I also came here to say thank you for the very pleasant update process via Fedora everything worked! So far no problems. I did not subscribe to the emails, but Fedora just proposed to run the update via the default process. 
After installing Windows on a spare NVME drive for the last bios update this was a great experience. Thanks again
Will try to run the EFI version of the update to also update the Intel CSME. Will update this post when done
Can someone confirm we have the stable version ?
Also the docs warn that the LVFS update does not update CSME, maybe itās be nice if the docs mentionned what it is and why should I care about it ? Am I to run into issues ?
See the edited opening posting:
So yes, the 3.17 BETA previously announced in this thread has been declared the 3.17 stable.
So today I got a notification about Framework FW update.
I then went to check the actual framework KB here to confirm. Indeed, stable 3.17 update has been released. While I was loosely following the FW update beta, I decided to stay on the stable 3.05 which shipped with the laptop. I got the update on a USB drive (wanted to update everything, including retimers), everything went smoothly.
1 thing to note, damn are the fans kicking in too often and too fast imo. Whenever I start a new program, the fans kick in like crazy and chill out in <1 second. Iāve kept reading about this 400 Mhz bug when the CPU sometimes gets locked to the lowest possible frequency and makes the laptop unusable. So far I have not encountered this issue with 3.17 update.
Running Pop!_OS 22.04 LTS with 12th gen i7 1280P.
This is a known problem confirmed by Framework, see above in this thread. But they still went ahead and released the update. (Iām not convinced that this was the best thing to do.)
Indeed, appreciate Framework devs fixing an issue with throttling on the machines to unlock performance, but the new behavior is really irritating.
I donāt know if this is potentially a bad idea, but this is where I ended up so I didnāt go crazy with the annoying fan curve or have to replace the machine. Iām monitoring my CPU frequency and I see it ramping up and down appropriately with system load. Iām also monitoring PECI and I see it spike quickly when the CPU starts to get a load, but CPU temps remain reasonable and the fan varies appropriately with system load instead of going to max RPM at the first sign of load.
> ectool thermalget
sensor warn high halt fan_off fan_max name
0 0 361 371 324 355 F75303_Local
1 0 361 371 324 342 F75303_CPU
2 0 360 370 313 342 F75303_DDR
3 0 323 333 313 323 Battery
4 430 440 450 320 410 PECI
5 0 361 371 324 342 F75397_VCCGT
(all temps in degrees Kelvin)
I am on 3.04, so I am trying to install 3.06, the oldest version I can find. That installation fails every time with this error:
Any advice?
Hi!
3.06 never made it to final. It was replaced by 3.08 as first public update for both, machines shipping with 3.04 and 3.05.
You should try 3.08 instead of 3.06 IMHO
Hi @Liam_Fruzyna
Some people had the same problem with 3.08:
And one user posted this solution:
Perhaps this helps you?
@Quin_Chou
Just out of curiosity: why this big version jump from 3.08 to 3.17?
Marketing considerations? ( Apple sends its regards 
)
Probably the same reason it happened for 11th gen. Somebody mistook number formats (its interpreted as hex number in some places). And after 3.10, hex and decimal format are are incompatible (3.A vs 3.11), so they skipped all the versions for which this would not be possible.
I.e. only using numbers for which it does not matter if its interpreted as hex or decimal numbers for checks which is newer.
And 3.10 was probably some internal version that never made it to public betaā¦
Edit> found the original post
Why 3.10 to 3.17?
In the 3.10 release, some of the SMBIOS/ESRT metadata was encoded using 0x310, which was converted from hex to decimal, and became 3.16. So we had to skip a few versions ahead so that both the hex and decimal versions are both incremented correctly.
Mhh. That is more embarrassing for Framework then I would have thought, that it was not a preexisting mistake on 11th gen, but somebody encoding the version 3.10 wrong at the creation of 3.10. Looks like they litterally redid a mistake they had already done and should have learned from 2 years priorā¦
after removing the CMOS battery, blowing compressed air on the fans and reinstalling 3.17, it seems Iām not hitting the 400MHz issue.
Yesterday I had a full day of meetings while running hyprland and meet videocalls, that combination was forbidden before.
Today Iāll have some more meetings where Iāll be sharing screen, thatās the ultimate test.
I first tried to update my laptop on battery (using the EFI method) and I donāt think it got very far before turning off. I think just the CSME and microcode were updated.
I plugged it into AC and tried again, and I think everything got updated.
It might have been helpful to have a warning/error message in the EFI script to say that it needs to be plugged in for all updates to proceed.
Just upgraded my i5-1240p laptop from BIOS 3.08 to 3.17 using the EFI upgrade script, without issue. No issues with ports or 400MHz CPU throttling. Thank you all.
Weāve received positive feedback after reverting the fan settings on 13th Gen platforms, which has resolved the 400MHz issue. A similar fix for 12th Gen platforms is in progress and scheduled for release by the end of June.
Resolved? Or hidden?
I did the update on my i5-1240P (dual boot Win11 and Debian Bookworm) yesterday. 3.08 - 3.17, using the Windows installer. No problems.
Only one thing is really bad now: the fans are really, really nervous. They spin up for short peaks, become very loud and then return to normal. This is very negativ compared to 3.08 and should definitly be adressed as it interrupts normal work!
No 400Mhz-problem until now.
One other question: how can I find out which version my retimers currently have?
