have you updated to the latest BIOS and enabled XMP? Also assuming you are running a legit copy of Windows?
 
Upvote 0 Downvote
MrHakisak

MrHakisak

Distinguished
Jul 3, 2013
43
0
18,540


I have the latest BIOS version. but I haven't touched any RAM configuration settings, i'll let you know what happens
 
Upvote 0 Downvote
MrHakisak

MrHakisak

Distinguished
Jul 3, 2013
43
0
18,540


Iv got XMP on Auto (it was always on auto, i think). but now the computer is straight up freezing (during games). any ideas?
 
Upvote 0 Downvote
D

Deleted member 217926

Guest
Does it allow you to set the XMP value as XMP 1 or XMP 2 or something like that? You can also set the memory manually. If that doesn't work then run memtest for at least 12 hours. The integrated Windows memory checker is useless.

http://www.memtest.org/

Also Skylake in particular can show gains of 30% or better in certain tasks with higher speed memory. CAD work is one of those things that will benefit. Gaming will not, not much at least.

 
Upvote 0 Downvote
MrHakisak

MrHakisak

Distinguished
Jul 3, 2013
43
0
18,540


I can set XMP to "Profile 1" (there is no "profile 2"). But it already has the RAM set to its proper standard for that kit (timing, voltages and clock). here is an Image do i need to change anything?

if memtest picks up anything, do you think its because that motherboard doesn't officially support that specific memory? or that the memory its-self is broken?
i will run memtest tonight, non the less.
 
Upvote 0 Downvote
D

Deleted member 217926

Guest
If memtest sees errors then try it with one stick at the time to see if there is an obvious bad module. Official support doesn't usually matter but with DDR4 being fairly new it might. If you can't isolate a bad stick with memtest then it may very well be a compatibility issue.
 
Upvote 0 Downvote
MrHakisak

MrHakisak

Distinguished
Jul 3, 2013
43
0
18,540


after 16 hours of memtest86+ (with both ram modules installed) here are the results (results) -_- . although it is displaying the wrong timings (but still correct in the bios).
maybe after the bios update; that fixed the blue-screening.
but now it could be a different problem as the computer is only freezing (not crashing), maybe a software error?
should I reinstall windows on the PC? or any other ideas before that?
 
Upvote 0 Downvote
D

Deleted member 217926

Guest
Well huh. I expected memory errors. You are using a legit version of Windows right? I did ask that before. If not that could be the problem.

Check the SMART values of the hard drive with something like Speedfan and make sure there are no disk errors.

When does it freeze? Can you replicate it or is it random?
 
Upvote 0 Downvote
MrHakisak

MrHakisak

Distinguished
Jul 3, 2013
43
0
18,540


I also expected memory errors :/
Yes it is a 100% Legit version of Windows 10 64-bit and is activated.
here are the HDD's results (images)
everything seems alright.

skyrim black screened at 2min 20sec (the whole PC is frozen, and the steam account went offline). but before it froze; the screen started flickering different colors with green dots.
TF2 freezes at 1min 30sec (no black screen or strange color problems) with the audio still looping though.

(also there are some missing textures in skyrim, but Im pretty sure that's just skyrim. although it is an official steam version).
 
Upvote 0 Downvote
D

Deleted member 217926

Guest
Artifacting like that sounds like the graphics card. Could also be the power supply. Do you have another of either you can test with?
 
Upvote 0 Downvote
MrHakisak

MrHakisak

Distinguished
Jul 3, 2013
43
0
18,540


yes i have a spare gt 440 i can test with.
but the power supply... that would be a bit more difficult.
I have a spare 400w silverstone strider (80+ Bronze) i can easily take out of my spare PC. but 400w would be cutting it close :/
we'll see what happens after the video card swap
 
Upvote 0 Downvote
D

Deleted member 217926

Guest
Just to rule it out make sure your temps are good. CPU and GPU. Could be something like an improperly seated CPU heatsink. Doubtful but possible.
 
Upvote 0 Downvote
MrHakisak

MrHakisak

Distinguished
Jul 3, 2013
43
0
18,540


the temps are fine.

I ran OCCT for the cpu stress and there were no problems.
I then ran Furmark which froze the PC within a min.
I then put in a GT440 and Furmark ran fine.

i took the gpu to another known working pc (which previously had a gtx 980) and the video card worked fine with OCCT, Furmark Unigine-Heaven&Valley.

looks like it could be a power supply problem when the video card pulls too much power. do you think that could be the problem? the cpu stress test ran fine and that should pull some power, but probably not as much as a video card.

I will have to connect another power supply to the system to rule that out :/
the current power supply is a 600w which should be plenty.
 
Upvote 0 Downvote
D

Deleted member 217926

Guest
Yep probably the power supply. Silverstone used to be a brand I recommended by name but their lower end stuff hasn't been too great lately.
 
Upvote 0 Downvote
MrHakisak

MrHakisak

Distinguished
Jul 3, 2013
43
0
18,540


alright, so today was a wild journey.

I deduced that when running Furmark; as soon as you touch (bump or move at all) the GPU power cables; the system will crash. just to make sure it was not a coincident (relating to faulty power supply) I tested it at least 10 times on 2 different power supplies (the power supply that was already in the system and an OCZ 850W 80+ Gold that was from another system). as soon as the cable is touched; the PC crashes (instantly).

I then took the Video card to a working PC (running win-7 that had a GTX 980 inside) and put in the 970. the computer crashed a couple of times (without touching the power cables), i then manged to get to 'device manager' only to find out that "Windows has stopped this device because it has reported problems. (Code 43)". i tried uninstalling and installing drivers and all that but nothing works. because the PC doesn't recognize it has a video card i cannot run Furmark or any 3D applications. I then took the video card to another PC running windows 7 and it has the exact same problem.

looks like the video card is faulty right?
its (obviously) under warranty so i can return it.
what do you think?


an after note: the video card worked on another windows-7-PC (with an array of benchmarks) once before, but now its getting those error problems.
 
Upvote 0 Downvote
D

Deleted member 217926

Guest
Yep pretty much has to be a faulty card if it crashes with 2 different systems and 2 different power supplies. EVGA has the best warranty service in the business so an RMA should be painless.
 
Upvote 0 Downvote
Solution
MrHakisak

MrHakisak

Distinguished
Jul 3, 2013
43
0
18,540


yer but you see... i live in Australia... so shipping it to murica would be a last resort >_<
ill try handing it to the company i got it off first (they are pretty easy-going with RMA's).
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom