Very recently built PC having start up problems

Associate
Joined
14 Nov 2015
Posts
18
Hi

I'm hoping to get some help/advice. My new PC is currently unusable. I bought these parts late November and put everything together in December. I haven't had a chance to use it much until recently to my great displeasure. A few weeks ago it would crash when browsing the web and the new Windows 10 update never seemed to take since then problems have been getting worse. I started getting a BSOD before login, then could get into the BIOS but now it only sometimes passes POST. The fans and lights always come on so not entirely dead!

Technical info below:

---Build---
Asus Z170 Pro Gaming (Flashed to latest version from website yesterday)
Intel Core i5-6600K
Noctua NH-D15S
XFX Radeon R9 390X
Corsair Vengeance LPX 8GB (2x4GB) DDR4 PC4-24000C15 3000MHz
Samsung 850 EVO 512GB SSD
Corsair RM750 Gold Modular PSU
Fractal Design Define R5

---BSOD codes--- (In a rather rough order of appearance)
Kernal security check faliure
Page fault in non paged areas
IRQL not less or equal
KMODE exception not handled
System service exception
0xc000021a (That was it)
Driver corrupted expool
system thread exception not handled
bad pool caller
resource not owned
driver overran stack buffer
bad system config info
-- At this point most codes don't get the whole message in, leading to a BSOD which only consisted of an eerily poetic
":(
y"


--What I have done about--
Arguably the important bit, well so far I have disconnected my disc drive and HDDs leaving only my SSD, although I have tried just booting to USB without that either. I updated the BIOS and took out the CMOS battery. I have taken the graphics card out which is sitting on an antistatic bag as I type. Strangely some of the times when it failed to POST the MB LEDs suggested a VGA problem. The only USB connections the the MB are to the empty case ports, kb&m and my memtest86 bootable USB drive, only other things plugged in are the HDMI to the MB, and power.

CPU temps seem fine, 30 in BIOS 33/35 in memtest which is running using only one core. I have tried each RAM stick separately and in different slots as well as running it at stock and turning on XMP for the first time. Due to the age of the hardware I don't know anyone I can swap parts out for except the graphics card or perhaps PSU.

Seeing as it now sometimes fails to get to the BIOS and would crash when booting to the Windows 10 recovery USB I made I'm assuming it is a hardware (or at least BIOS) issue. I am also assuming it isn't my graphics card or SSD. Unless both RAM sticks have failed it is unlikely to be that, and testing one of them is showing no errors halfway through pass 2 of memtest86. Although it does seem to look like a memory related issue.

Have I missed anything obvious so far? Anything else I should try before taking everything out/apart and building it again from pieces on my kitchen table? As I said I may have an old PSU laying around somewhere which I think works. If I can't get it to work what is the best way to tell what part I need to return/RMA?

Thanks to anyone who can help!
 
Associate
OP
Joined
14 Nov 2015
Posts
18
No errors on a 4 pass memtest on just core 0 and round robin (switches cores each test) on one of the sticks. I want to try the all CPU mode next although I am having trouble getting it to POST. The graphics card is out but sometimes the vga error light on the mb stays on.
Whilst all the errors point to the RAM I have a feeling it may be the mb.
 
Joined
5 Oct 2008
Posts
8,977
Location
Kent
Have you tried resetting the CMOS and then try to boot with default settings?

I would also recommend trying a RAM stick from elsewhere if the above does not solve the issue and see if the machine at least becomes usable.

The PSU being iffy can also cause issues so maybe try a spare if you can get hold of one?

Both RAM and PSU ruled out may indeed indicate the motherboard is at fault. It should however still be in warranty.
 
Associate
OP
Joined
14 Nov 2015
Posts
18
i would say its the ram, Sometimes faulty RAM can pass memtest
How likely is it that both sticks failed? Or do they tend to fail in batches? I don't have any I can switch it out for unfortunately.

Overclocked? If yes reduce OC slightly.

Try a notch higher voltages in 1) CPU vcore or 2) Memory.
Try slackening memory timings.
Nope, I wasn't planning on overclocking until it started to struggle and I'd worn it in so to speak. Only thing I touched as yet was turning on xmp to see if that stabilised the system, didn't seem to make a difference.

Have you tried resetting the CMOS and then try to boot with default settings?

I would also recommend trying a RAM stick from elsewhere if the above does not solve the issue and see if the machine at least becomes usable.

The PSU being iffy can also cause issues so maybe try a spare if you can get hold of one?

Both RAM and PSU ruled out may indeed indicate the motherboard is at fault. It should however still be in warranty.

I have tried resetting the CMOS, only way I can seem to get to the BIOS now. I don't know anyone in my area with DDR4, I'm completely out of sync with my friends upgrade cycles. I do have any old PSU lying around and tried that but it made absolutely no difference.

Apparently with Skylake the CPU can throw up memory errors, I'm in the process of returning the MB and CPU for fault testing soon. I'll post back with results. They are all still in warranty but I should have found time to do some stress testing much sooner to when I put it together!
 
Associate
OP
Joined
14 Nov 2015
Posts
18
Just for the sake of closing the circle so to speak. I returned both the mobo and CPU. It arrived and both were tested on the same day and they found the CPU was the cause and is now on its way back to Intel.
 
Back
Top Bottom