Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2025-05-15

00:40:38***Saving seen data "./dancer.seen"
01:00
01:17:15 Quit Burak_ (Remote host closed the connection)
01:17:41 Join Burak_ [0] (~Burak@185.25.123.34)
02:00
02:25:35 Quit Galois (Remote host closed the connection)
02:40:39***Saving seen data "./dancer.seen"
02:42:14 Quit advcomp2019__ (Read error: Connection reset by peer)
03:00
03:14:25 Join Galois [0] (djao@efnet.math.uwaterloo.ca)
03:15:04 Quit othello7 (Ping timeout: 244 seconds)
03:24:05 Nick JanC is now known as Guest138 (~janc@user/janc)
03:24:05 Quit Guest138 (Killed (zinc.libera.chat (Nickname regained by services)))
03:24:09 Join JanC [0] (~janc@user/janc)
03:43:18 Quit JanC (Killed (platinum.libera.chat (Nickname regained by services)))
03:43:22 Join JanC [0] (~janc@user/janc)
04:00
04:24:41 Quit JanC (Killed (copper.libera.chat (Nickname regained by services)))
04:24:46 Join JanC [0] (~janc@user/janc)
04:40:43***Saving seen data "./dancer.seen"
05:00
05:06:28 Quit JanC (Killed (tantalum.libera.chat (Nickname regained by services)))
05:06:33 Join JanC [0] (~janc@user/janc)
06:00
06:17:53 Quit Galois (Ping timeout: 248 seconds)
06:28:30 Quit JanC (Ping timeout: 252 seconds)
06:29:12 Join JanC [0] (~janc@user/janc)
06:40:46***Saving seen data "./dancer.seen"
07:00
07:04:28 Quit JanC (Killed (erbium.libera.chat (Nickname regained by services)))
07:04:32 Join JanC [0] (~janc@user/janc)
07:10:18 Quit Moriar (Ping timeout: 252 seconds)
07:30:44 Quit jj5 (Quit: Konversation terminated!)
07:31:01 Join jj5 [0] (~jj5@100.80.216.139.dynamic.dsl.dv.iprimus.net.au)
08:00
08:01:58 Join Galois [0] (djao@efnet.math.uwaterloo.ca)
08:14:09 Quit desowin (Ping timeout: 245 seconds)
08:17:46 Quit rogeliodh (Quit: The Lounge - https://thelounge.chat)
08:18:37 Join rogeliodh [0] (~rogeliodh@rogeliodh.dev)
08:35:39 Nick JanC is now known as Guest1903 (~janc@user/janc)
08:35:39 Quit Guest1903 (Killed (platinum.libera.chat (Nickname regained by services)))
08:35:43 Join JanC [0] (~janc@user/janc)
08:40:49***Saving seen data "./dancer.seen"
08:47:56 Quit braewoods (Remote host closed the connection)
08:48:15 Join braewoods [0] (~braewoods@user/braewoods)
09:00
09:17:57 Join advcomp2019 [0] (~advcomp20@user/advcomp2019)
09:18:18 Join braewoods_ [0] (~braewoods@user/braewoods)
09:18:23 Quit braewoods (Read error: Connection reset by peer)
09:25:49 Join desowin [0] (~desowin@rockbox/developer/desowin)
09:31:29skrzypHmm, why XM/S3M/IT files are played but this weird "external" libmikmod "application" with its own UI where MOD and SID files are opening in native player?
09:31:54skrzypWhat's needed to move the XM/IT replayer into the native player UI?
10:00
10:12:19skrzypActually, I've dug into this and seems like it should be a "codec" instead of "plugin"
10:12:34skrzypAnd looking at the codec sources, it doesn't seem to be that bad.
