Any ideas?

Ri_Nollaig

I have recently built a new machine, mostly new hardware, was able to reuse an SSD. When it works, it works perfectly! 1440p goodness in BF4 or Metro Last Light and pretty much anything else.

Specs:
*Windows 8.1
*i7 4770k
*Asus maximus impact
*750w Seasonic x-series
*EVGA 780ti SC version
*16GB Ram [Corsair Vengeance 1600mhz using the XMP profile]
*128GB Samsung SSD [boot disk]
*512GB Samsung SSD and a 2TB seagate HDD for storage

Its all in a corsair 250D mITX case and using a h100i cooler. Latest and greatest drivers/firmware currently available where possible and latest BIOS on the motherboard too.

However, right from the start I have been getting random crashes. Sometimes blue screen [CRITICAL_SYSTEM_PROCESS] other times it just completely dies and I get the Kernel power loss failure in event log when I eventually get back into it. Sometimes it doesn't even boot back up properly and instead turns on and then off again after a couple of seconds, this is before it hits the POST screen. When this happens I need to clear the CMOS to get it going again.

And thats the strange thing! It can work for a week near perfectly, crash once and then not even boot on multiple attempts.
This can happen in a demanding game like BF4 or just idling on desktop. Has happened with XMP profile enabled and disabled and even in Windows Safe mode. Has even happened in the BIOS screen!

I initial though, it has to be hardware related, most likely the PSU, I know seasonic are a good name, hence why I got it but I thought I got a lemon, so replaced that. Issue still occurs.
Thought maybe it was the RAM, so ran memtest on a couple of passes, and the built in windows one. No errors ever detected. Tried with both the default BIOS config and the XMP profile. As they are 2 sticks of 8 gigs, I have tried with just 1 stick in the machine and got the same result.

I thought perhaps I got a dodgy 780ti as I took it out and used the integrated graphics for a while and had no issue.
I replaced it and for a solid week had no problems. But then the same issue happened yet again.
I pulled everything and put everything back together, ensuring all cables were fully secure and nothing was grounding, etc. Kept an eye on the temperatures aswell to ensure it wasn't overheating and the 3.3/5/12v rails, all in expected ranges. Worked without issue for another week.

But still the same issue seems to happened. The next step could be to replace the CPU/Motherboard but I'd rather not and it would really only be a guess anyway. The primary problem is I haven't narrowed it down to the exact cause and I am really only taking guesses at this point.

I have used various benchmarks too and have stressed the system to full load [100% on CPU/RAM/GPU]. Have left it run for a good few days. All without issue, and then, seemingly random it will crash and just will not start up properly again. The last time this happened was Saturday and it just would not start up properly again. I reseated the RAM and has been fine since, but it feels like its just a count down to when it will happen again. I am at my wits end from this...

Anyone think of sometime I have missed here, might be something very obvious that I overlooked

Luck100

Sounds like you've ticked all the reasonable boxes so far. Next likely candidate is the motherboard. I assume you are not overclocking? Have you set anything in the BIOS manually, or is it running entirely default settings? Try going back to defaults if you've changed anything. It could also be the cpu itself although that's a lot less likely that the motherboard.

Tea_Bag

sounds awful.

might be odd/stupid, but have you tried switching SATA ports for your OS SSD? they're in AHCI right?

i still suspect RAM. how long have you tested it for? some problems can only be found after many hours of memtest.

its usually easier to throw in a known working ram stick though.

any OC on the CPU? remove it.

Have you tried a fresh OS install?

Ri_Nollaig

Thanks for the suggestions

I had suspected it was either the RAM, Motherboard or CPU alright [in that order]. I have never had a CPU act like this so I would only every blame that once every other possibility has been exhausted.
The RAM does seem like the most likely suspect. Given that the issue can sometimes occur in even the BIOS!

I have never tried changing the SATA ports alright, I have ensured they are 100% pushed in and they are differently set to AHCI alright.
No manual OC on the CPU but I haven't actually tried a fresh install of the OS now that you mention it...

zeromemory

I used to have random crashes and BSOD like you even while Idling(browsing) or playing. Though i havent got a beast of a system like yours. I tried eliminating components by components to see which is causing problem... Guess what it's my SDD + HDD (raid 0). I was reading your post and you never mentioned about checking your SDD/HDD. Try checking them maybe that's was causing your random crashes. Complete reformat and fresh install as a last resort.

Ri_Nollaig

I checked the SATA ports and the boot drive is definitely in port 1. ACHI is actually enabled by default anyway and RAID is disabled. Have had this issue with 100% default BIOS settings.

Gonna just have to try another shot in the dark and see what happens with another set of RAM, hassle and its hard to verify that it has definitely fixed the issue but it really has to be in the scope of RAM/Motherboard/CPU. Thanks for all of the suggestions lads!

Ri_Nollaig

I replaced the RAM and still had the same issue.
I managed to not throw it out the window in angry so I got that going for me.

Gonna try replacing the motherboard now, in a "what else could it be?" move.

SterlingArcher

Didn't have time to read all. Ssd? Have you swapped it out.

Ri_Nollaig

Update:
I got the motherboard replaced and it has been two weeks without a random restart, so [touch wood...] I believe this issue is fixed. Annoying but atleast I got to the bottom of it.