Bizare Screen Problem with HTPC

[th3pun1sh3r-1313573]

I have a HTPC that started developing issues about a week ago. The computer (which is always on) was found to be off one afternoon. (It was a warm day so I am wondering if it didn't over heat.) I turned it back on and Windows 8 gave some vague error that there was a config problem. It became unstable to the point that it was unusable after that. I was able to briefly look at the event logs and there were no specific errors that would explain the crash. The computer wouldn't stay up long enough to investigate further.

Yesterday I went to reinstall Windows 7 on it and the installer ran and eventually the screen started producing the below image, then it went purple and the computer eventually turned off. Over the past week when I turned it on it did produce the same image. Anyone have any ideas what this is a result of?

I'm leaning toward this having been caused by the processor overheating. I have the processor connected to a water cooler and the radiator is in front of the power supply. So the power supply fan pulls air from the case across the radiator. As long as the ambient temperature in the room is ok this has been a very stable configuration.


i3-2125 processor
ZS Series 550 Watt ATX Power Supply
Z68 Pro3-M Socket 1155 Z68 mATX Intel Motherboard
Ballistix Sport Series DDR3-1333 (PC3-10666) CL9 (there are two sets of memory for a total of 12 GB's)
OCZ Agility 3 AGT3-25SAT3-60G

Screen

 

[ronintexas]

There are two things I would investigate. First, try removing one set of RAM and see if the problem continues - if it does, swap the RAM.

Second - can you monitor the temperatures in the BIOS to see if it is overheating?

Last - what GPU are you using? Do you have one you can swap with it?

 

[th3pun1sh3r-1313573]

There are two things I would investigate. First, try removing one set of RAM and see if the problem continues - if it does, swap the RAM.

Second - can you monitor the temperatures in the BIOS to see if it is overheating?

Last - what GPU are you using? Do you have one you can swap with it?

I will try the RAM suggestion and post the results.

After I switched to the water cooler I monitored the temperature quite a bit to make certain the setup was working. All the temperatures were good, it kept the CPU well below Intel's specification (I forget what it is now). As long as the ambient temperature in the room is below 80 the setup works (I live in Minnesota and we run the AC as soon as it gets to 75). Last week we had a warm day but it was only in the 70's.

I am using the integrated graphics on the processor and I unfortunately do not have another GPU.

 

[th3pun1sh3r]

I tried testing the two sets of memory and got the same behavior. The behavior has changed since I first posted my problem. I tried a couple of different boot disks for testing hardware and they all eventually end in the PC powering off. I just booted into the BIOS to watch the power and CPU temp and the computer just shut off. Then it started into this power cycle where it turned on, then off, then on, then off. The computer never got to a point where anything showed on the screen. The entire cycle would take maybe five seconds. Any thoughts if this is the motherboard or the power supply? The voltage of the CPU looked good in the BIOS before the computer went dead.

 

[ronintexas]

There may be an issue with the motherboard or the power supply. Since the original issue was with the graphics, I would start with the motherboard.