10:13:33skrzypMoving out from libmikmod in favor of - for example - libxmp might also be beneficial, since it's more faithful and uses less resources from my own research, back then when I was creating simple C based player with tracker music support
10:40:50***Saving seen data "./dancer.seen"
11:00
11:07:59 Join thanosengine [0] (~thanos@user/thanosengine)
11:09:30 Quit JanC (Killed (lithium.libera.chat (Nickname regained by services)))
11:09:35 Join JanC [0] (~janc@user/janc)
11:11:56 Quit mrkrisprolls (Ping timeout: 272 seconds)
11:15:01 Quit COMPL_EXE (Ping timeout: 248 seconds)
11:15:22 Join COMPL_EXE [0] (~compl.exe@aosc/dev/origincode)
11:16:23 Quit aaabbb (Read error: Connection reset by peer)
11:24:05 Quit kirvesAxe (Ping timeout: 248 seconds)
11:39:14 Join kirvesAxe [0] (kirvesaxe@user/kirvesaxe)
11:47:49 Join mrkrisprolls [0] (~mrkrispro@2605:6400:20:5cc:e8af:da1c:ee53:f678)
11:50:57_bilgusskrzyp, the way plugins are implemented is pretty similar to the way codecs are done, there should be no reason you can't do that
12:00
12:01:31skrzyp_bilgus: so simply migrating from plugin to codec might work?
12:01:50skrzypTo have a consistent playback experience, as I have folders/playlists with lots of MOD + XM files
12:02:56_bilgusI doubt it will 1:1 or anything but they have the same basic building blocks and a core provided API for the common stuff
12:04:05_bilgusthe codecs basically are plugins just slightly different aims
12:28:26 Quit braewoods_ (Remote host closed the connection)
12:28:40 Join braewoods_ [0] (~braewoods@user/braewoods)
12:40:51***Saving seen data "./dancer.seen"
12:47:15 Join braewoods__ [0] (~braewoods@user/braewoods)
12:47:15 Quit braewoods_ (Read error: Connection reset by peer)
12:48:56 Quit braewoods__ (Remote host closed the connection)
12:49:11 Join braewoods__ [0] (~braewoods@user/braewoods)
12:57:26 Join lebellium [0] (~lebellium@2a01cb0405d03a004070cdf0d3f7e170.ipv6.abo.wanadoo.fr)
12:58:14_bilgusany thoughts on g#6549?
12:58:17rb-bluebotGerrit review #6549 at https://gerrit.rockbox.org/r/c/rockbox/+/6549 : Skin Engine FS #13594 - %t subline tag doesn't run at full speed below 0.5s WIP RFC by William Wilgus
13:00
13:00:23_bilgusthis patch adds a new event SYS_REFRESH_WAITING which allows lowering the button queue timeout, currently only the line alternator but could be extended to the peak meter and a few other places
13:01:58_bilgusOr! a better idea / mechanism??
13:37:25 Quit braewoods__ (Remote host closed the connection)
13:37:34 Join braewoods [0] (~braewoods@user/braewoods)
13:39:05 Join bahusoid [0] (~bahusoid@91.215.146.97)
13:40:47 Join braewoods_ [0] (~braewoods@user/braewoods)
13:40:58 Quit braewoods (Remote host closed the connection)
13:42:56 Quit braewoods_ (Remote host closed the connection)
13:43:10 Join braewoods_ [0] (~braewoods@user/braewoods)
13:44:56 Quit braewoods_ (Remote host closed the connection)
13:45:10 Join braewoods_ [0] (~braewoods@user/braewoods)
13:45:28bahusoidskrzyp: There is one catch with codec. Metadata handling is placed in core (e.g. see get_aac_metadata). So, you can't use libs methods for parsing and need to extract logic with minimal impact on core.
