Random bluescreens

[Constant Failure]

TL:DR: Random bluescreens once a month happen, IRQL NOT LESS OR EQUAL, drivers up to date, specs in signature.

 

Let's start off what my system settings are:

ASUS z170-A Bios = 1702 (which was the 2nd latest at the time when built and OC'd)

i5 6600k stock/4.5GHz

corsair vengeance ddr4 ram 2133/3000MHz(XMP)

Win 10 64bit home

 

Overclocking was stable* from 24h stress test with h.264 encoding run as well as withstood IBT extreme for 2hours without exceeding 70C (might have been 65-68C at most but I don't remember since it was done in august) and has not crashed or bluescreened repeatably ever. 

My overclock was done by just changing the following parameters: Multiplier to 45, SVID( I think this was the name for it in this Mobo and BIOS) off, Vcore 1,25V (Manual) and for the RAM XMP profile. LLC was left at auto, for some reason during my stability tests using LLC 5 would be unstable after x amount of time but with auto the same settings would prove to be stable. 

 

*= random bluescreens appear approximately once a month, mostly with code IRQL NOT LESS OR EQUAL, maybe once(or twice) I remember having CLOCK_WATCHDOG_TIMEOUT.

 

Ok so there's two problems, may or may not be unrelated to each other.

 

First problem: 

If I turn off my PSU (switch at the back) for whatever reason, let's say thunderstorm. And then turn it back on and turn on the PC, my system tries to start, lights from mobo, CPU cooler and GPU flicker and everything turns off. After approx 2 seconds the system restarts itself properly like nothing happened and everything works fine after that.

 

This only happens when my CPU is overclock by the settings above.

Stock CPU, stock ram. Turns on at first try.

Stock CPU, XMP ram. Turns on at first try.

OC CPU, stock ram. Turns on, then off, then on again. repeatably.

OC both. Turns on then off and then on again repeatably.

 

The question is, is this a problem for my PSU (corsair RM650x) or is my LLC setting too low so the processor doesn't get enough juice at the start or something else?

 

Second problem:

These random bluescreens, and why I say random is that I cannot repeat them ever, they occur rarely and the error codes are the two what I listed above.

 

The times I remember these happening have been, watching a youtube video, playing GTA V, playing something I forgot Fallout 4 (vanilla). My temperatures for CPU while gaming are 50-60C and for the GPU around 60-72C (depending on the game) and during one more different game I can't remember (most likely crysis 3 or deus ex humankind divided).

 

I've tried narrowing down what could be the problem by updating my drivers in order of listing:

1. USB 3/3.1 and Chipset drivers

2. audio drivers

3. GPU (zotac amp 1070) drivers

while the problem is still randomly persisting. Between a step I always waited until the bluescreen occurred before updating the next drivers.

 

I've also had this happen when my CPU is overclocked to 4.5GHz and ram is either stock or XMP.

I don't remember if in any of the bluescreen cases my GPU was overclocked manually to the maximum, but most of the cases the bluescreen happens it is running stock settings and GPU 3.0 does it's thing. 

The bluescreens have happened with my XBOX one wireless controller adapter connected and controller being used as well as unconnected.

Also with my steelseries headset connected and not connected (USB plugin). And with my stereospeakers  connected to a amplifier which is connected by a 3.5mm plugin.

 

Currently I've been running my system stock for all components with the mobo settings being on the stock settings aswell (ram 2133, CPU 3.5/3.9, GPU stock) and it's been probably a good 2 weeks now without any problems. Although not "enough" time has passed since my last BSOD to completely comfirm system stability even at stock settings.

 

Sorry for the wall of text, I tried to be as thorough as possible. TL:DR is at the top.

 

I'm planning on updating the BIOS if I get a bluescreen during stock run. But is there something that could improve my stability for the OC from this version of BIOS alone? LLC or other setting.

 

[Guest]

10 minutes ago, Constant Failure said:

TL:DR: Random bluescreens once a month happen, IRQL NOT LESS OR EQUAL, drivers up to date, specs in signature.

 

Let's start off what my system settings are:

ASUS z170-A Bios = 1702 (which was the 2nd latest at the time when built and OC'd)

i5 6600k stock/4.5GHz

corsair vengeance ddr4 ram 2133/3000MHz(XMP)

Win 10 64bit home

 

Overclocking was stable* from 24h stress test with h.264 encoding run as well as withstood IBT extreme for 2hours without exceeding 70C (might have been 65-68C at most but I don't remember since it was done in august) and has not crashed or bluescreened repeatably ever. 

My overclock was done by just changing the following parameters: Multiplier to 45, SVID( I think this was the name for it in this Mobo and BIOS) off, Vcore 1,25V (Manual) and for the RAM XMP profile. LLC was left at auto, for some reason during my stability tests using LLC 5 would be unstable after x amount of time but with auto the same settings would prove to be stable. 

 

*= random bluescreens appear approximately once a month, mostly with code IRQL NOT LESS OR EQUAL, maybe once(or twice) I remember having CLOCK_WATCHDOG_TIMEOUT.

 

Ok so there's two problems, may or may not be unrelated to each other.

 

First problem: 

If I turn off my PSU (switch at the back) for whatever reason, let's say thunderstorm. And then turn it back on and turn on the PC, my system tries to start, lights from mobo, CPU cooler and GPU flicker and everything turns off. After approx 2 seconds the system restarts itself properly like nothing happened and everything works fine after that.

 

This only happens when my CPU is overclock by the settings above.

Stock CPU, stock ram. Turns on at first try.

Stock CPU, XMP ram. Turns on at first try.

OC CPU, stock ram. Turns on, then off, then on again. repeatably.

OC both. Turns on then off and then on again repeatably.

 

The question is, is this a problem for my PSU (corsair RM650x) or is my LLC setting too low so the processor doesn't get enough juice at the start or something else?

 

Second problem:

These random bluescreens, and why I say random is that I cannot repeat them ever, they occur rarely and the error codes are the two what I listed above.

 

The times I remember these happening have been, watching a youtube video, playing GTA V, playing something I forgot Fallout 4 (vanilla). My temperatures for CPU while gaming are 50-60C and for the GPU around 60-72C (depending on the game) and during one more different game I can't remember (most likely crysis 3 or deus ex humankind divided).

 

I've tried narrowing down what could be the problem by updating my drivers in order of listing:

1. USB 3/3.1 and Chipset drivers

2. audio drivers

3. GPU (zotac amp 1070) drivers

while the problem is still randomly persisting. Between a step I always waited until the bluescreen occurred before updating the next drivers.

 

I've also had this happen when my CPU is overclocked to 4.5GHz and ram is either stock or XMP.

I don't remember if in any of the bluescreen cases my GPU was overclocked manually to the maximum, but most of the cases the bluescreen happens it is running stock settings and GPU 3.0 does it's thing. 

The bluescreens have happened with my XBOX one wireless controller adapter connected and controller being used as well as unconnected.

Also with my steelseries headset connected and not connected (USB plugin). And with my stereospeakers  connected to a amplifier which is connected by a 3.5mm plugin.

 

Currently I've been running my system stock for all components with the mobo settings being on the stock settings aswell (ram 2133, CPU 3.5/3.9, GPU stock) and it's been probably a good 2 weeks now without any problems. Although not "enough" time has passed since my last BSOD to completely comfirm system stability even at stock settings.

 

Sorry for the wall of text, I tried to be as thorough as possible. TL:DR is at the top.

 

I'm planning on updating the BIOS if I get a bluescreen during stock run. But is there something that could improve my stability for the OC from this version of BIOS alone? LLC or other setting.

 

I think your overclock wasn't stable. 4.5Ghz with 1.25V seems too good. Try 4.5 Ghz on 1.3V and report back?

[Constant Failure]

19 hours ago, FTL said:

I think your overclock wasn't stable. 4.5Ghz with 1.25V seems too good. Try 4.5 Ghz on 1.3V and report back?

It's just that the time frame of the BSODs is so random and far apart that I was hoping there could be some other factor I'm not fully utilizing than just voltage.

[Guest]

1 hour ago, Constant Failure said:

It's just that the time frame of the BSODs is so random and far apart that I was hoping there could be some other factor I'm not fully utilizing than just voltage.

If it works at stock, but crashes if you oc that defitnely means that your overclock wasn't stable. Have you tried 1.3V?

You said that your PC at stock hasn't been BSOD'ing since 2weeks (it's enough time). If that's the case, it's defitnely some stability issue.

BSOD can also happen from software issues, not all problems have to come from your hardware.

[Constant Failure]

31 minutes ago, FTL said:

If it works at stock, but crashes if you oc that defitnely means that your overclock wasn't stable. Have you tried 1.3V?

You said that your PC at stock hasn't been BSOD'ing since 2weeks (it's enough time). If that's the case, it's defitnely some stability issue.

BSOD can also happen from software issues, not all problems have to come from your hardware.

I haven't tried it yet, I'll run it stock for a few more days until I have time to redo my stability tests with higher voltages. But the problem is that the stability issues even with 4.5ghz 1.25V is random and rare so I can't really reproduce any of the BSODs.

[Guest]

It isnt random, it happens when you draw too much power.

[Constant Failure]

2 hours ago, FTL said:

It isnt random, it happens when you draw too much power.

Aside from the youtube video bluescreen I would agree. But applying more voltage to the OC wouldn't solve the other problem or should it?

[Guest]

1 hour ago, Constant Failure said:

Aside from the youtube video bluescreen I would agree. But applying more voltage to the OC wouldn't solve the other problem or should it?

In theory it should solve your problems.

[Constant Failure]

On 17/12/2016 at 3:50 PM, FTL said:

In theory it should solve your problems.

Well,

 

I did some testing.

IBT, stock or OC (4.5GHz, 1.3V, LLC6 ), 30minutes doesn't crash.

Prime95 (28.7) same deal

Both of the above also survived having 2133mhz and 3000mhz(XMP) ram speeds

 

x264 stability test v2.06, crashes almost instantly, both OC and stock.

 

BSOD caused by x264 ->  IRQL NOT LESS OR EQUAL

 

However, during summer when I overclocked previously x264 worked fine ( the same version) and passed 12h stresstest with 4.5ghz 1.25v instead of 1.3v used now

[Constant Failure]

@FTL, update:

 

I probably found out what the problem was with x264. My pc passed 1 loop with it at abovenormal priority (16 threads). But crashes about a minute into with high priority. Even the guide from the user says to use normal priority to check the stability. Going to run 12h stability test tonight with the "stable" settings from IBT and prime95 OC trials.