Stability Problems with P7N Diamond Continues

Question

Big bummer. I've had this system since April, took a month to get it working as the first board I ordered was bad, this one seems to be a bum board. I've came here and a few other places, but help has been kind of hard to find.
Here's what I got:
MSI P7N Diamond
Intel QX9650
2x 2GB mushkin DDR2 1066 (XP2-8500) Timings: 5-5-5-15 2T
I have 4 HD's: all are seagate baracudas: 250, 500, 750, 1TB
MSI NX8600GT1GEZ
At this point overclocking is out of the question. All I want is a reliable PC that runs to its specifcations. RMA-ing this board is a last resort as often enough the RMA boards aren't worth a crap anyway, dont know if this is just the case with lower-end boards, but I hate having to wait, and I'm really ticked about the whole thing.
The computer reboots at random, I've found "more stable" configurations, but usually the machine will not stay running for more than 24 hours without a reboot. The most stable configurations are with the official 1.2 BIOS, setting the NB to 1.250 v, CPU Volt: Auto, SB: Auto, DRAM: 2.10, and I have manually set the memory timings: CL: 5, trcd: 5, trp: 5, tras: 15, and Command Rate: 2T. The FSB Speed: 1333, and the DRAM Speed: 1066. Temps are normal, and range from about 100 F at idle, and 120 F load on the CPU.
The voltage on the CPU is on the high side on Auto, the QX9650 can run on as little as .98 v at stock and this board is giving it 1.23 v on auto. The board also tries to set the timings on the memory to 7-7-7-24, which I feel has something to do with it.
I'm not an overclocker, I don't know what all these voltages do, I've only changed settings that have either been reccomended or have a warning setting (ie turn red).
I'll try a beta BIOS if one is available, but I was unaware that this board doesn't fully support the 45-nm CPU's.
There is NOTHING wrong with any of the hardware (except the board perhaps), I've tried other memory, PSU, even RMA the CPU, does the same thing. Everything has been tested, every possibility I can imagine, this problem has been haunting this system for months, and I'm almost to the point of buying a cheap as board that has some credibility that my hardware will work.
This is my most critical computer, I do everything on it, but I didn't buy a $300 motherboard to have these problems, or find that my CPU scores half of what it should when it is running stock, everything on this board is below stock, benchmarks are way below what they should be.
I'm up for some ideas, suggestions, if needed, I'll RMA the board, but it the return board does the same thing, I've lost all my faith in MSI. My last system had a P4M900M2 and it kicked butt, and it's a $50 board!  Go figure these boards are worse than imported cars.

Answer 1

Quote from: NovJoe on 27-September-08, 12:40:17
By means of instability, can you describe the symptons of events that happen?
Can I have a list of the PSU models and RAMs models which you have tested?
Have you try the hardware on another board? If you have done so, what are the results?
I have not tried any of my hardware on this board, I have tried other memory on this board, this is the only board I have avialable to me that support LGA775 so I'm stuck between a rock and a hard place on that. The memory I tired, couldn't tell you the exact model, but it was Apacer DDR2 800, and was listed under this board as tested memory, i do remember that much- if that helps any.
As far as overclocking, I was told that overclocking my rig would make it stable, which i feel in this case is a little more than bumping the FSB eitherway. But no this happens under "stock" settings. Orginally with the 1.0 BIOS the machine would sometimes give me a multitude of BSOD, CLOCK_INTERRUPT (A clock inttrupt was not recieved in a timely manner on a secondary processor), DRIVER_IRQ_NOT_LESS_OR_EQUAL.
Only way to get the machine to run "stable" is to severely underclock, with the FSB at 1137 and DRAM at 1034.
Other things. Benchmarks (SiSoft Sandra Lite 2007)
CPU: Drystone: 39000 MIPS / Whetstone: 28000 MFLOPS
Memory: 7 GB/sec (1066 with correct timings)
Both appear to be under par, CPU particularly, and in the event of tinkering, only see a jump in performance when adjusting the multiplier from 9 to 10.5, which isn't stable, stable enough to run a benchmark. So I'm not sure where the problem is.
This might help, the original thread from back in June.
https://forum-en.msi.com/index.php?topic=117779.0