|
Hey
My Build + Show Spoiler + Windows 8.1 64bit Mobo- Asus z87-a CPU- i5 4670k running at stock (Noctua cooler (Currently going to waste :<)) gpu- Asus GTX780-DC2OC-3GD5 PSU - Corsair RM750 Case- define r4, stock fans 8 gig (2x4gb) kingston hyperx 1866 ram Have a 120gb ssd with windows + league/wildstar and a 2tb drive with steam + everything else
An old deathadder black edition and some random peripherals
I've been having some repeated crashes on my relatively new desktop pc. They're seemingly random, sometimes none for hours, sometimes 5 in as many minutes.
Basically the graphics driver times out and tries to recover, sometimes this is just a brief black flicker.
Sometimes (Mainly when playing wildstar) the computer locks up with the screen showing a solid colour, or a frozen image with massive artifacting for 1-2 minutes, before ctrl+alt+delete mashing gets me to the ctrlaltdlt screen, and returning to desktop from there gives a "memory referenced at xxxx could not be read yyy" type error and wildstar has crashed.
Sometimes (Mainly cs go) it just locks up indefinitely, and I have to reboot.
None of these give crash dumps or bsods, and the only event in event viewer is a lot of "driver has stopped responding" logs. + Show Spoiler +
It very rarely (maybe 2-3 times) has crashed when i've just been at desktop/chrome, (And I believe thse were times where the flickering started while in csgo/wildstar, and i closed the program. Then the crash happened a few seconds/minutes later.) And I haven't crashed once it league at all, despite playing it probably the most out of the three.
Sometimes sound will skip/loop, sometimes it keeps playing in the background. I can often still talk to people on ts/skype despite my monitor having "lost connection/feed" from the pc. Sometimes I can still talk and lsiten to skype, but the game sounds are looping/skipping in the background.
It's been happening for a while, but it's gotten a LOT more common over the last week or so, from once every few days to multiple times an hour.
I've tried maybe 6-7 different gpu drivers (Including the super old ones which came from the manufacturers site. 327 or something) and it's been crashing with all of them. Some drivers had different visuals during the crash (different kinds of tearing/artifacting.) Cleaning drivers with ddu between installs.
I've run memtest on two occasions, once 5 passes (took 4-5 hours) and once 10 passes (9 and a half hours) and both returned 0 errors. I've also tried both xmp profiles available in the bios, as well as disabling xmp, all had crashes.
At all points the temperatures are super stable. CPU below 50, gpu below 60 (maybe rarely peaking at 70) but generally gpu was below 50 too. It's even happened when both were sub 40.
Ran furmark burn in test for 45 minutes with nothing happening. (One of the subsequent furmark tests had some screen flickering occasionally at points, but it never crashed, just the black flicker 3-4 times over 15 minutes.)
Side note, after installing the old drivers, then reinstalling the current ones, my curse client (addon manager for wildstar) will no longer load, giving a .net error message. + Show Spoiler +
SFC scannow says everythings fine, DISM checkhealth commands say everythings fine.
Not sure what to do/check next, any help much appreciated, i'm going crazy ~
(Sorry for any typo's/missing info, I'm a bit annoyed/tired right now so I might have forgotten somethign :<)
|
United Kingdom20145 Posts
Try 340 driver and set pci-e to 2.0 in bios, dunno if it'll help
|
On August 21 2014 08:23 Cyro wrote: Try 340 driver and set pci-e to 2.0 in bios, dunno if it'll help I've changed "pci-e speed" from "auto" to "gen2", i'll test it today
Strangely there was no gen3 option, only auto, gen1 and gen2.
Isn't there meant to be a gen3 option? The mobo definitely supports pcie 3.0
edit: so that made me reinvestigate bios, seems (despite asus software saying it was up to date) My bios version was super old, like from 2003. I've updated it to the latest version (which advertised itself as intel gen4 compatible) so maybe that was the problem. Time to try and force it to crash again :D
|
Bios updated now but crashes still happening.
With pcie speed at gen 2 it's still just as unstable. Played for 15 minutes then crashed. Took a break, started again and within 15s the screen was flickering green
What would be the next thing to test? Not sure where to go next. Do the memtest results guarantee the rams fine, or would it be worth trying to get hold of some extra ram to swap that out?
|
United Kingdom20145 Posts
I've changed "pci-e speed" from "auto" to "gen2", i'll test it today
Strangely there was no gen3 option, only auto, gen1 and gen2.
You're probably changing the wrong one, that you're not using (first two slots are x16 or x8/x8 from cpu, the other one is 2.0 x8 from pch)
There's no 2003 bios for 2013 motherboard, but good to be up to date
I don't think it's memory problem. Try setting 2.0 in the right place if possible and underclocking GPU by 100mhz on the core and memory with msi afterburner
|
On August 21 2014 18:56 Cyro wrote:Show nested quote +I've changed "pci-e speed" from "auto" to "gen2", i'll test it today
Strangely there was no gen3 option, only auto, gen1 and gen2. You're probably changing the wrong one, that you're not using (first two slots are x16 or x8/x8 from cpu, the other one is 2.0 x8 from pch) There's no 2003 bios for 2013 motherboard, but good to be up to date I don't think it's memory problem. Try setting 2.0 in the right place if possible and underclocking GPU by 100mhz on the core and memory with msi afterburner I just checked through msinfo32 and it said it was version 16xxsomething from june 2003 lol, dunno what was going on there.
