Firefox dev says Intel Raptor Lake crashes....

erek

erek

8-D
2FA
Joined
Dec 19, 2005
Messages
18,491
... are increasing with rising temperatures in record European heat wave — Mozilla staff's tracking overwhelmed

“These CPU instability issues have caused Intel to lose ground to AMD in the desktop PC market share, and its Arrow Lake chips performed poorly in sales, exacerbating the company’s financial problems. In comparison, AMD has been releasing excellent gaming processors, especially the AMD Ryzen 7 9800X3D chip, which eclipses even Intel’s latest offerings in several gaming benchmarks.”

Source: https://www.tomshardware.com/pc-com...ntel-crash-reports-team-disables-the-function
 
Last edited by a moderator:
It's weird that a difference of 10°C ambient temperatures from ~25-30°C up to ~35-40°C would cause CPU's to die.

These are undoubtedly just CPU's that were dying already, either without updated BIOS/microcode or that were permanently damaged to the point where failure was inevitable before the new BIOS/microcode was applied.
 
Last edited:
Zarathustra[H] said:
It's weird that a difference of 10°C ambient temperatures from ~25-30°C up to 35-40°C would cause CPU's to die.

These are undoubtedly just CPU's that were dying already, either without updated BIOS/microcode or that were permanently damaged to the point where failure was inevitable before the new BIOS/microcode was applied.
Click to expand...
Agreed. I doubt these claims too.
 
This seems odd to me, because areas in the US see those temperatures on a regular yearly basis, yet this doesn't appear to be an issue there?
So why are those ambient temperatures a problem in Europe but not the US?

Here's a link to the Bugzilla page for the issue:
https://bugzilla.mozilla.org/show_bug.cgi?id=1950764

And here are some people discussing it:
https://mas.to/@gabrielesvelto/114814117276254003

And here's the important bit of the analysis:
" The real issue I’m seeing is with gaming motherboards that set PL1 and PL2 to 4095 and folks are using regular cooling in hot climates, so the CPU is running too much power and doing so at too hot a temperature. "
 
Last edited:
Hey, I'm planning to upgrade after my 5800x died. I heard the i5s were less susceptible, should I still be looking at a 14600k?
 
Haswellbeast said:
Hey, I'm planning to upgrade after my 5800x died. I heard the i5s were less susceptible, should I still be looking at a 14600k?
Click to expand...
The microcode updates have solved this, so a new cpu will be fine. Also, Intel gave an extra 2 years on the warranty for these, so it's five years total (double check that it's still valid).
 
GoldenTiger said:
The microcode updates have solved this, so a new cpu will be fine. Also, Intel gave an extra 2 years on the warranty for these, so it's five years total (double check that it's still valid).
Click to expand...
Ah, okay, I also heard that there's something like "VID limit" that you can set in order to stop the CPU from asking for crazy voltages like they were with the 6ghz CPUs in particular
 
Just need to enable SVID behavior in the Bios from Auto to Typical usage unless your on a laptop. I had high temps since 2023 untill I upgraded my cooler to the Noctua G2. When Dragon Age came out last year I was getting 100c on a few cores my Ram would flash red so decided my older cooler wasn't good enough.
 
Last edited:
Lakados said:
This seems odd to me, because areas in the US see those temperatures on a regular yearly basis, yet this doesn't appear to be an issue there?
So why are those ambient temperatures a problem in Europe but not the US?

Here's a link to the Bugzilla page for the issue:
https://bugzilla.mozilla.org/show_bug.cgi?id=1950764

And here are some people discussing it:
https://mas.to/@gabrielesvelto/114814117276254003

And here's the important bit of the analysis:
" The real issue I’m seeing is with gaming motherboards that set PL1 and PL2 to 4095 and folks are using regular cooling in hot climates, so the CPU is running too much power and doing so at too hot a temperature. "
Click to expand...

The reason is that Europe has a disproportionately high percentage of the population without AC, unlike the US.

1752873049073.png
 
Zarathustra[H] said:
It's weird that a difference of 10°C ambient temperatures from ~25-30°C up to ~35-40°C would cause CPU's to die.

These are undoubtedly just CPU's that were dying already, either without updated BIOS/microcode or that were permanently damaged to the point where failure was inevitable before the new BIOS/microcode was applied.
Click to expand...

If your system is broiling with the CPU above 95C on a comfortable day when ambient is 10-15C higher you're going to be at red line and risking trouble.

On my older systems I'd do post-build stress testing around 85F (vs 65-75F normal room temperature) because a day or two in a hotbox would consistently trigger errors that would otherwise lead to once a month or two system crashes.

With Intel's last few generations being factory OCed to just below the redline I'm not surprised heat waves are triggering crashes in systems that are already struggling.
 
Isn't this the first time that the 13/14th gen problems are associated with temperature?
 
uOpt said:
Isn't this the first time that the 13/14th gen problems are associated with temperature?
Click to expand...
Maybe, but a chip that's already predisposed to degradation, will generally experience worse degradation when heat is compounding the issue.
 
  • Like
Reactions: uOpt
like this
PeaKr said:
Why, what'd you do to it?
Click to expand...
I’ve had more than one B350 and maybe a B450 board pop on me in the past 6 months taking the CPU with it. Cant speak towards this person but there are plenty of AM4 systems out there old enough to encounter a critical failure.

Seriously considering replacing them with some 8700G’s, give them 48Gb in ram so I can safely dedicate 16gb to the GPU….
Could make some decent little multi media rigs.
 
PeaKr said:
Why, what'd you do to it?
Click to expand...
I left it sitting for a week... And then it started bootlooping, and then the board posted just fine once I put a 1400 in it. The weirdest failure I've ever had. I also had a b550 Asus prime board that died when I tested it on a box right after I received it. I'm pretty sure it was my thermaltake smart power supply, as every capacitor after the EPS power was shorted. Still, it worked, I applied a pbo undervolt and it instantly shut off, never to turn back on

I've had terrible AM4 luck
 
Last edited:
Haswellbeast said:
I left it sitting for a week... And then it started bootlooping, and then the board posted just fine once I put a 1400 in it.
Click to expand...
My boards started having power delivery issues. I didn’t have any lower power CPU’s to try but I do wonder if I’d placed some 65w or lower processors in them if it would have booted.
 
You must log in or register to reply here.
Back
Top