BSODs and hard crashes (Possibly the CPU)?

[EGM92]

Hi everyone,

I have a i7 2600k + 2 motherboards a Asus P8P67 Vanilla and a Asus P8Z77-V LK. For the longest time I've had nothing but CPU problems. I've RMA'd the CPU a couple of times, every time getting a new replacement and every time getting one with issues.

Here's what's happening:

Normal computer useage the PC would randomly BSOD mainly during gaming. I'd get 0x000001E errors with KMODE_EXCEPTION_NOT_HANDLED. On the odd time I would get 0x0000024 error codes. I would also get crashes related to Intelppm.sys. There are no CPU overclocks of any kind.

I was experiencing data corruption on my SSD, I even went out and purchased a new one and the problems still persisted. I replaced all SATA cables. Tested all SATA ports on the boards. Tested all the RAM sticks (4x4GB) one at a time. Disconnected all devices. I formatted the SSD, and now when I attempt to install windows on it, sometime, randomly during the install of Windows 7/8 I get the KMODE BSOD flash for a split second on screen then the system reboots. So now I'm unable to even install windows directly on that system.

I've tried the CPU on both boards and the same problem occur. Could this be my CPU? Is there anyway to diagnose it for sure? I'm pretty sure I'm out of warranty for the CPU and Intel got really crappy about me RMA'ing my last CPU even though they told me that one was from a batch with known issues.

I would appreciate any help/suggestions.

 

[enaberif]

At one point yes there was a known issue with the i7 2600k CPUs. A shop I worked in got a bad batch of them and we had to return quite a few.

What revisions are your boards? If they aren't the updated versions they were known for data corruption due to sata failure.

 

[EGM92]

All the boards are B3 revisions and have the latest available BIOS/UEFI. Is there any way to coax Intel into RMA'ing this CPU if there is a known fault?

 

[EGM92]

Thanks to those that PM'd me, I ended up getting a really nice CSR who listened to the steps that I went through and just skipped me right to an advanced RMA. Great service from Intel.

 

[bignick277]

Tested all the RAM sticks (4x4GB) one at a time.

While it's good you tested all 4 sticks one at a time to eliminate the possibility of RAM module being faulty, that doesn't mean the problem isn't with the memory. You also need to run a memtest 86+ with all 4 modules installed at the same time. I've seen it plenty of times where a computer would run fine with 1, 2 or 3 modules installed, but be unstable when running all 4 modules installed populating all 4 slots. I even had it happen to me. Specifically it was with an Asus Rampage Formula motherboard and 4x2GB sticks for Corsair DDR2 memory @ 1066Mhz. There was nothing wrong with the modules, but the motherboard just didn't like them populating all 4 slots. So I ended up having to remove 1 of the modules to stabilize my computer. It turned out it was actually a known issue with my motherboard and the memory modules I was using, which was surprising since my modules were specifically highlighted as being compatible with my board.

Also fair warning, when I had this issue, if I ran a memtest with all 4 slots populated from a cold start, it passed. But after playing minecraft for a while, my computer would BSOD and then if I ran the memtest again, it would fail. Then if I removed one module, any module and ran again, it would pass. Put the module in again and test. Fail again. Pull out a different random module and test again. Passed. This is what happened to me and this was how I finally found what was causing my stability issues. Memory was perfectly fine. My motherboard could use them correctly. My motherboard just didn't like having all 4 slots populated with that specific memory. So I ended up just running with only 3 slots populated until I did a core upgrade (motherboard, CPU, Memory) a year or 2 later.

Lesson learned. Test memory thoroughly in a number of different scenarios to completely rule out a memory issue. Hope this helps.