I am currently experiencing the 544MHz throttle bug on this 4.02 bios. Dissapointing….
A sleep-wake cycle while the dGPU was in use (running glxgears) brought it out of this throttle mode.
I had it too. But the laptop was not on sleep. It switched from Daylight to Night-light (KDE Neon - Plasma) when it happened.
It happened to me while the device was idle but not asleep. I got around it with a reboot.
These BIOS bugs are the sole reason I begrudgingly decided that the FW 16 isn’t quite mature enough to be my only device. As a result, I set my desktop back up, which had been taken down as I was planning to sell it.
is the update process steps guide wrong? kb update says 4.0 bios required before kb update, but they list kb uipdate before bios update…
Just update them both. No reason to over analyze it.
yes, but whats the order LOL
I did BIOS first and am still alive.
1 Like
I did KB first and then BIOS… I’m also still breathing.
I’m pretty sure the order doesn’t matter. It’s not going to prevent your keyboard from typing, it’s just whether or not the lid-close fix is implemented or not. I’ve used the updated keyboard on 3.05 and 4.02.
1 Like
*“*We’ve changed the behavior on Keyboards, Numpads, and Macropads to keep them suspended when the lid is closed. This requires updating both BIOS (4.01 or newer on 7040 Series along with all 300 Series) and keyboard firmware (0.3.1 and newer). You can find the most recent BIOS version for your Laptop here.”
now after bios update, system refusing to boot because out of date kb firmware…and this is why i asked…….
It will flag it hasn’t had the key update showing the image on the monitor with a QR code. Just press ESC to continue boot.
you just need to press power button once and it will let you boot
Oh my mistake, was it the power button? I forgot. I had to do the same thing but it was a while ago. Thanks for the clarification.
Then your bios is different than everyone elses, because ours still let us boot even though it has that message if the keyboard was not updated.
If you are on Linux, the standard update tool will flash the keyboard and stage the bios for next reboot. Not very easy to mess that up.
The manual way to update the keyboard firmware is also quite easy. Works in both linux and windows.
- download the keyboard firmware (being really really carefull you have the right varient for your keyboard)
Keyboard Firmware Update (Framework Laptop 16)
Section: “How to update with other OS (UF2)” - slide the touchpad out so the keyboard powers off
- slide the touchpad back in while pressing a specific key on the keyboard
- a new drive appears, just like you plugging in a usb stick.
- use the touchpad to copy/paste the firmware onto the newly appeared drive. It has the “.uf2” file extension.
- keyboard automatically detects it, reboots itself, update done.
1 Like
Funny enough, I tried that method several times on mine a while back when I needed to update. I tried the .exe first which kept failing because for some reason it couldn’t detect the keyboard. I then moved to the manual way. While I could get the “drive” to appear, it would not accept the update either, and I tried several times with disconnecting/reconnecting/rebooting to no avail. I then resorted back to the .exe route, which for whatever reason magically worked after the struggles mentioned. Honestly it was a pain to do when it shouldn’t have been.
I am also experiencing lower performance. I cannot for certain confirm that it is due to the BIOS (I’ll try downgrading later), but when using the “balanced” amd_pstate power profile (in Linux), and running stress in s-tui, power consumption quickly drops and stays at just 30W when powered with the official power brick. If I unplug it, it jumps up and than stabilizes at 35W!
Previously, with the power brick connected it would stay at >50W for quite a while and then drop gradually as it got hot.
It is not due to thermal throttling, since all cores are far bellow 90°C.
Yes, my performance is also drastically reduced as it caps around 30W instead of 50+. It looks to be thermal related, so I suspect the voltage curve was modified.
I’m on Linux so I can’t comment on how it performs on Windows.