Computer freezing randomly for half a second!

Yes Hauk that is the right one and I am thinking maybe go onto update the SATA and chipset stuff.

Now I have been scratching around re the kernel 41 error an it seems to be linked with a temporary loss of power. I know we have done the PSU but I am also thinking maybe the problem lies within the device and after it has been running for a bit because there is circuitry in the PSU that does a self test and if everything as in the voltage is ok then there is a signal sent to pin 8 (grey cable on the 24 pin plug) that is known a the Power_Good signal if that is out of whack for any reason it will shut the machine down. see this
Voltage Rails - Power Supply 101: A Reference Of Specifications go to the Power_Good section and it will tell you what the operating voltage has to be high or low of those volts will cause a shut down. But if you are going to use a meter to measure make sure it is a digital one not the analogue as I think I previously described injects volts and it is typically 9v so in a 5v rail - not good.

It could be something as simple as just one bad capacitor in the unit.

No wait and just try the PSU out with this

Using HW Info PSU

Download HW Info and pick the right bit version - HWiNFO, HWiNFO32/64 -Download -(copy and paste thissite into the search bar) If you do not need the running information in thesmaller right hand panel close it. In the left hand window click on the + forthe individual device groupings and they will open out to the variouscomponents.

Go on opening out further until you get to the device itself. click on it andit will highlight. In the right hand window there will be a very detaileddescription of that device as in brand, speeds model number etc etc These arevery detailed and are just what is needed for searching for drivers etc.

To get the PSU volts you need to select the Sensor buttoncircled in yellow in the pic

See my pic for an example - in this I have shown my power but it is a laptopand a desktop will show more details if you scroll down that listing. (see pic it is an Asus board and yours if different may have a different name for that section)
or this
PSU - Test DC Output Voltage but use a digital meter not the old analogue type it injects typically 9v into a circuit to test it and would do no good in a 3.3v or 5v rail.

There is a part pic of one on the tutorial and they are really very cheap these days I can pick an adequate one up for about $10 - some places even cheaper.

Pic of meter attached

I haven't been able to get one of those tools yet. But something I just realized while playing a game is that my GPU got really hot! It made the monitors go black for a second when it hit 81 degrees C, and the fans were suddenly making loads of noise, probably going full throttle.
Since the monitors went black, it reminded me of that rare crash I've had a few times, as mentioned in #1.
Does the GPU have anything to do with the freezes?

Edit; Under no stress, surfing the web, it is around 40 degrees C.

Edit 2; Thank you!

Hauk it sounds very much like the GPU is not quite right with the temps like that. Now I know the others can advise you better really how to check that card with some stress tests. I know of them but not enough about them.
I might throw this back to more help before going any further.

Now on editing all you need to do is go to the Edit on the return post and it will go back and let you change text etc and then click on save and it will repost the original with your added stuff:)

Hauk, do you use a utility to monitor temps? Obviously you do to know the temp your GPU was during gaming. Which one do you use? Most have the capability to use a custom fan profile to keep the catd from getting that hot. Use a pretty aggressive profile if you do it, I do. Once a card gets very hot, it's difficult to cool it doen, but if you use the fan profile to keep it from getting that hot, it will work much better. I set mine to run 40% until 40C, after that I have the fan go up 1% for every 1C the temp increases. I will have it at 100% around 70-75C. Just a suggestion. does your utility have that capability? If not, try MSI Afterburner, MSI Afterburner