#asahi-dev · 2023-02-06 13:26
povikit's up to you but i think it would be best to save it for posterity#asahi-dev · 2023-02-06 13:22
povikmarcan: please commit the notebook from the stream, i will see if i can write the exponentials below :p#asahi-dev · 2023-02-06 09:33
povikbut i saw the issue a couple of times#asahi-dev · 2023-02-06 09:33
povik( https://github.com/povik/macaudio-userutils/blob/master/scripts/stresstest.sh )#asahi-dev · 2023-02-06 09:33
povikchadmed: not that i know of, even rapid restarting didn't trigger it for me#asahi-dev · 2023-02-05 21:18
povikprobable fix for the stuck ADMAC IRQs: https://tpaste.us/Vqbz#asahi-dev · 2023-02-05 08:01
poviknot necessarily, just very cold :p#asahi-dev · 2023-02-05 08:01
povikfor estimating the parameters of the model#asahi-dev · 2023-02-05 08:01
povikmeasurements on a laptop that's initially freezing might be the most accurate#asahi-dev · 2023-02-05 07:55
povikerr, i mean you need to *comment out* disabling of the switches#asahi-dev · 2023-02-05 07:54
povikbtw to bring it up on a new model, you need to: 1. describe the sense data slots in DT; 2. uncomment the ISENSE/VSENSE switches being disabled in macaudio#asahi-dev · 2023-02-05 07:48
povikmarcan: that repo is where the research you may need to do can go as well :)#asahi-dev · 2023-02-05 07:47
povikthis counts joules...#asahi-dev · 2023-02-05 07:47
povikhttps://github.com/povik/macaudio-userutils/blob/master/scripts/live_powersum.py#asahi-dev · 2023-02-05 07:47
povikwell, i have been doing some experiments myself :p#asahi-dev · 2023-02-05 07:46
poviki have been sitting on it as i still want to polish a few things#asahi-dev · 2023-02-05 07:46
povik(at least on j314)#asahi-dev · 2023-02-05 07:46
povikthe status is that it's pretty much ready in https://github.com/povik/linux/tree/ivsense#asahi-dev · 2023-02-05 07:46
povikmarcan: i did ask and you haven't answered, but since then i started getting samples on j274 so indeed it looks like it's wired up on all models!#asahi-dev · 2023-01-24 14:33
povikhow come this wasn't fixed when the other m2 devices were being brought up?#asahi-dev · 2023-01-23 08:42
povikyeah, that could be more straightforward#asahi-dev · 2023-01-23 08:30
povikinternally routed to a different frontend#asahi-dev · 2023-01-23 08:30
povikit's because the sense data uses a different slot width that it needs to be done like this#asahi-dev · 2023-01-23 08:29
povikit's still generated by MCA but in a different frontend#asahi-dev · 2023-01-23 08:29
povikyes, in a sense#asahi-dev · 2023-01-23 08:29
povikhaven't done much suspending myself O:)#asahi-dev · 2023-01-23 08:29
povikah, i guess it applies then#asahi-dev · 2023-01-23 08:26
poviki'm taking name suggestions#asahi-dev · 2023-01-23 08:26
povikno, this is about something that happens with ivsense capture#asahi-dev · 2023-01-23 08:23
povikmarcan: opinion on the above?#asahi-dev · 2023-01-23 08:23
povikhttps://github.com/povik/linux/commit/5fb55674eb82744012a49691cc6cdbcf8ad7b4d9#asahi-dev · 2023-01-21 20:33
poviki plan to continue working on it monday, to test it on j314#asahi-dev · 2023-01-21 20:32
povikchadmed: not at the moment (while i pushed an ivsense branch yesterday it's not ready to go)#asahi-dev · 2023-01-19 07:23
povikthat never shipped in the reference distro, right?#asahi-dev · 2023-01-19 07:23
povikoh dear, so it seems fedora shipped an asahi kernel with the jack distortion bug#asahi-dev · 2023-01-18 09:17
povikwho's that?#asahi-dev · 2023-01-18 09:13
povikchadmed: i only remember certain quotes about lemons...#asahi-dev · 2023-01-18 09:11
poviki think we are all in agreement here#asahi-dev · 2023-01-18 09:11
povikno, i was thinking of UX, not safety#asahi-dev · 2023-01-18 08:51
povikah#asahi-dev · 2023-01-18 08:50
povikand if pipewire respects UCM, it shouldn't offer that sink to users#asahi-dev · 2023-01-18 08:50
povikor we can *not* describe the speakers in UCM, making that woofer baseline unavailable by default#asahi-dev · 2023-01-18 08:48
povikbut that is pretty separate from the safety issue, you are right#asahi-dev · 2023-01-18 08:48
povikyeah, i was thinking of not exposing the direct speaker sinks#asahi-dev · 2023-01-18 08:47
povikbut i don't have to think about that right now O:-)#asahi-dev · 2023-01-18 08:47
povikthere's of course still the issue of how the pipewire DSP layer falls into the scheme#asahi-dev · 2023-01-18 08:46
povikas dsharshakov points out, if we don't describe it in UCM it won't show up as a microphone in user interfaces#asahi-dev · 2023-01-18 08:45
povikbut a a separate PCM, yes#asahi-dev · 2023-01-18 08:45
povikno, no new soundcard#asahi-dev · 2023-01-18 08:45
povikmarcan: not really. this must be more than a quantization noise, it's above the level of quanta, but it's all in accord with the noise figures from datasheet -- i just drove a weak signal to the speaker when i captured the trace