Fake CPU Failure warning

Event ID 41 is not a Windows error. It is a hardware error which can be caused by the system detecting anomalys such as different stepping levels on cores of the same CPU. It's potentially serious, and you definitely want to get it handled.


If you are Overclocking: Reset the CMOS to the defaults, by removing the battery (while powered off, OFC), or by using whatever means of resetting your mobo maker provides.

Ensure you are running the latest bios for your system.

And try to check/repair any Windows System files which may have been (or may be) damaged. To to that, open an elevated command prompt by typing CMD in the search box, right clicking the return and selecting "Run As Administrator".

Then run the following command: sfc /scannow



IIRC, it's also possible for these kinds of errors to be cause by some perticularly nasty virus.

Hi Scotteq, thanks for the reply. I am not OC at all and have even set my BIOS to "Load Fail-Safe Defaults"
Now I've done this
1. Right-click My Computer, and then click Properties.
2. Click Advanced system settings.
3. Click the Advanced tab.
4. In the Startup and Recovery section, click Settings.
5. Click to clear the Automatically restart check box.
to maybe see if there is any change.

I am doind the system files scan at the moment ....let's see ...I will give an update asap.

I may turn off a setting in the BIOS Labelled Thermal Temperature Control to see if that helps ...I think that lowers the voltage etc for core when it gets too hot but not sure if that would report anything to the OS.

Frank

Also, check in the bios, on my board, there is an option about auto temp fan control or something. Essentially as processor gets toward a lighter load, the fan slows down. I turned that setting off which makes the fan run full speed all the time. If the two things don't work, for intel, they have speedstep, I think AMD's function is called cool n quiet, essentially it downclocks the chip when it's idle. You may turn that setting off and see if the errors go away.

Lastly, what power supply are you running? If you have a cheaper, off brand power supply, it could be to that maybe there is some kind of voltage fluctuation going on to the chip, and causing the errors.

Thanks for the input Scott. I have had no re-occurance since the 11th. The last thing I did was to remove the EasyTune6 (Gigabyte utility)..not sure how good the software of those Gigabyte apps is ....
I have latest BIOS (there is BETA version ....but that is for PhenomII 6 core support mainly)
1.Beta BIOS
2.Update CPU AGESA 3.7.0.1
3.Support GIGABYTE On/Off Charge function

What is on/off charge function (3) ?????
Not too sure how AGESA is used at Bootstrap time.

I have changed PSU from a 750 to an 850 W (kept the other spare) a while back and that did not stop the re-occurance ...also changed the CPU Fan as well ...
I suspect this is a software issue with one of the apps I was running.


Thanks for the Technet entry, that's pretty generic but can be useful if there any data in the eventlog but in mycase it is after the event so not much use.

Well let's hope it keeps stable....

Frank

Depending on the brand & model the 750W PSU could have been better than the 850W one.
Gigabyte's on/off charge basically gives the USB ports 3x the power specified in the USB Specification, it makes it so if you charge with the USB port (phone, iPod, etc) it charges quicker.
http://www.gigabyte.com.tw/microsite/185/on-off-charge....

I would do a virus scan, Norton and Kaspersky are my favorites, I'm pretty sure Norton gives a free scan.

I'm going to flash the bios with the latest version even though it is a BETA as they have mentioned they have improved the CPU microcode. (reply from Gigabyte).
Been checking the voltages and they look pretty stable even when I ran some soak tests with Orthos and Prime95 .....The CPU core seems to get to 58C in about a minute ...at that point I usually kill the test....is that kinda normal for a 955BE ???

Let's see what happens with BIOS Flash.

Frank

Update:
I think I kind of understand what is happening now.
Yesterday I installed a brand new 120mm Fan (2 wire... so that it is permantently on on SYSFAN1) .....Previously I had an 80mm Fan (3 wire at approx 3000rpm). Within an hour I got my my 1st failure and then another about 2 hours later. All my previous failures also had been on SYSFAN1 with a 120mm FAN (3 wire) which I now put on SYSFAN2.
It seems that if you have a 120mm FAN (these do not reach more than 1500 rpm normally) that the io port reading these rpm signals maybe missing some of them and deciding the FAN has failed. Or maybe it expects an 3000 rpm fan there going a lot faster than a 120mm would.

Anyone has any info on this function.

I have now put the 80mm Fan back on SYSFAN1 and will monitor.

Best answer selected by fxv300.

No more failures since I replaced the 120mm fan with an 80mm fan at SYSFAN1. As the problem ahd

originally started when I had replaced the 80mm with an 120mm I suspect that the OS or the BIOS

sometimes see the low RPM as a failure.

Thanks for all the replies.