Diagnosing boot failure. Suggestions please.

[kawzman]

I have various success booting up my desktop pc. Most of the time it will self reboot at the bios slash screen or during the memory test. When it does this reboot the screen remains black and does not attempt to post. Occasionally when I can get the OS to load it will freeze at the login screen. On the rarer occasions that I can actually login, it eventually self reboots or more often freezes with garbeled colors on one or more monitors (have 3).

I've attempted to update my drivers and firmware, but because I can't remain logged in long enough I haven't been able to complete some of the peripherials. I've updated all internal hardware drivers including mobo and gpu. To eliminate the peripherials, I've disconnected everything except the keyboard, mouse, and one dvi conection to a monitor. I've disconnected my hdd's to eliminate those.

I pulled the lower gpu (have 2 ati hd 4850's not in crossfire) and sound card, mainly so I could reset the cmos. Doing this enabled me to successfully login last night after three resets/self reboots. I left it running last night and found it had self rebooted to a black screen again this morning. Tried rebooting and that lead me to a windows update that became a garbled blue colored screen. Manual reboot went a black screen.

All cooling fans in the case, psu, gpu's, and heatsink are operating properly and I verified that case and proc temp are not the issue. Side panel has been removed, so definitely plenty of airflow. Early on in the issue, I used the Windows 7 DVD to perform a memory test which passed.

Alright, so after all that, does anyone have any ideas? Could the psu or mobo be dying? Could really use some help here. Thanks for your time!

__________________________
-Mobo: GIGABYTE GA-X48-DS4, Rev 1.3, Bios F3B
-CPU: Core 2 Quad Q9450 @ 2.66GHz (not oc'd)
-Cooling: Thermalright Ultra-120 Extreme w/ Yate Loon 120mm & AS 5
-Mem: 4x2GB Corsair Dominator DDR2 1066 @ 5-5-5-15/dual channel mode (not oc'd)
-Mem: 4x2GB G Skill DDR2 1066 @ 5-5-5-15/dual channel mode (not oc'd)
-Video: 2x VisionTek ATI Radeon HD 4850 NOT in Crossfire (not oc'd)
-Sound: Creative Sound Blaster X-Fi Titanium Fatal1ty Pro
-HDD0 and HDD1: Seagate Barracuda 7200.11 500GB in Matrix RAID 0, striped 128kb
-HDD2 and HDD3: Seagate Barracuda 7200.11 500GB, storage
-HDD4: Seagate Barracuda 7200.10 750GB, storage
-Case/PSU: Antec 1200/Antec TPQ-850W
-OS: Windows 7 Home Premium SP1 64bit

 

[thestrangebrew1]

Post your specs. Is this a new build? Did this just start happening? Why would you have 2 4850s installed and not xfired?

 

[denis280]

Post your specs. Is this a new build? Did this just start happening? Why would you have 2 4850s installed and not xfired?

all spec.important

 

[kawzman]

Sorry about that guys. My sig was turned off. Specs are below. This was a build from the summer of 2008. This issue just started Sunday. GPUs not in crossfire because I have three monitors. Haven't hardly gamed since upgrading to three monitors, so wasn't a big deal to drop out of crossfire. Thanks!

_____________________
-Mobo: GIGABYTE GA-X48-DS4, Rev 1.3, Bios F3B
-CPU: Core 2 Quad Q9450 @ 2.66GHz (not oc'd)
-Cooling: Thermalright Ultra-120 Extreme w/ Yate Loon 120mm & AS 5
-Mem: 4x2GB Corsair Dominator DDR2 1066 @ 5-5-5-15/dual channel mode (not oc'd)
-Mem: 4x2GB G Skill DDR2 1066 @ 5-5-5-15/dual channel mode (not oc'd)
-Video: 2x VisionTek ATI Radeon HD 4850 NOT in Crossfire (not oc'd)
-Sound: Creative Sound Blaster X-Fi Titanium Fatal1ty Pro
-HDD0 and HDD1: Seagate Barracuda 7200.11 500GB in Matrix RAID 0, striped 128kb
-HDD2 and HDD3: Seagate Barracuda 7200.11 500GB, storage
-HDD4: Seagate Barracuda 7200.10 750GB, storage
-Case/PSU: Antec 1200/Antec TPQ-850W
-OS: Windows 7 Home Premium SP1 64bit

 

[mikeymikec]

Garbled colours might suggest dodgy graphics (though before that I'm thinking PSU or mobo, do you have a spare PSU to try?). Normally I would suggest trying to find a scenario where the computer always fails (prime95, memtest, graphics test, something). Once you have that then you've got something that you can work with a lot easier.

Is the computer more likely to fail from cold (ie. switched off overnight and switched on for the first time in a day) or after it has been running for a while?

Has this instability been happening with increasing frequency?

 

[kawzman]

Thanks for the response mikeymikec, but from both gpu's at the same time? The issue is occurring regardless of gpu. Its occurring in a configuration of both and only one gpu. I've swapped them out while only running one with no change. The issue is occurring before and after the driver update to version 13.1. I stay on top of my drivers and firmware, updating monthly.

Unfortunately I don't have a spare psu. The pc is only off when it's down for troubleshooting. Needless to say this build has been extremely stable over the years, so downtime doesn't happen often. Running for a prolonged period of time is not possible at the moment. On the few instances I have been able to boot into the os (normal or safe mode) its only a matter of time before it freezes or self reboots and the reboot go to a black screen.

There was no frequency change from the first occurrence of this issue. It has been as described in my original post since its first occurrence Sunday.

 

[thestrangebrew1]

You've got a lot of components in your rig, I'd hate to suggest this but disconnect everything, clean everything out, and put together piecemeal. PSU mobo and RAM, power on, power off and then a single device each time till you can replicate the problem. I was initially thinking maybe psu then gpu. Do you have a different gpu you can throw in there? It's highly unlikely both have gone out on you, but I wouldn't rule it out. I've learned to not rule anything out when troubleshooting, especially lately with the probs I thought my server was having.

 

[Steltek]

I presume that it does this with either set of memory listed in your signature installed?

If so, were it my system I'd try replacing the PSU before I did anything else.

 

[mikeymikec]

Thanks for the response mikeymikec, but from both gpu's at the same time? The issue is occurring regardless of gpu. Its occurring in a configuration of both and only one gpu. I've swapped them out while only running one with no change. The issue is occurring before and after the driver update to version 13.1. I stay on top of my drivers and firmware, updating monthly.

It's probably not a GPU problem, but bear in mind, it does happen. I once bought two identical-model PSUs at the same time, and neither worked properly. They weren't a bad brand either, and they were blatantly over-spec'd for the job.

Given that you're saying it doesn't boot properly most of the time, I would probably go with the 'strip it down' approach to begin with. If the machine normally has problems say 2 out of 3 boot attempts, I would probably want 5 consecutive successful boots before adding another component back in.