GA-X58A-UD3R (Rev 2.0) - Hard Drive corruptions and BSoD

[Galastorm]

Please help (also posted on Toms Hardware Gigabyte area). This has been a problem for about a week now and I'm about to send the GA-X58A-UD3R (Rev 2.0) back to Ebuyer as I'm almost certain it's faulted in some way.

Just to be clear, I'm *not* trying to OC the motherboard in anyway - I'm using the MB Optimal settings with the exception of the memory which is set using XMS profile 1. All I want is a stable system - standard settings are more than ample enough for what I intend to use it for.

It looks like I keep getting random disk corruption failures:
- occasionally (about 50% of the time) the MB reports that there is a failure due to "Incorrect overclock or voltage settings" (or similar). This, even though nothing has changed and there is nothing over clocked.
- many occasions, windows refuses to install (hangs on re-start while updating registry)
- many occasions, it manages to get all the way through installation to the point of installing MB drivers then, on re-start, I get an unrecoverable BSoD
- On a couple of occasions the machine installed Windows updates then BSoD with disk-check disk corruptions.
- on one occasion the system worked fine (all MB and graphics drivers installed correctly and re-boot was ok) until I tried to installed a game (Dungeons and Dragons Online). After install, the game then reported that it's data files were corrupt and needed to be re-installed. On re-start check disk found a huge amount of corrupted file entries which it fixed - but the system just repeatedly cycled disk-check and then BSoD.
- on another occasion everything installed ok and the game played without an error for around 3hours (Dx10, 1920x1200, Max settings). Then, on power down and power up, the MB believed it had been OC'd again, the file system was corrupted and the BSoD cycle repeats - this is the point at which I am writing this...

My system is this:
1 x Gigabyte GA-X58A-UD3R Intel X58 (Socket 1366) DDR3 8 Channel Audio ATX Motherboard
1 x Corsair 6GB (3x2GB) DDR3 1600MHz XMS3 Memory Kit CL9(9-9-9-24) 1.65V
1 x Intel Core i7 930 2.8GHz Socket 1366 8MB L3 Cache OEM Processor
1 x Corsair CWCH50-1 Cooling Hydro Series H50 High-performance CPU Cooler
1 x Gigabyte GTX 460 768MB GDDR5 Dual DVI Mini HDMI PCI-E Graphics Card
1 x Corsair 850W TX Series PSU - 140mm Fan - 5 Year Warranty
1 x Western Digital 1TB Hard Drive SATAIII 7200rpm 64MB Cache - OEM Caviar Black
1 x Microsoft Windows 7 Home Premium 64-bit - Licence and media - 1 PC - OEM - DVD - English
1 x Samsung SH-S223 22x DVD±RW DL & RAM SATA Optical Drive - OEM Black
1 x Coolermaster CM 690 II Advanced Mid Tower Case
1 x FractalDesign - 120mm Silent Series Cooling Fan
2 x FractalDesign - 140mm Silent Series Cooling Fan

There were no signs of crush damage (as per another post) to the MB box.

The case has 5 fans installed (all directly connected to 12V power - except for front panel which is connected to MB fan 1):
- (incoming) 1x120(?) front, 1x140 Side (directly over graphics card and the X58 chipset) and 1x120 rear (with the H50 radiator)
- (exhaust) 2x140 in the top (plus the entire case is pretty much open-mesh)

Both the optical drive and the WD HDD are connected to the ICH10R (because I read there may be problems with the onboard SATA 3 chipset).

I've installed the HD in another machine and completed a 3hour full format with no problems reported.
I've flashed the MB board bios to the latest version (FB) and used the latest drivers for the MB and graphics card (as well as those drivers on disk with the previous version of BIOS).
I've tried only installing the Audio and Ethernet drivers (because on one occasion the problem occurred after installing the MB .inf file)
I've used other windows installation disks which I know to work fine (including Win 7 Home Premium 32bit to see if it was some sort of 64bit driver issue)
I've disabled windows updates to varying degrees (none, critical all etc) without a noticeable difference

I haven't installed temperature monitoring programs etc (it's hard enough getting windows installed) but the highest temps I've seen for the processors in the BIOS screen have been around 37 after a re-boot failure. On cold starts they're normally around 30-32.

Any thoughts or suggestions before I dismantle the machine and return the MB?

[Dark Mantis]

Hi and welcome to the forum. I am sorry that you seem to be having such a rough ride so far. I would suggest that it is down to one of two things, either bent pins on the CPU socket or a bad memory module. You will need to remove the processor to check the pins so I would advise checking the memory first. Download and run Memtest86+ and let it check each module seperately for several loops. Then swap to the next stick.

[fatbobuk]

Just to let you know your not the only one
But this only happens for me when I try to update the BIOS from FA ( default ) to any other BIOS

My Optimized defaults setting do not run my PC ( corrupted i think ) and It is hard to find settings that will run it

I do not have bent pins on my CPU socket.

[Dark Mantis]

Just to let you know your not the only one
But this only happens for me when I try to update the BIOS from FA ( default ) to any other BIOS

My Optimized defaults setting do not run my PC ( corrupted i think ) and It is hard to find settings that will run it

I do not have bent pins on my CPU socket.

Fatbob I still think that there is quite a possibilty that you have bent pins on your socket. I have just been back through your initial thread and it would certainly seem probable. You have all the indicators.

[Peteruk]

Intel does not back 1600MHz RAM speed to be 100% with ever CPU made thats why they list supported speeds so its worth checking the CPU at stock speeds as the RAM controller is in the CPU.

This isn’t to say that your CPU can't handle 1600MHz as it might just need tweaking but try it at stock speeds without XMP first to rule out the CPU RAM controller.

Other thing:
Put a jumper on to limit the HDD to 3Gb/s
http://wdc.custhelp.com/cgi-bin/wdc.cfg/php/enduser/std_adp.php?p_faqid=5387&p_created=&p_cats=185&p_cv=1.185&p_pv=2.279&p_prods=227%2C279#jumper

[Galastorm]

Mem test worked fine I used the earlier version and tested all memory modules at once.

After reading Peteruk's post about potentially needed to tweak 1600Mhz RAM I decided to go the whole hog and overclock the system...

So far I've the processor running at 3520MHz (HT off) and the memory at 1600Mhz - CPU temps of around 60-65 running Prime95.

Stupid thing is the system looks stable!  Will keep you informed and the working final freq/voltages for comment in a day or two if it stays up.