Multiple different GPU's in multiple system builds all CTD

Status
Not open for further replies.

sLaDeY

Honorable
Apr 26, 2012
12
0
10,510
First off, long time viewer, rare poster... the help these forums provide has meant i rarely have to post a question, and most if not all theads are answered by people more tech savvy than me... anyway, i digress....

this one has me completely stumped, I will start from the beginning, bare with me as i want to give out as much information as possible

first off I purchased all the gear for my dream x79 build

ASUS Rapage Formula (x79)
ares 32 gb ram kit (quad channel) (1600)
intel 3930K 6 core
agility 3 ssd
storage drive
evga gtx680
silverstone strider 850w
water cooling loop

well thats the basic first build

so i put all this together, crank it up, temps ok, benchmarks ok, all is well, over the moon.... then i go to play a game.. crash to desktop...over and over again across all games. I then returned the gtx680 and swapped for a gainward gtx680 phantom, hoping this was a a dogy card issue i threw it in and.....same thing.

I finally got a fix i was not happy about at all, witch was to downclock the gpu significantly, along with leaving the cpu standard speed. this was fine but started noticing it crashing to desktop but only after a much greater time.. by this stage i was fed up and after much forum browsing and trying EVERYTHING i thought 'stuff it' I will get a 7970 instead (got 2, actually) well this time, after the same amount of time, it did not crash to desktop but hung..you know, that horrible hang where the audio hangs as well....

DAMIT! i said.. wth is going on...power supply perhaps? so i replaced with a corsair ax850.. SAME PROBLEM. replaced memory with corsair dominator gt quad channel...SAME PROBLEM.

by this stage i was getting seriously depressed and dipping into money reserved for other things (the missus was not happy) but made the big choice of doing a new build on the z77 platform incase it was a mobo/cpu issue (bottlenecking or something). so along came the second build

z77 sabertooth
intel 3730k
intel 520 ssd
asus zonar audio card
new storage drive
new cooler
using the brand new quad channel dominator ram in it... should still work if i have my facts right
using the ax850 psu I purchased as a possible fix on the old build
and trying either the gtx680 (the second one i got after replacing the 1st in the hope it was a gpu issue)
and the 7970


all went together, and with the loony bin on standby I fired up bf3 and skyrim...................... SAME BLOODY THING, both 7970 and 680 presenting the same issues

at this point, as you can imagine, i was damn near ready to cry, i mean how can an entirely new build using either gfx card (confirmed working in friends builds) present the same issues!!!!

so this brings me here, with many thousands spent, 2 of the most modern 'all out' pc builds sitting here, with the option of a 680 or a 7970 in either of them... i should be loving it, but im not

I am at a loss and about ready to go and buy an xbox

edit- i have tried all manor of connectors, cables, dvi-d hdmi...etc etc to connect to my 2560x1440 monitor.. makes no difference, my last and final idea is the power from the wall is dodgey.. spikes? drops? could this do it? and would a psu not handle that flux and deliver stable power to the rails anyway? (no other devices connected present with issues)

i appreciate all of the help I have gained from you guys in the past and would love some thoughts on what direction to take

warm regards
slade
 
Solution
Take EVERYTHING back to stock clocks. CPU/RAM/GPU, insure they all work perfectly during stress tests and gaming before doing anything else. Have you done stress tests with Furmark (gpu) or Intel burn (cpu)? I would run them for atleast 15 minutes and check if your temperatures are fine. What games are you having a problem with? are all games doing this? Does it not do this in certain games? Are you using a completely new/fresh drive (IE did you reformat it if you transferred it over from one pc to the next?)

Are you using the latest drivers and firmware for your various things? Latest SSD firmware?

What are your temps/voltages/etc of EVERYTHING. What does Windows Event Viewer report as being the reason the game crashed? Have you run...

sLaDeY

Honorable
Apr 26, 2012
12
0
10,510
problems persisted before and after the bios update of the motherboards.. on both builds

im looking at voltage regulation issues now but have no idea how to problem solve in this area


looking for other ideas before i spend the last of my savings on a UPS, power boards, mutimeter and new power cable....
 

AdioKIP

Distinguished
Jul 10, 2008
789
0
19,160
As you stated, the issue could have to do with power from the wall. Have you tried adding a UPS?

Have you run stress tests on the memory (even though you tried 2 kits). When filling all the ram banks this is even more important. Run Mem86 for a few hours and make sure your ram isnt having any errors.
 

sLaDeY

Honorable
Apr 26, 2012
12
0
10,510
must have edited my last post while you were posting adio, yea, tested the memory and all ok, looking at ups and the like, will take system to a mates place (brand new house, modern electrics, lol) and see if problem persists before lashing out on a new ups and surge boards,

cheers, appreciate the input
 

AdioKIP

Distinguished
Jul 10, 2008
789
0
19,160
Smart move going to your friends first before buying a UPS. Hopefully that will give you some type of results, you seem to have thoroughly tested most possibilities.

Also, did you manually set your ram timings and voltage in the bios? If not I would definitely try manually doing it.
 
Take EVERYTHING back to stock clocks. CPU/RAM/GPU, insure they all work perfectly during stress tests and gaming before doing anything else. Have you done stress tests with Furmark (gpu) or Intel burn (cpu)? I would run them for atleast 15 minutes and check if your temperatures are fine. What games are you having a problem with? are all games doing this? Does it not do this in certain games? Are you using a completely new/fresh drive (IE did you reformat it if you transferred it over from one pc to the next?)

Are you using the latest drivers and firmware for your various things? Latest SSD firmware?

What are your temps/voltages/etc of EVERYTHING. What does Windows Event Viewer report as being the reason the game crashed? Have you run memory diagnostic tests such as windows memory diagnostics or Memtest64?

Does it crash when you only have say 2/4 sticks in or 1/2 in? Reduce your ram timings to stock (if its higher than 1600, reduce it back to 1600, as that is technically overclocked.)

Edit: didn't see your edit.
 
Solution

sLaDeY

Honorable
Apr 26, 2012
12
0
10,510
yep, completely stock clocks.. fresh install, brand new drives, nothing but essential testing software, origin, steam and assorted games (a big one i play, and always test first, is bf3... but others crash as well)

prime95 no errors
furmark -all maxed out settings (duel gpu version, resolution set at 2560x1600 (massive res) - 71 degrees celsius stable
intel burntest no errors after run, temps measured via coretemp and all below 65c
will update ssd firmware but doubtful as issue presents on multiple install on different drives


will get back with a response to the event viewer readout on crash.. (not sure it gave one, other than the 'stopped responding' message
will try with 2 mem sticks in
 

sLaDeY

Honorable
Apr 26, 2012
12
0
10,510
i'll add my unigine heaven score

7970 (gigabyte ghtz edition) in corssfire (both x-bridges installed)

1920x1200/ 4xaa/ 8xAF/ tesselation extreme

....... score: 2576 (thats bloody good as 2 standard 7970's get about 2100)

ran temp testing as it ran... 77 degree max

AND STILL CRASHING IN GAMES!! skyrim, bf3, borderlands 2

ohh well back to the drawing board...


......have not ruled out gremlins in the wiring..those sneaky bastards will eat anything
 
Status
Not open for further replies.