Still good after a week

Got a new motherboard and when I went to swap it out I noticed I had a standoff in the wrong spot 🤦🏽‍♂️

Anyway I tested my RMA'd RAM and my system's all in order again. Though I'm gonna keep MemTesting for a while to catch it early if there's still something wrong

They should probably respell Arkansas but it's kinda funny how much it trips up everyone so I get why they don't

Only if they're #crazy

But anyway thank you so much for all your help! I appreciate the advice and guidance, it definitely steered me in good directions and made some of the decisions easier, ty

Yeah considering that the last 2 kits tested fine initially I'm willing to believe they've never been to blame. I'm still gonna RMA them again, but while that goes thru I guess I'll shop around and see what I can get in terms of a motherboard replacement. Not sure I wanna risk staying w this one now

And I tested the PSU today! Everything was within the acceptable tolerances and the readings were very consistent, didn't notice anything fluctuating wildly. So I feel I can rule out the PSU at this point. Question is whether to trust that the BIOS was the problem or not before I risk my next kit

TuxTimings screenshot

Wasn't able to update the BIOS through the physical button, but doing it from within the bios itself worked fine. Memtest86 still gives errors. And this is TuxTimings with the new bios, looks very similar. Seems GDM is now enabled and command rate changed from 2T to 1T?

My boot was already broken anyway, already planned to fresh install once this is in order. I've been using a USB live environment to rescue files + do this tuxtimings thing.

Yeah I'll give the bios an update then. But aside from that I guess the current plan is just test the PSU at some point tmrw

*LR glasses

And finally the BIOS *is* out of date by like 2 versions, though it was the most recent version for the day I built the PC. And the listed patch notes don't seem to be relevant to this issue as far as I can tell.

Screenshot of TuxTimings. In the RAM tab under Voltages, the non-empty items are: VSOC 0.9750 V CLD0 VDDP 0.9002 V VDDG CCD 0.9002 V VDDG IOD 0.9002 V VCORE 1.4500 V PPT 15.8 W

And in terms of the live voltage view I did find github.com/Death4two/Tu... which, assuming I've installed everything correctly, does give me some voltages, but seemingly not all of them. And I've no clue what most of those acronyms mean yet but here they are

MemTest86 failing with 645 errors

Ok so I ran the test with the lower frequency and it still failed, but the errors did drop significantly from 1716 down to 645.

My current thought has been I'll test the PSU (apparently there's dedicated little tester devices, but you can also use a multimeter?) and if it looks fine then it's most likely the motherboard (esp. if CPU can't cause it as you say).

Ah I don't have Windows installed, and l don't have a spare SSD for it. I'll have to look into a Linux alternative.

At least from looking at the BIOS I'd checked the voltage was reported at 1.20 V which is consistent with the specs for my RAM kit when not running XMP (which I have never enabled)

But from the stuff I've read it seems that the PSU, motherboard, or bent CPU pins could all be causes of dead RAM so I need to figure out which I might need to replace

This kit only lasted me 2 months before this filesystem failure (but the test results show much less errors than before). I spent a year with the previous kit before I realized it had gone bad. Both kits tested perfectly fine when I initially received them.

I think I'd updated the BIOS? Not sure..

Oh and this is what MemTest86 does if you cancel it early

This is how my system boots rn

Anyway, for now I suppose I'll see if I can still recover anything of my files, the corruption shouldn't be that big. Most important stuff is backed up as of a few months ago at least.

At this point I don't know whether I should just suck it up and go buy a different set of RAM (💸) instead of try for a 3rd RMA replacement. Or is it something else that's the problem, like is it my motherboard that is killing my RAM sticks? Because what are the odds at this point?

So this time it's even more annoying because I think it corrupted my filesystem. I was always lucky my fs didn't corrupt before when I had regular crashes on bad RAM for a good while. Now with a relatively fresh set it corrupts immediately, before I even begin to suspect the RAM could be bad again.

god fucking damn it

A glass of lemonade

It's made, I guess 👍

I might have to get a bit creative

A bowl full of lemons, with more lemons littered around it

Thanks, life

Trans people: I'm trans

Society: ok

Ludwig in New Super Mario Bros U:

Did anyone else read Quandale Dingle more like "kwandalay" instead of "kwondail"?