#
✕ nick:chadmed
#asahi-dev · 2023-08-04 12:48
chadmedthey have it on the DVC too but we dont touch that
#asahi-dev · 2023-08-04 12:48
chadmedbut it evidently sucks at it
#asahi-dev · 2023-08-04 12:48
chadmedthe chip tries to do it itself to "minimise audible artefacts"
#asahi-dev · 2023-08-04 12:47
chadmedoh its in the datasheet right there
#asahi-dev · 2023-08-04 12:47
chadmedi cant remember where i saw it
#asahi-dev · 2023-08-04 12:46
chadmedhttps://www.ti.com/lit/ds/symlink/tas2764.pdf?ts=1691153095843
#asahi-dev · 2023-08-04 12:46
chadmedanyway re confidentiality, the codecs are just TAS2764 with a different name lasered on to the top
#asahi-dev · 2023-08-04 12:45
chadmedits a cirrus hda thing, the driver is an out-of-tree monstrosity but it works ~the same as ours does (raw access to the individual drivers)
#asahi-dev · 2023-08-04 12:41
chadmedim there :)
#asahi-dev · 2023-08-04 12:40
chadmedi can crank absolutely heinous volumes through those and get some very obvious overexcursion but none of this digital-sounding popping noise
#asahi-dev · 2023-08-04 12:40
chadmednow that ive started testing on a 13" with different codecs but the same woofers, im leaning towards this being some limitation in the digital side of the codecs
#asahi-dev · 2023-08-04 12:38
chadmedthe bass hits cause the popping above ~70% volume
#asahi-dev · 2023-08-04 12:38
chadmedgood example is town.mid (the microsoft one) played through fluidsynth
#asahi-dev · 2023-08-04 12:37
chadmedeverything ive done has been by ear
#asahi-dev · 2023-08-04 12:37
chadmedand i would assume it adds _some_ latency of course but i havent benched or RE-ed any of the coreaudio stack beyond reading strings in plists
#asahi-dev · 2023-08-04 12:36
chadmedand you cant call it up from anywhere in macos
#asahi-dev · 2023-08-04 12:36
chadmedtheres a coreaudio preset which gives "raw" access to the output but i would wager this never, ever, ever gets used
#asahi-dev · 2023-08-04 12:35
chadmed(i actually even managed to get macos to pop once)
#asahi-dev · 2023-08-04 12:35
chadmedand then reading about popping and sudden changes in volumes in the asoc and TI documentation and put the pieces of the puzzle together
#asahi-dev · 2023-08-04 12:35
chadmedall i remember is kinda speculating as to how it works based on some strings in the obfuscated plists
#asahi-dev · 2023-08-04 12:34
chadmedits just a coreaudio plugin (audio unit)
#asahi-dev · 2023-08-04 12:34
chadmedits inserted into the dsp chain in macos just like our filter chain works
#asahi-dev · 2023-08-04 12:33
chadmedbut it doesnt
#asahi-dev · 2023-08-04 12:33
chadmedfor all i know it could actually just be overexcursion but if that were the case then it would appear with my 0 dBFS sine wave
#asahi-dev · 2023-08-04 12:32
chadmedim not really entirely sure how it works and its been over a year since i even looked at coreaudio's graph
#asahi-dev · 2023-08-04 12:32
chadmedyeah its a bit more than that though, its a lookahead delay with a ramping feature
#asahi-dev · 2023-08-04 12:30
chadmeddsharshakov: only when the content changes, pipewire's volume knob is silent
#asahi-dev · 2023-08-04 11:02
chadmedrapid volume changes below ~400hz cause popping which i have been thinking all this time is the result of overexcursion
#asahi-dev · 2023-08-04 10:59
chadmedok im starting to think that funky volume-ramping thing macos does might actually be required
#asahi-dev · 2023-08-04 08:49
chadmeddoing those glmark2 runs ages ago really made some of those keys seem like nonsense
#asahi-dev · 2023-08-04 08:48
chadmedah ok
#asahi-dev · 2023-08-04 08:47
chadmedoh right its not encrypted
#asahi-dev · 2023-08-04 08:46
chadmedtime to scratch up some power planes >:)
#asahi-dev · 2023-08-04 08:44
chadmedhmmm ok i think we're a bit higher than that with EAS at the desktop so there might be some gains left on the table in the cpus
#asahi-dev · 2023-08-04 08:43
chadmedwhat does PHPC say
#asahi-dev · 2023-08-01 06:05
chadmedi _think_ that timer intterupt FIQ thing only applies to the original AIC too but dont quote me on that
#asahi-dev · 2023-08-01 06:05
chadmedyeah AIC sits where GIC does normally aiui
#asahi-dev · 2023-08-01 06:01
chadmedAIC routes them to the application cores
#asahi-dev · 2023-08-01 05:58
chadmedaiui these socs do not implement gic
#asahi-dev · 2023-08-01 05:58
chadmedah theyre not in the logs
#asahi-dev · 2023-08-01 05:57
chadmeddid my messages not send
#asahi-dev · 2023-08-01 03:36
chadmedmarcan: were you planning to adopt speakersafetyd? ive merged your changes back into my repo and we can yeet the local version of the alsa crate now
#asahi-dev · 2023-07-24 11:48
chadmedhttps://github.com/chadmed/asahi-audio/tree/wip/j293 <-- can i get a peer review from someone who knows how to install this atm
#asahi-dev · 2023-07-23 10:41
chadmedi might push what i have for the 13" pros and get someone to test it to see if im right, or if i have to actually go buy a 13" m1/m2 pro
#asahi-dev · 2023-07-23 10:40
chadmedmy hypothesis is that nothing would have materially changed about the 13" pros from the introduction of that chassis
#asahi-dev · 2023-07-23 10:39
chadmedyeah that was expected
#asahi-dev · 2023-07-23 10:39
chadmedohhh yeah sorry misread
#asahi-dev · 2023-07-23 10:35
chadmedif there are differences between the m1 and m2 13" pros thats unfortunate though
#asahi-dev · 2023-07-23 10:34
chadmedkettenis: thanks :) just wanted to make sure im not going crazy
#asahi-dev · 2023-07-23 03:44
chadmedjust curious if the m1 13" is noticeably quieter than the 14" too

plus ›