I've gotten what I believe to be a workaround. If you disable EIST and C-State's in the BIOS the system seems to become stable. I can't say that this is actually a fix or a proper workarond. All I can say is that the PC's haven't so much as twitched since I changed those settings.
If anyone else runs into a similar issue can you try this and let me know if you get the same results as me.
I am having exactly the same issue.
I first noticed this issue when I used the default Intel driver install from the CD that came with the board. The problem was severe, the entire screen would flicker to black and come back. After a few times doing this, the system would reboot within 30 seconds of coming up. I know the memory is good, as I had used it on a prior system for some time and know it was stable (I upgraded that system to 8gb and took the 4gb and stuck it in this system immediately prior and that system, a Intel PW55B was perfectly fine)
I have installed the latest BIOS and video drivers with and it decreased the frequency of the reboots, from immediate every time within 30 seconds, to randomly every few minutes to an hour where the screen has a dark diagonal line about 1" wide goes across the whole screen left to right at a slight 20 degree angle.
This problem does not occur at all in SAFE mode at all, which tends to lead me to believe it is software related. Not one flicker or reboot while I am using the problem computer to type this.
Turning off the Intel Speedstep and C5 power state seems to have fixed the issue, though we give up the Speedstep capability of the processor for stability!
I have a similar problem and I have not been able to solve this issue.
About two years ago our primary system builds were based off of the DQ67SW and the DQ67OW desktop boards. About a year or so after they were built we started seeing a few of these system having an issue were they would simply shut off during usage. No warning no blue screen just shut off.
Replacing the system board has been the only way we have been able to solve this issue.
Now, I have an epidemic of these systems doing the same thing. All of these systems were built within a 15 or so month period and all of them are the DQ67SW or DQ67OW system boards. It has been a hit or misses
whether or not the boards returned from Intel would suffer this same issue. I literally have hundreds of these systems now showing up with this same issue. Friday of last week I had a single business customer bring me 4 systems all with the same issue and he told me that he had more. I had two replacement boards in
our inventory that already have been RMA’ed for this same issue. Both still suffered a random power failure issue. We have tested all of the other components in the system and still the system board is the faulting part. A new behavior issue was discovered on Monday shed some light on the subject. I replaced the system board and ran a burn in test for 6 hours on one of the above mentioned computers. The system was then returned to the customer. Once the customer put the system back in it failed to stay up long enough to log in.
I was confused as I just ran the system for 6 hours. After a little experimenting I discovered that the
colder the room is the more likely or more often it power off. I also discovered that if I dropped the temperature of the board down to 45 degrees F it would fail to power on at all. Being somewhat of a geek I decided to do some more experiments. I took every DQ67OW and DQ67SW system I could get a hold of and performed this test in a controlled environment.
With nine systems in total I was able with the exception of one to get them to fail to power on by dropping the temp of the boards between 40 and 45 degrees. This is after I tested each one before chilling them. Every
system powered on successfully just before I chilled the system boards. Just as a matter of clarification I only chilled the board and no other component. Using a laser thermometer I was able to determine that most of these systems would start to power on at between 55 and 60 degrees. How I discovered this anomaly is that one of the system was delivered VIA motorcycle in the cold weather. I also discovered that this particular system was placed in a part of the building that was without heat. All nine systems experienced more frequent
power failure if the ambient room was around 60 degrees. There is about 8 month’s period between the build time on all nine systems.
I also tested the theory that in a hotter environment if it would run more stably. In almost all nine cases the system experienced less power failure if the room temp was higher the 75 or 80. I have spent countless
hours trying new processors, power supplies, and memory. I have also eliminated
the chassis as being the culprit by building the system out on the counter without a chassis. We probably sold 1200 of these systems over a 15 or so month period. Not every system is doing this but I have reports from our techs and our customers starting to pile up quick. we are getting more DQ67OW boards
doing this than the SW.
I have escalated this with Intel but no results yet, if
somebody has a workaround for this issue please let me know.
Manfred1, I am sorry to hear you are having problems with all these motherboards. Would you please send me a private message with the Intel service ticket for the escalation?
In addition to that Manfred1, please let me know the following information of at least 2 systems:
- Motherboard AA#
- Current BIOS version: