Motherboard is probably ok. Memory is likely the problem. If you check the spec sheets of the RAM you mentioned:
According to Corsair CMX4GX3M2A1600C8 requires 1.65v
with 8-8-8-24-2T timings
According to Kingston
Kingston KVR1333D3N9K2/4G requires 1.5v
9-9-9-??-?? timings (not specified)
These modules are very different are not desgined to run together as a dual channel kit!
But, to get it to work you can try:
_Manually_ Set the ram timings to the slowest settings.
So, Speed: 1333; timings: 9-9-9-24 (or slower), tRAS of 2T or AUTO and voltage to 1.5v
If that works, try Dual channel. Not work?
Try increasing DRAM voltage one increment at a time (0.02v or something) until it posts with dual channel; do not exceed to 1.65v. Any higher and I think you need to change the uncore voltage to 0.5 V within the DRAM voltage. But you should research that. I don't know if the Kingston KVR1333D3N9K2/4G will handle this speed.
Run memtest86+ to check RAM stability (search web for Hiren's Boot CD for a Command Line version of this tool on a bootable CD). If you get errors, add a little more voltage or slower timings.
My apologies for not being clearer. I wrote my first post in haste. To clarify, I first tried a set of Corsair sticks and I adjusted the voltage to 1.6v (the max condition my mb was guaranteed to not be damaged) at 8-8-8-24-default. I recently RMA'd the sticks after working with the Corsair tech team trying to get them to work under the assumption that the ram was the problem. They eventually suggested my mb could be at fault, but I decided to try different ram first. This is what brought me to the Kingston sticks. I popped them in, running the default memory settings and I got the same results. I concluded this:
whenever I try to run two approved sticks of ram that are exactly the same on the same channel, my computer won't even boot into bios.
I will try tweaking bios settings with the Kingston sticks and report back with my findings.
*** Update ***
My problem is not solved. I have returned my mb to its default settings.
Try loosening the timings even more... 9-9-9-25 maybe..looser and 1.6v and see. The Kingstons should post fine at 9-9-9-25, i would think.
Other than that, I also wonder if the power supply is weak...or
Perhaps clearing the CMOS and trying from scratch, setting the timings manually and loosely at first will be the last resort before calling up Intel and begininng the RMA process. At this point, most of the obvious stuff is out of the way.
Maybe this board has its own particular quirks we have yet to discover.
I am using G.SKILL Ripjaws Series 4GB (2 x 2GB) 240-Pin DDR3 SDRAM DDR3 1600 (PC3 12800) Desktop Memory Model F3-12800CL9D-4GBRL in two of my systems with this motherboard with no issues. I am running them at 1600 MHz with the XMP settings and at 1.5 volts.
UPDATE: My bad, I thought you were using the 2x4Gb memory kit. Ignore what I say below.
I'm about to purchase the same combination of motherboard and memory modules, have you found out if these work together yet? The other Kingston modules from the KVR1333D3N9 series seems to work, both in kit a single versions. Also people seems to be able to use 2 kits of KVR1333D3N9K2/4G.
The 8Gb kit (KVR1333D3N9K2/8G) consisting of 2x4Gb modules is built from 16 x 256Mb chips per DIMM module, while the 4Gb kit (KVR1333D3N9K2/4G) is built from 16 x 512Mb chips per DIMM module, both in 8 banks. So no difference there.
However I noticed that the tRFC is greatly different for the two modules:
KVR1333D3N9K2/8G: Refresh to Active/Refresh Command Time (tRFCmin) 160ns
KVR1333D3N9K2/4G: Refresh to Active/Refresh Command Time (tRFCmin) 90ns
Time for an update. I got my replacement motherboard from my retailer today and everything's working just fine. I had a defective board. Thanks for all the comments.