Games Crashing (GTX 1080)

[PierreLTrili]

Hello, I created a thread about the same problem (Games crashing) some times ago but this time I wanted to go a bit further as the problem isn't solved yet.

My Setup :

i7-6700k with its own watercooling system (R360 Pacific from Thermaltake)
Z170-A
Galax Hall of Fame GTX 1080
2*8GB of Ram (3000 MHz Corsair Dominator Platinum)
Cooler Master M2 Silent Pro 850W Silver PSU
Thermaltake Core P5

So my issue is that I'm playing 2 games, League of Legends and H1Z1 King of the Kill. LoL tends to crash like 1 game out of 20 (it crashes suddenly and let me join back the game immediately) while on H1Z1 the game freezes for 1-2 seconds before closing, sending me back on Desktop. It seems to happen like every 3 or 4 games depending on game's length. Sometimes it can freeze after few minutes while sometimes it will freeze 30 minutes later, and sometimes it just won't freeze at all. It seems to be pretty randomly, it's not related to what I'm doing in the game (It happened while walking, while shooting etc..) on H1Z1 I'm also experiencing some really little freezes (Maybe more something like huge FPS drops ?) when opening/closing doors or moving fast in little spaces, again it's randomly and not at all systematic here.

My drivers are up to date (I'm updating them through Nvidia Geforce Experience software), same for my Windows Update (Windows 10 Pro Edition). So this leave me to some hypothesis :

1) There is an issue with my PSU
2) There is an issue with my GPU
3) There is an issue with my RAM
4) There is an issue with my Watercooling (I do not think as it's only cooling my CPU)
5) There is an issue with the PCI-E Riser Extender cable (With its core P5 Thermaltake provide a Riser cable which is considered as garbage, most of them were broken when received by buyers).
6) There is an issue with the games (I contacted Daybreak, creator of H1Z1 to get their opinion as well).

I would personnally think it's related to either PSU or PCI-E Riser Extender cable. I will get another PSU soon (Snow Silent 1050W from Seasonic) and will probably change my PCI-E Riser Extender cable, the fact is considering the complex build I have, I don't want to change part one by one without knowing where the problem came from. Is there a way to identify the source of the problem easily ? (Through benchmarks etc.) because it seems to be hard to reject the fault on either the PSU, the GPU, the Ram or even the PCI-E. What disappoint me is that, the game crash but I do not get any black screen, I mean if the problem was coming from PSU/GPU/PCI-E I should get a black screen coz the GPU would be disconnected for a moment no ? (I do not know if I'm clear right now, I apologize for my english by the way, trying my best to make it as clear as possible).

I downloaded OpenHardwareMonitor to get some information about temperatures/load from CPU/GPU and I didn't see anything abnormal. The highest temperature reached by my CPU was 60C during H1Z1, rest of the time it's around 21-25C. GPU's Idle temperature is 35-40 and the maximum reached was.. 62C if I remember correctly. The only number that triggered me was that my CPU Core #2 reached when the game crashed 92% of load (Other Cores #1, #3 and #4 were at way lower load rates) considering it's an i7 is this normal ?

Oh I almost forgot, I have another little problem, that may give you an hint about the problem. When the computer is on standby mode (Fans and GPU aren't running, computer sleeping), once I wake him up, it takes 1-2 seconds for the keyboard to be detected AND the computer is lagging during 4-5 seconds. I can't enter my password at this moment to access the desktop, if I do it (Let's say my password is qwerty123) the password will bug, ending something like qqqqqqwwwwwerty1111123, it's lagging. I do not know why this is happening, it is systematic, every time the computer is sleeping, once I wake it up it lags on the locking screen. I didn't experience this on my older desktop even on Windows 10. What could be the reason of that and is this related to my crash problem ?

Thank you for reading me and helping me, again I apologize for my english, I hope it was clear enough.

Pierre

 

[Makentox]

Hello, I created a thread about the same problem (Games crashing) some times ago but this time I wanted to go a bit further as the problem isn't solved yet.

Have you overclocked your 1080?
When next time u crash/freeze go to:
Right click windows>event viewer>windows logs>system, check if u get any warnings or errors or post it here.

 

[Makentox]

Have u fixed it??? What was an issue?

 

[PierreLTrili]

Sorry I did a mistake selecting your message as the solution.

Well I would have answered you that No I never overclocked it, but when running GPU-Z I can see that my GPU Clock and Boost are equal to the default Clock (So the GPU is normally running at its stock clock) BUT when I go into the " sensors " tab (or when I check in OpenHardwareMonitor) it says that my current frequency (while playing h1z1 is 2000-2012MHz. Is this normal ? Sorry I have no idea on how frequencies work when it's about the GPU.

 

[Makentox]

Did this issue begin right after u built/purchase your computer or after(when?)
Well considering the fact that u freeze for couple of seconds and dont end up with bsod that requires a restart there can be 2 common things that i experienced before. GPU or RAM. Im not an expert, but if i were u i would:
1. Like i said before check event viewer!
2. If you have overclocked your ram reset back to default to test.
3. (most likely not) Download msi afterburner to check built in monitor. For me HWmonitor shows faulty gpu frequency readings and i haven't checked other softwares. So msi afterburner will show correct value. Just to make 100% sure.
PCI-E Riser Extender cable can be the reason but i have 0 experience about it so cant say for sure.
Personaly for me OpenHardwareMonitor is garbage, it lied so many times to me.

 

[PierreLTrili]

I just experienced another crash, so I went into event viewer and there was nothing (I have lots of " informations " and only few " errors " I checked these errors are like this way " Authorization parameters Application-specific do not give the Authorization " Local Activation " for the server app COM with the CLSID etc.. etc.. " I think there is nothing related to the crash ? Especially considering the times do not match with crashes).

What triggered me was that BOTH OpenHardwareMonitor and GPU-Z when the game crashed told me that the GPU was at a 278 MHz Frequency (after some seconds it keeps staying at this frequency, I assume it's just the low frequency of the GPU when no game is running, so it showed me 278 because the game having already crashed, the GPU went back to this low frequency.)

and yes I did OC my Ram (I mean it's a 3000MHz one, I used the XMP profile to change it from 2133 to 3000). I should try with 2133MHz ? I will install MSI Afterburner.