Understanding Errors

[Mokki]

Hey,

I've been troubleshooting my PC for some time now. It crash at random and doesn't turn on again, or it sometimes does come back on after a few days. I suspected the PSU to be faulty so got a new one and the issue still persist. Decided to remove all but 1 stick of ram and it turned on again, swaped stick for another one and it didn't turn on. So I thought 1 ram stick was faulty, ofc I managed to mix them up, so I had to do it again to find the faulty stick but now they all work. So I installed Memtest, ran it 4 times (4times on each stick (4 sticks)). The error seems to be
identical to one another, so I'm pretty confused atm.

Could anyone with brain look at these 4 logs and tell me what's up?
(Could only upload 3, but you get the point, 4th log is the same)

[David (PassMark)]

Was this machine ever stable?
Did you install any new hardware components recently?

Looks more like a BIOS bug where the BIOS is claiming a particular memory address (0x13FFFFFFC) is free to be used, but in fact it isn't free at all, but used by some other piece of hardware.

[Mokki]

The PC is closing in on 6 years now, it has had its ups and downs. First crash was when it was 2 years old, then the PSU died. Then after another 2 years the GPU died. Now I thought it was the PSU again but turns out it's something else. The machine is located in a machine shop, so there's quite a bit of oil mist in the air which gets through the fans, this is how the GPU died at least. Filled with oil dust which I guess shorted the GPU. CPU cooler (intel stock) is also filled with the same residue, but has been cleaned several times to allow propper cooling.

But to answer your question, yes, the machine has been running flawless if you don't consider the above crashes. Only now very recently has it become a nightmare to work with as it turns off and takes days with trying to get it up and running again. As for new HW there is no new components.

The BIOS has been the same over the past 2 years I believe. I could try and update it and run a test again? But would you agree that it looks strange with 4 different sticks reporting same exact errors?

[Mokki]

Unable to edit previous post as it's yet to be approved, so to add from previous response I just updated the BIOS.

And there's a significant change.
I tested will all 4 sticks in at same time this time, and this is the result. I still wonder what the outcome means, any translation for us less gifted?

[David (PassMark)]

So all that is a bit strange. As if there was a BIOS bug, it would have been in the machine and causing problems from day 1. Maybe it was a combination of changes (e.g. a Windows update, plus an old BIOS bug)

After the BIOS update is the machine now stable (or more stable) in Windows?

New test result looks a lot more like a bad RAM stick. There is just one byte with 1 bit in error, which almost always means bad RAM. Might be tempting to now go back to testing 1 stick at a time.

[Mokki]

I'll run the test again with 1 stick at a time when I get back to work on Monday. I didn't get to use the machine much after the BIOS update, but it did crash ones during that time so would assume it's still unstable.

[Mokki]

I have now tested all sticks individually and got no errors. Then tested with two sticks and got no errors. Then back to four sticks and got 1 error.
Then I tested all four individually again, but only ran Test 8 (which was the where I got error) and ran 12 passes. The fourth stick I put in finally revealed an
error at pass 7. Can I safely assume this is a faulty stick? Either way, I'm now just running 2 sticks to see if the PC crashes at all.

[David (PassMark)]

If the machine was stable with the 2 sticks, then yes, I would replace the 4th stick as assume it was indeed faulty.

[Mokki]

PC seems stable, I just couldn't figure out if it was the ram, or one of the last two ram slots on the MB that was faulty. Thnx for answers David!
Good idea to update bios though