#asahi-dev · 2023-09-01 14:23
chadmedno i mean the offical advice is the opposite#asahi-dev · 2023-09-01 14:14
chadmedPaulFertser: they meant safe to use absent speakersafetyd. the official advice is that they are _not_ safe to use without speakersafetyd#asahi-dev · 2023-09-01 14:12
chadmedeiln: i have on the city shore in my collection, its one of my favourite albums :)#asahi-dev · 2023-09-01 13:06
chadmedthey dont have tweeters#asahi-dev · 2023-09-01 13:05
chadmedwell yeah i dont think you can break the 13" pros#asahi-dev · 2023-09-01 12:23
chadmedthe 13" pro m1 is j293 which is in wip/j293#asahi-dev · 2023-09-01 12:20
chadmedthe 13" airs need doing though! #asahi-dev · 2023-09-01 12:20
chadmedknedlik: ive already done the 13" pros.... sorry :/#asahi-dev · 2023-09-01 10:09
chadmedbah change of plans the gentoo livecd doesnt like the fedora kernel at all #asahi-dev · 2023-09-01 07:08
chadmedlets fix that#asahi-dev · 2023-09-01 07:08
chadmedthis thing actually sounds ass above like 35% volume lmfao wtf #asahi-dev · 2023-09-01 07:07
chadmedooh interesting theyve gone even harder on the compressor than with the 14" pro#asahi-dev · 2023-09-01 07:04
chadmedthats ok i need to update the gentoo stuff to work with fedora now anyway :) #asahi-dev · 2023-09-01 06:59
chadmedok 15" air booted at 16:57, how quick can i get from here to gentoo to speakers#asahi-dev · 2023-08-31 08:06
chadmedjannau: ah i was looking in the wrong -wip tree :)#asahi-dev · 2023-08-23 13:46
chadmedtobhe: yes it is, you basically just have to scrape bits of it together from various android source dumps, theres no single source of truth for it afaik #asahi-dev · 2023-08-21 09:50
chadmedoh lmfao#asahi-dev · 2023-08-21 09:50
chadmedi can edit#asahi-dev · 2023-08-20 13:39
chadmedit took me days to wrap my head around why sample-wise multiplication works#asahi-dev · 2023-08-20 13:39
chadmedso am i, im too dumb to implement DSP#asahi-dev · 2023-08-20 13:37
chadmedunfortunately that means the multiband is back, but LSP's is better than macos's and uses way less cpu so we're still doing better#asahi-dev · 2023-08-20 13:37
chadmedthe same way macos does, just throw more compressor at the input until it stops doing it :)#asahi-dev · 2023-08-20 13:26
chadmed(no point in anyone testing btw, you need pipewire and wireplumber special sauce)#asahi-dev · 2023-08-20 13:25
chadmeda ~done j314 dsp graph is now in asahi-audio/main#asahi-dev · 2023-08-20 13:24
chadmedpopping noises dealt with#asahi-dev · 2023-08-20 13:24
chadmedone again spite proves to be the greatest of motivators#asahi-dev · 2023-08-20 01:00
chadmedthe LSP one introduces too much delay#asahi-dev · 2023-08-20 01:00
chadmed*sigh* might have to write an equal loudness plugin too....#asahi-dev · 2023-08-19 13:29
chadmedhttps://chadmed.au/res/buzzing.mp4#asahi-dev · 2023-08-19 13:17
chadmedwhen i was running frequency sweeps in macos for fun ages ago i hit it#asahi-dev · 2023-08-19 13:17
chadmedOH thats right macos is also prone to this! #asahi-dev · 2023-08-19 13:16
chadmedill upload some videos hang on#asahi-dev · 2023-08-19 13:16
chadmedthough the fact that we start getting harmonics in the output of a pure sine wave just before we turn the volume up enough to get the noise (and the fact that its a hard cut in) suggests that the digital side of the amp is saturated to me#asahi-dev · 2023-08-19 13:16
chadmedits certainly possible and its something i can and will test once the visense stuff is merged#asahi-dev · 2023-08-19 11:17
chadmedi should note that it doesnt matter if the sine wave is being generated at 0 dBFS or less, its definitely a codec thing#asahi-dev · 2023-08-19 11:09
chadmedas i said last night the quirks povik added to sound/ivsense-improvements seem to do *something* but i didnt get around to doing much testing on that branch#asahi-dev · 2023-08-19 11:08
chadmedfurthermore the 160hz sine wave contains very audible harmonics so the digital side of the chip seems to be saturated#asahi-dev · 2023-08-19 11:08
chadmedan A4 sine wave does not #asahi-dev · 2023-08-19 11:08
chadmedmarcan: a sine wave at 160 hz will cause hard digital clipping at 65% going straight into the woofers #asahi-dev · 2023-08-18 13:38
chadmedit still fails reliably on machines with tas2770 supposedly but idk whether thats due to the codecs or how apple has wired up the gpio pins on those particular boards#asahi-dev · 2023-08-18 13:38
chadmedhmm okay#asahi-dev · 2023-08-18 13:33
chadmedwill test that too #asahi-dev · 2023-08-18 13:33
chadmedgiven the only failure mode left seems to be rapidly cycling awake-sleep-awake the issue now is probably just that the sleeps in the pm ops are not long enough#asahi-dev · 2023-08-18 13:32
chadmedand finally get sleep at least halfway towards working #asahi-dev · 2023-08-18 13:31
chadmedshould pull https://github.com/chadmed/linux/tree/snd-pm-v3 in too when we merge this #asahi-dev · 2023-08-18 13:29
chadmedall the ivsense/sample rate detection stuff worked fine#asahi-dev · 2023-08-18 13:29
chadmednah they were related to something else, something about cpu pstates#asahi-dev · 2023-08-18 13:28
chadmed(i did update the dtbs)#asahi-dev · 2023-08-18 13:28
chadmeddidnt test long enough to hit it, got a load of kernel oopses and stuff so just came back to 6.4-asahi-6 #asahi-dev · 2023-08-18 13:27
chadmedi pulled sound/ivsense-improvements and tried it earlier and there was an audible difference in the sound so those quirks must do something important