Sign-in / Sign-up
Your question
Solved

BSOD on new build. Corsair CMX8GX3M2A1333C9

Tags:
  • Blue Screen
  • CPUs
  • Corsair
  • Memory
Last response: in Memory
a b à CPUs
January 5, 2014 10:04:17 AM

I just built a new 4770k with 16gig of corsair memory and have been getting regular blue screens. Last night while I was sleeping I ran memtest and it completed 5 passes in 8 hours. Below are the results. Does anyone have any suggestions on how to proceed at this point?

Summary

Report Date 2014-01-05 10:36:11
Generated by MemTest86 V5.0 Free (64-bit)
Result FAIL

System Information


EFI Specifications 2.31
CPU type Intel Core i7-4770K @ 3.50GHz
Measured CPU Speed 3492 MHz
# Logical Processors 8
L1 Cache 64K (208379 MB/s)
L2 Cache 256K (54359 MB/s)
L3 Cache 8192K (38424 MB/s)
RAM 16439M (16218 MB/s)
DIMM #0 4GB DDR3 PC3-10600
Corsair / CMX8GX3M2A1333C9
9-9-9-24 / 666 MHz / 1.5V
DIMM #1 4GB DDR3 XMP PC3-10600
Corsair / CMX8GX3M2A1333C9
9-9-9-24 / 666 MHz / 1.600V
DIMM #2 4GB DDR3 PC3-10600
Corsair / CMX8GX3M2A1333C9
9-9-9-24 / 666 MHz / 1.5V
DIMM #3 4GB DDR3 XMP PC3-10600
Corsair / CMX8GX3M2A1333C9
9-9-9-24 / 666 MHz / 1.600V

Result summary


Test Start time 2014-01-05 02:19:13
Elapsed Time 8:16:21
Memory Range Tested 0x0 - 42F000000 (17136MB)
CPU Selection Mode Round Robin
# Tests Passed 50/58 (86%)

Lowest Error Address 0x4360370 (67MB)
Highest Error Address 0x42E5206B0 (17125MB)
Bits in Error Mask 0000000000E00000
Bits in Error 3
Max Contiguous Errors 1

Test # Tests Passed Errors
Test 0 [Address test, walking ones, 1 CPU] 6/6 (100%) 0
Test 1 [Address test, own address, 1 CPU] 6/6 (100%) 0
Test 2 [Address test, own address] 6/6 (100%) 0
Test 3 [Moving inversions, ones & zeroes] 5/5 (100%) 0
Test 4 [Moving inversions, 8-bit pattern] 5/5 (100%) 0
Test 5 [Moving inversions, random pattern] 5/5 (100%) 0
Test 6 [Block move, 64-byte blocks] 2/5 (40%) 10
Test 7 [Moving inversions, 32-bit pattern] 5/5 (100%) 0
Test 8 [Random number sequence] 5/5 (100%) 0
Test 9 [Modulo 20, ones & zeros] 0/5 (0%) 72
Test 10 [Bit fade test, 2 patterns, 1 CPU] 5/5 (100%) 0

Last 10 Errors
[Data Error] Test: 9, CPU: 7, Address: 1F88A0058, Expected: 2DCF255D, Actual: 2D8F255D
[Data Error] Test: 9, CPU: 7, Address: 1E08E00D8, Expected: D55EEEDE, Actual: D51EEEDE
[Data Error] Test: 9, CPU: 7, Address: 1D07C00D8, Expected: 9D4AD315, Actual: 9D0AD315
[Data Error] Test: 9, CPU: 7, Address: 1A4AA0058, Expected: 215074B0, Actual: 211074B0
[Data Error] Test: 9, CPU: 7, Address: 190380298, Expected: 254D8E60, Actual: 250D8E60
[Data Error] Test: 9, CPU: 7, Address: 1707C00D8, Expected: 7F5B181F, Actual: 7F1B181F
[Data Error] Test: 9, CPU: 7, Address: 15C7801D8, Expected: 43C60CC5, Actual: 43860CC5
[Data Error] Test: 9, CPU: 7, Address: 15C780058, Expected: CECEFC01, Actual: CE8EFC01
[Data Error] Test: 9, CPU: 7, Address: 1547C0498, Expected: FAB52645, Actual: FAF52645
[Data Error] Test: 9, CPU: 7, Address: 150380298, Expected: E0496E33, Actual: E0096E33

Certification


This document certifies that the Tests described above have been carried out by a suitably qualified technician on the System described above.

