[RESPONDED] AMD Framework minimum brightness too high (now with measurements)

Framework Laptop 13
22

Just to set expectations for other operating systems, my understanding is the 6.14 fix is a workaround and the underlying issue is a misconfigured minimum value in the display controller board firmware (panel?). The kernel fix overrides the firmware value. Is this correct?

23

Pretty much exactly that.

24

How to measure the improvement without professional instruments? Put a solar panel in front of the screen?

25

With your bare eyes. It is very noticable.
Or make a picture with your phone with different kernels.

26

What about those who have to use Windows for various purposes? Firmware update coming to address this?

27

No idea. I’m a Linux user/dev.

28

The latest statement from Framework is in this very thread. There is no software workaround for Windows since source is not open. You are totally reliant on a firmware fix from the hardware vendor.

29

Any updates on this? It’s been almost a year. This bug makes Windows on my Framework 13 AMD almost unusable at night

Wondering if there’s been at least some sort of “debug/insider patch” we can apply? I’m willing to sacrifice VariBright features

30

The latest statement from Framework is in this very thread. There is no software workaround for Windows since source is not open. You are totally reliant on a firmware fix from the hardware vendor.

[RESPONDED] AMD Framework minimum brightness too high (now with measurements) Framework Laptop 13

31

Oh got it, do you work for Framework? Any estimates on a firmware fix from the hardware vendor? And is the issue here the display, or the video driver?

Also, was curious about software workaround because I’ve seen people hack AMD drivers and software in the past by modifying the binary hex code. Should be possible, but curious if anyone has done that.

32

If I preorder one of the new 13s is it likely to have this problem still?

But, since it’s a known problem, I shouldn’t really worry anyhow because it’s fixable in firmware? Or is this firmware not conveniently flashable from software and I have to take the panel out and do some JTAG sort of thing (or just send it in I guess)?

Or, since I’ll only ever be using Linux and don’t mind using a newer kernel, the workaround is fine? Or is the workaround a compromise that’d impact quality compared to the firmware fix eg. I’d have less dynamic range at low brightness or something?

33

The kernel quirk is working fine. There is nothing more the firmware could be doing.