08:05 <marcan> leio: I've seen some LG monitors report 4K30 EDIDs over HDMI even though they accept 4K60. if yours is like that, nothing we can do about that, I'd consider that a monitor firmware bug.

08:06 <marcan> we're effectively using the macOS display stack with DCP, so if macOS does up to 30Hz, then Asahi will do the same, since it's the same firmware enumerating the modes

08:07 <sven> it looks like we also can’t easily fix monitors that negotiate 4 lane DP when they could do 2 lane + usb 3 instead :(

08:14 <marcan> usb-c firmware quality seems to be particularly bad :/

08:15 <sven> there’s a single flag that stored on flash which disables most customization we could do afaict

08:15 <sven> *that‘s

08:16 <sven> I wonder if the tipd swd is connected to some soc gpio though. but then again, the newer typec controllers are probably more locked down if they are similar to the new iPhone ones

08:16 <sven> apparently there’s „port dfu“ now which might require personalized firmware updates :/

08:20 <PaulFertser> Do modern displays generate EDID on the fly somehow or is still a dedicated EEPROM IC inside which can be ~easily reflashed? Or probably there's a market for a small board which would pass through the video lines but allow to present a custom tweakable EDID?

08:22 <sven> i have a cheap hdmi splitter that allows to change the edid

08:22 <janneg> marcan: I have the edid, it reports a 3840x2160 59.996625 Hz mode

08:22 <sven> pretty sure i could easily hack that to allow fully custom edids. and also pretty sure something like that exists on aliexpress ;)

08:24 <sven> for display port the edid is sent over the aux channel which is slow but still a differential pair iirc

08:25 <janneg> dcp supports uploading edid so we should be able to provide a way for custom edids without anything extra

08:25 <sven> ah, nice

08:25 <PaulFertser> Very nice indeed.

09:12 <eiln> janneg: do you might know where to find vp9 test vectors / gnarly samples

09:12 <eiln> or do I have to tinker with the encoder

09:12 <eiln> mpegs have the itu-t draft conformance samples but vp9 I can't seem to find any

09:28 <janneg> eiln: at the bottom of https://www.webmproject.org/vp9/levels/ are vp9 performance test streams. not as nice as the h264/h265 conformance suite

09:31 <eiln> ah, thank you

12:00 <leio> I'm happy to worry about the 30Hz pick over 59.99Hz after there's working 30Hz or 60Hz on the 6K monitor that exposes a 60Hz exact 4K EDID ;) But yeah, macOS only exposes 24, 25 and 30Hz for it, no 59.997, the dmesg doesn't seem to have listed what the firmware enumerates to know if we'll be able to do better out of the box. Vaguely recall it had 60Hz on a mac mini, but it was last intel era one

12:02 <leio> I'm not going to be using the 59.997Hz monitor until there's dual-monitor support (dp-alt) for it, but happy to test things to make it better for others; I wonder if I could test it in macOS over displayport somehow, but I don't have any such cable if it's supposed to be possible to drive it over thunderbolt

14:24 <millenialhacker> Hello!! long time without chatting over here...

14:24 <millenialhacker> congrats to eiln and company for the work on the webcam. I was so excited today when I saw the camera finally working after @world upgrade.

14:32 <eiln> millenialhacker: happy to hear :)

14:32 <eiln> I did initially base the tracer off of yours, though it got a quite the facelift since (lina worked on it too)

14:32 <eiln> all changes are upstreamed in m1n1 fwiw

14:33 <millenialhacker> thanks, I did a tiny thing compared to the work both of you did. :)

14:35 <millenialhacker> Is there any peripheral in M1 family not taken / being worked yet?

14:36 <millenialhacker> I checked the feature support wiki but it seems almost everything is already in progress

14:45 <eiln> millenialhacker: the last big one is prolly touch id/sep. there's also some misc bits e.g. jpeg (ask r), apple light sensor (ask sasha)

14:47 <millenialhacker> Thanks eiln. Who was working on SEP stuff?

15:09 <janneg> millenialhacker: amarioguy

15:10 <millenialhacker> amarioguy do you have any branch docs I could give a look?

15:10 <millenialhacker> janneg thanks! :)

15:21 <marcan> microphone also needs someone to pick it up I think

15:21 <marcan> povik had a version bypassing AOP, but I'm really not comfortable with that after where we ended up with ANE

15:21 <millenialhacker> Mic + Cam would be a nice atractor for folks required a decent ViCo

15:22 <marcan> (and those branches are now very outdated anyway)

15:22 <marcan> I'd like to see a version that does it the macos/AOP way

15:22 <marcan> since we need AOP for the ALS anyway and possibly other stuff

15:31 <millenialhacker> marcan: ALS stands for?

15:31 <millenialhacker> marcan: Do you want some kind of AOP signalling to turn on/off mic on-demand?

15:31 <j`ey> ambient light sensor

16:00 <marcan> AOP is supposed to be involved in the mic config/signal path as far as I know

16:00 <marcan> basically just "mics the way macos does it"

16:00 <marcan> (AIUI the only reason AOP deals with this is "hey siri" on phones needs it and got carried over)

18:47 <janneg> so sway apparently has a red screen of death

18:53 <nicolas17> marcan: doing hey siri detection on the AOP is probably still a power saving even on Mac... negligible compared to phones, but

22:06 <ArcaneNibble> millenialhacker: AVE?

22:06 <ArcaneNibble> oh wait, not online

23:09 <ArcaneNibble> millenialhacker: AVE? don't think anyone has looked at that in detail