#asahi · 2024-10-25 21:13
jannaunote that Control links into the steam store#asahi · 2024-10-25 21:08
jannauwine is not supported in this preview/alpha: https://docs.fedoraproject.org/en-US/fedora-asahi-remix/x86-support/#_can_i_run_windows_applications_outside_of_steam#asahi · 2024-10-25 20:59
jannautry interacting with menues. that's the main issue. no scaling is second issue#asahi · 2024-10-25 20:49
jannauyou can exit big picture mode to convince yourself from the brokenness#asahi · 2024-10-25 20:48
jannaubecause it is. the windowed interface is broken due sommelier/xwayland#asahi · 2024-10-16 17:35
jannaum1n1 1.4.16 suspends the display on boot so it should work directly after booting in linux#asahi · 2024-10-16 17:18
jannaurussell--: does it work after closing and opening the lid / have you updated macos to sequoia? That's an macos bug but the latest m1n1 release has a workaround#asahi · 2024-10-12 08:29
jannauTramtrist: you can add entries to $STEAM/config/libraryfolders.vdf just the path and label is enough#asahi · 2024-10-09 14:51
jannauprobably not but Longhorn is cclaimed that it's not necessary with nvidia and the driver for old AMD GPUs (most likely slower than agx) apparently doesn't need unaligned accesses anymore#asahi · 2024-10-09 14:43
jannaumarcan got the pcie controller working on the mac pro (probably) but all hardware is behind a large PCIe switch#asahi · 2024-09-28 13:46
jannauno, we don't support out of tree modules and I don't think anyone is looking specifically at zfs + asahi#asahi · 2024-09-26 18:44
jannauwhich macos version? 13.5 should work newer ones might be broken. Apple added features / checks which made running macos under HV harder. Might be macos boot args as well. run with "debug=0x14e serial=3 apcie=0xfffffffe -enable-kprintf-spam wdt=-1 clpc=1"#asahi · 2024-09-11 08:04
jannauyes, needs dsc. for DP we need to parse the DisplayID data#asahi · 2024-09-11 07:45
jannauhttps://git.linuxtv.org/edid-decode.git/#asahi · 2024-09-11 07:44
jannauI'm using whatever fedora packages#asahi · 2024-09-11 07:31
jannauthat one is new#asahi · 2024-09-11 07:31
jannaufrom "Block 2, CTA-861 Extension Block"#asahi · 2024-09-11 07:30
jannauVTDB 1: 3840x2160 239.990737 Hz 16:9 582.938 kHz 2331.750000 MHz (RB, aspect 16:9, no 3D stereo)#asahi · 2024-09-11 07:18
jannaubut it shouldn't make a difference because the review is recent#asahi · 2024-09-11 07:16
jannauare you testing with the latest macOS version?#asahi · 2024-09-11 07:16
jannau"Sonoma 14.4.1" from that link doesn't sound good#asahi · 2024-09-11 07:13
jannau4k 240 Hz monitors?#asahi · 2024-09-11 07:12
jannauagreed#asahi · 2024-09-11 07:12
jannaurtings also hasn't gotten more than 120Hz on macOS#asahi · 2024-09-11 07:10
jannau "However, on 07/11/24, Samsung released the 1003.3 firmware version" does Samsung provide firmware upgrades?#asahi · 2024-09-11 07:07
jannauI still haven't parsed the DisplayID timings seen with DP#asahi · 2024-09-11 07:00
jannauI think Apple uses https://www.paradetech.com/products/ps196/ or something based on that. So it supports HDMI 2.1 FRL6 with 48Gbit/s#asahi · 2024-09-11 06:55
jannauoh, I missed your "yes" to the question if there's a stable picture#asahi · 2024-09-11 06:54
jannaure: VTDB (Video Timing Display Bandwidth), so that's something different than we thought#asahi · 2024-09-11 06:52
jannauif yes please try to read it from sysfs#asahi · 2024-09-11 06:50
jannauleio: does the display works now with 'apple_dcp.unstable_edid'#asahi · 2024-09-11 06:35
jannauleio: while you have a stable picture?#asahi · 2024-09-10 21:31
jannaucan you get the EDID from your monitor connected via HDMI to the M2 Ultra? Since asahi-6.10.7-2 the linux driver supports querying the EDID if "apple_dcp.unstable_edid" is specified on the kernel command line#asahi · 2024-09-10 21:26
jannauleio: that's just FullHD with 240 Hz which is less of a problem. it requires the same bandwidth as 4k at 60 Hz i.e. is fully in the limits of HDMI 2.0#asahi · 2024-09-10 16:50
jannauhdmi one is unchanged, the DP one has different peak luminance values#asahi · 2024-09-10 15:50
jannauleio: new EDIDs please if it didn't start working#asahi · 2024-09-10 09:07
jannauyes, the connector of choice for DP is USB-C as far as apple is concerned#asahi · 2024-09-10 09:06
jannauhmm, apple's CDN is slow today. 10h estimated download time for 14.5GB sequoia beta update with 350MB aleready downloaded#asahi · 2024-09-10 09:02
jannauI'd say it's more likely that they remove the HDMI port than replacing it with full size DP#asahi · 2024-09-10 09:01
jannaualso for TV production it might be useful to be able to drive a TV directly#asahi · 2024-09-10 09:00
jannauyou obviously are supposed to buy an apple thunderbolt display. I don't see them removing the hdmi port on mac minis. If you have it already there and on the laptops just use the same solution on the low volume desktop parts and do not make more work for yourself#asahi · 2024-09-10 08:12
jannausounds like bad timing. I was surprised that you went ahead after our conversation#asahi · 2024-09-10 07:42
jannauit would be helpful to have the edid from a display for which dcp supports 4k 240Hz#asahi · 2024-09-10 07:41
jannaubut I had to remove CTA timing extensions since I can only inject 2 blocks / 256 bytes#asahi · 2024-09-10 07:39
jannauleio: that probably won't work. at least it doesn't with my way of injecting edids#asahi · 2024-09-09 21:41
jannauleio: https://paste.centos.org/view/ae7841c0 that's the edid from the arch linux forum recreated#asahi · 2024-09-09 20:30
jannaudifferent tool capabilities#asahi · 2024-09-09 20:29
jannauthat output is from https://people.freedesktop.org/~imirkin/edid-decode/#asahi · 2024-09-09 20:29
jannauno, that's for the 384 byte one. the hdmi has just 2 blocks of 128 bytes. the DisplayID is in the 3rd block#asahi · 2024-09-09 20:26
jannauI get https://paste.centos.org/view/7463149b