More about : bsod build corsair cmx8gx3m2a1333c9

a c 1388 à CPUs
a c 5048 } Memory
January 5, 2014 2:11:21 PM

OK you've got a 4770K and 4 sticks of 4GB 1333 Corsair DRAM and it appears like you bought two packages of 2x4GB sticks - is this correct? IF so, any time you mix packages of DRAM it can be problematic, even with 2 of the exact same set....would suggest try running them with the DRAM voltage manually set to 1.55 or 1.6 and see if that straightens tings out first, may not be a bad stick, may just be the two sets don't want to play
m
0
l
a b à CPUs
January 5, 2014 4:18:23 PM

The CPUID shots of the SPD tab. Not even close considering the same part number. 1 of the 2 does not even have XMP.

The 2 modules with XMP alone seem to work get prime 95 stable for an hour (I know its short). The 2 modules without XMP alone start failing workers in prime after about 30 seconds. This is running 9-9-9-24 at 1.5 volts at 1333/666mhz. After reading one of your other posts I did notice the non working pair is using a tRFC of 174 while the good pair use 107

https://skydrive.live.com/embed?cid=D5BC9A6BC8B01533&re...

https://skydrive.live.com/embed?cid=D5BC9A6BC8B01533&re...
m
0
l
Related resources
a c 1388 à CPUs
a c 5048 } Memory
January 5, 2014 4:57:57 PM

Was this two packages of 2x4GB?
m
0
l
a b à CPUs
January 5, 2014 5:10:11 PM

Yes, It was 2 separate kits of 2x4GB. I had used multiple kits before and never had this issue... too bad I didn't know before now.
m
0
l
a c 1388 à CPUs
a c 5048 } Memory
January 5, 2014 5:34:58 PM

Try raising the DRAM voltage as mentioned and let me know what mobo
m
0
l
a b à CPUs
January 5, 2014 6:03:57 PM

The motherboard is a Gigabyte GA-Z87X-UD3H. I increased the voltage to 1.6 and still no luck. 3 of the sticks appear to work (very short tests) with one failing prime after only a couple of minutes. The link to my memtest run from last night posted below (5 passes in 8 hours 15 mins). I have 2 more slots to give each a 1 hour test however currently it looks as though I have 1 bad stick in the version 7.21 kit (no available xmp and runs tRFC 174).

https://skydrive.live.com/redir?resid=D5BC9A6BC8B01533!207&authkey=!AKh1BJWPh0b3siA&ithint=file%2c.html
m
0
l

Best solution

a c 1388 à CPUs
a c 5048 } Memory
January 5, 2014 6:07:42 PM

Might just check the single stick with memtest, if errors then prob RMA time
Share
a b à CPUs
January 5, 2014 6:22:20 PM

I have not ran memtest86 on the single stick in my 4770k however memtest running on my old i7 920 on the single stick has yet to find a problem with the stick.
m
0
l
a b à CPUs
January 5, 2014 9:22:06 PM

Update,

With memtest86 running for the last 3 hours on the suspected "faulty stick" it has completed 3 full passes and is finished test 9 on pass 5. According to memtest this stick appears to be just fine however as soon as I install it for realtime use the crashing begins. All motherboard slots seem to be good as I have tried moving the other sticks around and everything seems to behave well. Looks like its RMA time unless you have any other ideas.
m
0
l
a c 1388 à CPUs
a c 5048 } Memory
January 5, 2014 9:30:18 PM

One more, leave DRAM voltage as is and try raising the the VCCSA (System agent voltage) to about 1.1
m
0
l
a b à CPUs
January 5, 2014 10:28:37 PM

In terms of a -0.3 to +0.4 adjustment range how far do I go?
m
0
l
a c 1388 à CPUs
a c 5048 } Memory
January 6, 2014 7:38:01 AM

Either is safe, try the + 0.04, if stable could try dropping or go up a bit more from there
m
0
l
a b à CPUs
January 6, 2014 7:17:40 PM

Thanks for all the help, that stick is a lost cause. Submitted RMA for replacement memory kit.
m
0
l
a c 1388 à CPUs
a c 5048 } Memory
January 6, 2014 7:28:24 PM

Hope the RMA goes quick, drop us an update when you get the replacement
m
0
l
a b à CPUs
January 6, 2014 7:45:51 PM

It should, the place I order from is usually very good and the tech support person said they will RMA it... at least I still have 8 gigs.
m
0
l
a c 1388 à CPUs
a c 5048 } Memory
January 6, 2014 9:10:53 PM

Will be awaiting your update, hope all goes quick and well ;) 
m
0
l