F3-12800CL9D-4GBRL (2 kits) + ASUS P7P55D-E PRO *Win 7 BSODs*

[joshuachin888]

Hello,

I've recently purchased a computer with the following specs:

Motherboard: ASUS P7P55D-E PRO (BIOS version 0610)
Memory: 2 kits of F3-12800CL9D-4GBRL (8 gb, 4 DIMMs total)
CPU: Intel Core i7 860
OS: Windows 7 Home Premium

I've been occasionally (at a rate of 1-2 times per day) getting BSODs in Windows 7 (especially within minutes after loading into Windows from a cold boot) with the following bug check strings:

0x0A IRQL_NOT_LESS_OR_EQUAL
0X7E SYSTEM_THREAD_EXCEPTION_NOT_HANDLED
0X1A MEMORY_MANAGEMENT
0X4A IRQL_GT_ZERO_AT_SYETEM_SERVICE
0X1E KMODE_EXCEPTION_NOT_HANDLED
0X3B SYSTEM_SERVICE_EXCEPTION
0XD1 DRIVER_IRQL_NOT_LESS_OR_EQUAL

These crashes occured with Ntfs.sys and ntoskrnl.exe, and I suspect it is due to the memory settings. Memtest86+ detected no errors after 9 passes with the below settings.

I'm using the factory settings for the BIOS (although I've set the DRAM freq. to DDR3-1333MHz which is probably the same as the factory defaults). It's also my first time running a system with all DIMM slots filled, so I don't know if e.g. the voltages will have to be tweaked.

Here's the BIOS settings and what CPU-Z sees:

Ai Tweaker

CPU Level Up – [Auto]

Ai Overclock Tuner – [Auto]
Intel(R) SpeedStep(TM) Tech – [Enabled]
Intel(R) TurboMode Tech – [Enabled]
Xtreme Phase Full Power Mode – [Auto]
DRAM Frequency – [DDR3-1333MHz]
QPI Frequency – [Auto]
ASUS/3rd Party UI Priority – [ASUS Utility]

OC Tuner – [Turbo Profile]

DRAM Timing Control *every parameter set to [Auto]*
1st Information: 9-9-9-24-4-74-10-7-20
2nd Information: 2N-52-53
3rd Information: 5-5-16-10-10-10-7-6-4-7-7-4

CPU Differential Amplitude – [Auto]
CPU Clock Skew – [Auto]

CPU Voltage Mode – [Offset]
Offset Voltage – [Auto]
Current CPU Core Voltage – [1.080V]
IMC Voltage – [Auto]
Current IMC Voltage – [1.118V]
DRAM Voltage – [Auto]
Current DRAM Voltage – [1.550V]
CPU PLL Voltage - [Auto]
Current CPU PLL Voltage – [1.803V]
PCH Voltage – [Auto]
Current PCH Voltage – [1.050V]
DRAM DATA REF Voltage on CHA – [Auto]
DRAM DATA REF Voltage on CHB – [Auto]

Load-Line Calibration – [Auto]
CPU Spread Spectrum – [Auto]
PCIE Spread Spectrum – [Auto]

Advanced
CPU Configuration
CPU Ratio Setting – [Auto]
C1E Support – [Enabled]
Hardware Prefetcher – [Enabled]
Adjacent Catch Line Prefetch – [Enabled]
Max CPUID Value Limit – [Disabled]
Intel(R) Virtualization Tech – [Enabled]
CPU TM Function – [Enabled]
Execute-Disable Bit Capability – [Enabled]
Intel(R) HT Technology – [Enabled]
Active Processor Cores – [All]
A20M – [Disabled]
Intel(R) SpeedStep(TM) Tech – [Enabled]
Intel(R) TurboMode Tech – [Enabled]
Intel(R) C-STATE Tech – [Enabled]
C State package limit setting – [Auto]

My goal is to run a stable system with minimal (or no) overclocking. I'm quite the novice when it comes to managing memory settings so I hope I can find the help I need in these forums. Many thanks to those who contribute to this thread!

Cheers! XD

Josh

[GSKILL TECH]

Increasing the IMC Voltage should stabilize the system. Don't forget to also raise DRAM Frequency to DDR3-1600.

Thank you
GSKILL TECH

[joshuachin888]

I've upped the IMC voltage by 2 notches (i.e. from stock 1.100V to 1.118V) and set the DRAM frequency to DDR3-1600 by enabling XMP. So far, no problems, but I'll have to cold boot for the next few days (I turn off my PC during the night) to be certain. My CPU runs ~5 degrees C hotter as expected, but isn't a problem for my cooling system. I'll also be runnning memtest86+ now. Thanks for the prompt reply!

Josh

*Update: It appears that running the sticks at DDR3-1600 causes errors by memtest86+ when tested with IMC = 1.10000V and 1.11875V...will I need to crank IMC voltage up even more and by how much? (By the way, is there a reasonably safe upper limit to the IMC voltage or a max level that will not void the CPU warranty?) I have since lowered the DRAM frequency to DDR3-1333 while keeping the IMC voltage up at 1.11875V (no errors by memtest86+). I'm still testing out the cold boot issues.

[GSKILL TECH]

IMC Voltage may need up to 1.30V. That is Intel's max recommendation.

Thank you
GSKILL TECH

[joshuachin888]

Thanks for the reply. I'll be testing at that voltage. Will I also need to change the CPU voltage mode (from offset to manual) or increase the DRAM voltage to stabilize 4 DIMMs?

[GSKILL TECH]

CPU Voltage should be Manual or Absolute. DRAM Voltage should not need more voltage than rated.

Thank you
GSKILL TECH

[joshuachin888]

Unfortunately, the cold boot BSODs still persist after upping the IMC voltage to 1.30V.

[joshuachin888]

Okay, this morning, I started running each of the two kits (2x2gb = 4gb total) separately under Memtest86+. I've kept all parameters constant between both kits (i.e. stock voltages, occupying slots A1 and B1 of the motherboard, stock timings, etc.) and ran 1 pass in Memtest86+. The results are as follows:

Kit #1
DDR3-1333: Memtest86+ fail, many errors
DDR3-1600 (XMP): Memtest86+ fail, many errors

Kit #2
DDR3-1333: Memtest86+ pass
DDR3-1600 (XMP): Memtest86+ pass

It appears from the results that Kit #1 contains faulty RAM (I didn't have time yet to test the modules individually) and might be eligible for an RMA.

Just a small question: My previous test for all 4 sticks together (8gb total) passes Memtest86+ under DDR3-1333 but fails DDR3-1600. Is it common that 4 sticks passes Memtest86+ but (as in my case) fails when testing the individual kits?

Another question: When changing from DDR3-1333 to DDR3-1600, will I need to boost the IMC voltage when working with 2 sticks?

[GSKILL TECH]

Yes, with 4 modules installed, many times the memtest won't detect the error. This is why we ask people to test individually.

From DDR3-1333 to DDR3-1600, it will require more IMC Voltage, but the motherboard is able to automatically support that which is why "AUTO" should be fine. Only DDR3-2000+ may require manually setting the IMC Voltage. With your specific combination, we tested 8GB of DDR3-2200 without a manual setting. One of the easiest combinations to reach high performance.

For the bad kit, simply send it in for new replacements.

Thank you
GSKILL TECH