this post was submitted on 24 Aug 2023
23 points (96.0% liked)

Public Blue Screens Of Death

813 readers
1 users here now

Public Blue Screens Of Death

Public displays and digital infrastructure software failing to do their job because of blue screens, crashes or other problems

founded 1 year ago
MODERATORS
 

Image description:

A memory test screen that claims the test passed with no errors, while somehow the display itself has at least three errors on it.

The issue turned out to be bad capacitors.

top 13 comments
sorted by: hot top controversial new old
[–] [email protected] 10 points 1 year ago (1 children)

Nah, it's says there's no0errors at the bottom. It's all good!

[–] [email protected] 5 points 1 year ago (2 children)

I'm surprised that it ran through even one full test with no errors. IME when RAM has any issues, you get a flood of errors during testing.

It can't even display a screen full of text without 3 errors, how did it get past an entire test without catching any?

[–] [email protected] 5 points 1 year ago

A theory I have about why that happened that way and didn't get detected...

The bad caps might have been for the GPU slot power, and the glitch might have actually occurred in the GPU VRAM, not the system RAM.

Sigh, we'll never know for sure...

[–] [email protected] 2 points 1 year ago

I know right? That's the only time I've ever seen such a thing, and I probably would have totally missed it if it wasn't for the no0errors glitch.

A few new capacitors fixed it right up.

Weird huh?

[–] [email protected] 8 points 1 year ago (5 children)
  • Test #0
  • 𝚽Moving
  • no0errors at the bottom

Also I ask:

  • No L3 cache yet there is L2 and L1?
  • DDR669?
[–] [email protected] 9 points 1 year ago* (last edited 1 year ago) (1 children)

~~Damn, I think you spotted two more errors that I hadn't even noticed. I hadn't noticed Test #0 or DDR669.~~

But you did miss one, "Press Ecc to exit" Should be Esc

I found that one extra funny, along with no0errors 😂🤣

[–] [email protected] 4 points 1 year ago* (last edited 1 year ago) (1 children)

I don't believe either of those are errors, unless I missed something on the test count. Counting starts at 0 and you already covered the DDR rate.

Also L3 cache "none" isn't an error. The CPU has no L3 cache.

[–] [email protected] 1 points 1 year ago

Memtest does start at Test #0?

Okay, it's been forever since I've needed to run Memtest, I kinda forget these days.

That Ecc is definitely an error though, should be Esc

[–] [email protected] 5 points 1 year ago (1 children)
[–] [email protected] 3 points 1 year ago

Just... Take my upvote and GTFO... 😂

[–] [email protected] 2 points 1 year ago

The "DDR669" might be DDR2-667 with a slightly inaccurate clock

[–] [email protected] 2 points 1 year ago

This PC is anywhere from 2003-2008. Pretty sure K8 didn't have L3. And DDR2 RAM tends to show as mhz+mhz=DDRmhz.

[–] [email protected] 1 points 1 year ago

I did a little number crunching to see about that DDR669 thing...

If indeed the CPU speed is correct at 3014MHz, then the clock multiplier must be 4.5, yielding a RAM speed of 669.777MHz.

That part just might check out, but I dunno, I took that photo back in like 2011.