You're probably right on the pcie thing, i'll check the other type.
I've tried underclocking before, had no effect, but i'll try again in conjunction with pcie 2 and see if that makes any difference.
If this doesn't do it i'll probably try and reformat later today, I've been getting some wierd windows behaviour throughout the process, like the windows installer module service just randomly disabling itself when i try and run sfc o.0
|
IMHO the best test for this would be using an identical GPU in your system for a few hours to a few days. If you have the option to buy one and pick it up from a store, you could then get a full refund (or keep it, depending on the results).
In the past I've had issues with RivaTuner (and I assume similar software could have the same problems) causing artifacting and BSODs if I changed (updated or downgraded, "clean install" checked) video drivers without reinstalling it, though that may be an issue pertaining to just that one PC (Win7 32-bit).
Other than that, I'd try using chkdsk on the Steam partition, as you mentioned CS:GO shows the most severe symptom - lockup requiring a hard reset.
Whenever I've experienced driver crashes and recoveries, changing driver versions worked for ATI cards, but for nVidia GPUs the cards themselves were to blame and I had to RMA. Just my 2c.
|
I had a very similar problem and i was close to exchange my graphic card. If you have installed daemon tools (or any other image mounting tool), kill the process in your task manager.
|
Switching pcie speed didn't do anything (I haven't crashed yet, but there's been screen tearing/flickering every few minutes so it's only a matter of time i think.) Did a semi-reformat with my win8 disk, (reinstalled windows but kept user files.) and that did nothing either.
On August 21 2014 23:58 chiflutz wrote: IMHO the best test for this would be using an identical GPU in your system for a few hours to a few days. If you have the option to buy one and pick it up from a store, you could then get a full refund (or keep it, depending on the results).
In the past I've had issues with RivaTuner (and I assume similar software could have the same problems) causing artifacting and BSODs if I changed (updated or downgraded, "clean install" checked) video drivers without reinstalling it, though that may be an issue pertaining to just that one PC (Win7 32-bit).
Other than that, I'd try using chkdsk on the Steam partition, as you mentioned CS:GO shows the most severe symptom - lockup requiring a hard reset.
Whenever I've experienced driver crashes and recoveries, changing driver versions worked for ATI cards, but for nVidia GPUs the cards themselves were to blame and I had to RMA. Just my 2c. Not sure where i'd easily pick up a new gtx 780 though, I got it through an online shop, there aren't many actual hardware stores near where I live, and I'd imagine it'd be pretty hard to just return it.
What kind of program is rivatuner? I was having some very odd behaviour earlier where some of the times the display driver crashed, the computer would get SUPER slow (as in clicks take 1-2 minutes to register and everything seems frozen) but if i slowly opened task manager, I could see that gputweak (Or one time msi afterburner) were using 30-40% of my cpu (maxing out a core) and ending those processes restored everything to normal. It seemed to stop after I uninstalled gputweak, but could it be msiafterburner locking everything up on gpu timeouts? Either way I might try uninstalling them and repatching drivers later.
My brother has a 660ti, but I don't see much point in testing that in this system, because it's just as likely to be a 780 specific driver conflict or something as it is to be the card itself, not sure if it's worth the bother of dismantling his computer too.
On August 22 2014 00:37 Striker.superfreunde wrote: I had a very similar problem and i was close to exchange my graphic card. If you have installed daemon tools (or any other image mounting tool), kill the process in your task manager. Don't have any software like that, it's a pretty barebones pc (especially now after reformatting.)
I suppose it looks like I might just have to rma the card, but that'll take a week or two at least and my wildstar guild is about to start raiding, so it would suck to find out in 2 weeks that that wasn't the problem and then have to wait another 2 weeks to rma the mobo or something and miss out on everything T.T
edit: oh, and I just ran chkdsk on the 2tb drive (the one csgo is on) and it returned no problems.
|
I don't get this, I can sit watching unigine heaven 4.0 go on extreme for over an hour, then I boot up wildstar on lowest, and 2 minutes later everythings locked up T.T
Sad times
|
|
United Kingdom20145 Posts
I didn't really hear of those issues with these hardware setups (or even widespread issues at all on decent systems) and it was affecting multiple games
|
It's happening in cs go as well, just crashing every other round right now :<
|
Rivatuner does pretty much what MSI Afterburner does, except it's far older (and thus supposedly more mature and stable). Have you tried shutting it down and playing without it ?
|
I guess I'd give up... you tried everything that's reasonable to suggest. You tried different drivers, made sure no files are corrupted, memtest finds no errors, forced PCI-E 2.0, etc.
The only other thing I remember is, I once had issues on a driver version that could be fixed by disabling the "threaded optimization" setting in the NV control panel.
As a last thing to try before sending the card back to the shop, you could reinstall Windows from scratch, then next time don't install any drivers for the board etc., just let Windows Update set things up how it wants to.
Another thing to try is, take out the GTX780 and either play with a different card if you have access to one, or play with the integrated graphics from Intel. The Intel stuff should be good enough to make CS:Go fun, and you could play that for a few hours or so until you are convinced there's no crashes. That would then point to the GTX780 having an issue.
|
|
|
|