13:47:21 Quit bahusoid (Quit: Client closed)
14:00
14:04:57 Quit braewoods_ (Remote host closed the connection)
14:05:10 Join braewoods_ [0] (~braewoods@user/braewoods)
14:11:05 Join braewoods [0] (~braewoods@user/braewoods)
14:11:27 Quit braewoods_ (Read error: Connection reset by peer)
14:14:22 Join braewoods_ [0] (~braewoods@user/braewoods)
14:14:41 Quit braewoods (Read error: Connection reset by peer)
14:16:52 Join braewoods__ [0] (~braewoods@user/braewoods)
14:17:43 Quit braewoods_ (Read error: Connection reset by peer)
14:30:33 Quit braewoods__ (Remote host closed the connection)
14:30:41 Join braewoods [0] (~braewoods@user/braewoods)
14:37:03 Quit braewoods (Ping timeout: 245 seconds)
14:37:30 Join braewoods [0] (~braewoods@user/braewoods)
14:40:55***Saving seen data "./dancer.seen"
14:42:34 Quit braewoods (Remote host closed the connection)
14:42:43 Join braewoods [0] (~braewoods@user/braewoods)
14:48:10 Join facepunch01 [0] (~facepunch@45.131.79.73)
14:50:44speachyskrzyp: mikmod vs xmp's fidelity is both about as bad in a general sense
14:51:35speachybut the main reason why most modules play externally is due to how we have to load the entire file at once; modern tracker files are simply too large for our "codec" allocation.
14:52:11speachyplus you need to parse the whole file to figure out how long it is.
15:00
15:13:42speachyour codec API is written assuming a fixed codec state, streaming compressed data in and getting raw audio out.
15:14:19speachythe codec only sees a small portion of the file at a time, and that read buffer is managed externally.
15:16:09jssfr(which is a very sensible design)
15:28:23speachybut yeah, metadata is another problem; tracker files don't have anything formally structured.
16:00
16:40:57***Saving seen data "./dancer.seen"
17:00
17:09:58 Join Moriar [0] (~moriar@107-200-193-159.lightspeed.stlsmo.sbcglobal.net)
17:46:43skrzypsong "title" is enough, all the "metadata" are done in instrument names, and maybe the only way to show that is as "lyrics"
17:47:28speachythere's no "title", just a filename.
17:48:21skrzypanyways, MOD still works as a codec, and while its size is limited to 2MB (due to maximum amount of chipram in real world Amiga with AGA), the other module formats are not that much larger
17:48:48skrzypSIDs or SAPs are worse, since this is mostly executing 6502 code bitbanging data onto registers
17:48:55skrzypbut these are still codecs :P
17:48:58speachywell, 2MB minus the state of the player itself
17:49:47speachySIDs are quite tiny
17:50:00speachyand can be streamed
17:51:44speachysame argument applies to midi files btw
17:51:58speachythe actual "song" data is fairly small, but the instruments are what get you.
17:51:58 Quit lebellium (Quit: Leaving)
17:52:19speachy(plus there's still the problem of having to effectively parse the whole file to work out how long it is)
17:53:49speachy(smaller RAM targets get a lot less memory for codecs btw)
17:55:08speachyactually most targets only have 1MB for the codec's private use.
17:56:05speachya bunch have 512K and the smallest, only 288K. Those also have a restricted codec set.
18:00
18:04:43speachygranted, if one excludes Impulse Tracker (with its 16-bit instrument support) file sizes get a lot smaller.
18:05:27speachyaverage size of (zipped) IT files in modarchive's 2022 collection is 988K, XM is 617K, S3M is a mere 167K.
18:21:50speachyour mikmod plugin clocks in at ~246K for code and static data structures on an ipod6g.
18:22:26speachyalmost none of that is UI-related.
18:28:50speachyIn other words, at best, module files have to be under 700K to fit if played via the regular codec system.
18:31:46speachyour (admittedly crappy) midi player clocks in at 146K, but 128K of that is an output buffer.
18:34:13speachysorry, 64K of that mikmod size is an output buffer too. so maybe we'd be able to squeeze an 800K module in there?
18:41:01***Saving seen data "./dancer.seen"
20:00
20:03:32 Join massiveH [0] (~massiveH@108.50.181.86)
20:41:04***No seen item changed, no save performed.
21:00
21:15:50 Quit baltazar (Ping timeout: 252 seconds)
21:17:33 Join baltazar [0] (~baltazar@user/baltazar)
22:00
22:41:06***Saving seen data "./dancer.seen"
23:00
23:00:43 Join aaabbb [0] (sitku@user/aaabbb)
23:39:35 Quit massiveH (Quit: Leaving)

Previous day | Next day