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 2022-10-11

00:30:01 Quit tchan (Read error: Connection reset by peer)
00:37:10 Quit m01 (Quit: Konversation terminated.)
00:39:11 Join m01 [0] (
00:41:12 Quit speachy (Ping timeout: 264 seconds)
00:42:55 Join speachy [0] (~speachy@
00:42:55 Quit speachy (Changing host)
00:42:55 Join speachy [0] (~speachy@rockbox/developer/speachy)
00:42:55Mode"#rockbox +v speachy" by ChanServ (
00:55:47***Saving seen data "./dancer.seen"
01:25:35 Quit hexadecagram (Ping timeout: 252 seconds)
01:32:01 Join hexadecagram [0] (~acc@user/hexadecagram)
01:44:30__builtinCH23: cycles=1000 should have one period be 1,000/100,000 = 1/100 second ~= 100 Hz
01:44:52__builtinif you want 10 kHz, you need cycles=10
01:45:19__builtinspeachy: yeah, I looked into doing a scummvm port once
01:45:59__builtinlooked like a good candidate except for the C++ thing
01:46:30__builtinand also it being way to large to manually rewrite as C
01:46:43__builtin(which is what I did for xworld, actually)
01:47:15__builtinso yeah, if you get a working C++ plugin toolchain, I'll perhaps do a port
01:52:44sporkwould some of the arguments against mpegplayer not also apply to scummvm ?
01:53:22sporknot that i want to steal your fun
02:47:03__builtinwell, it wouldn't need any core support, presumably
02:47:20__builtinalthough now that I think about it, I think I did bring floating-point into the core to make quake run...
02:48:54__builtinah yep,
02:49:16__builtinbut also:
02:55:32__builtinspeachy: also, re: mpegplayer, I don't think "requires re-encoding to work" is a good enough reason to throw away the entire plugin
02:55:51***Saving seen data "./dancer.seen"
02:56:38__builtinremoving it from master now effectively kills it going forward, since it won't be maintained against future API changes
02:57:06__builtinand I'm sure there are users who use it and have movies/music videos encoded for on-device playback
02:57:10__builtinI do, for one
02:59:02__builtinand besides, rockbox is chock-full of obscure features that "nobody" uses, but arguably that's what makes rockbox... rockbox?
02:59:32sporkon some of the more modern players it might work well enough
02:59:56__builtin^ yep, on the full-size color ipods, it's actually quite usable
03:00:56__builtinand having mpegplayer brings us closer to feature parity to OF on those devices (although admittedly it's not H.264)
03:02:56paulcarrotymore features always better than less
03:05:50__builtinI don't want to make a slippery slope argument, but it's worth thinking about what the point of rockbox is at the end of the day - IMO, it's always existed as a "cool" alternative to OF, with lots of neat features, some of questionabe utility
03:08:42__builtinvideo playback is arguably a more mundane feature among those - but even so, not having it, if only to the limited extent we have now, is a big loss for rockbox, I think
03:11:42__builtinand of course, there's also features like doom/quake/duke3d/wolf3d, each of which is probably comparable to mpegplayer in degree-of-functionality ("it works, but do you really want to use it?"), and also which add some (minor) weight to the core
03:12:15sporkdoom and friends are harder to get to work too
03:12:23__builtinbut also arguably less-practical (but naturally also more cool)
03:12:39__builtinand I don't think we want to get rid of those
03:26:58 Quit CH23 (Quit: Ping timeout (120 seconds))
03:30:57 Join CH23 [0] (~CH23@revspace/participant/ch23)
04:09:12_bilgusso like I said make it self contained is a decent compromise
04:55:53***Saving seen data "./dancer.seen"
05:09:40 Join mendel_munkis [0] (
05:12:11 Quit munkis (Ping timeout: 260 seconds)
06:55:56***Saving seen data "./dancer.seen"
08:06:37 Join tchan [0] (
08:55:57***No seen item changed, no save performed.
09:25:14 Quit braewoods (Quit: WeeChat 3.5)
09:30:54 Join braewoods [0] (~braewoods@user/braewoods)
09:34:48speachyfwiw, I think the games are a lot more "useful" in their own right, especially on a device that has more/sufficient buttons.
09:40:44 Join massiveH [0] (~massiveH@2600:4040:a993:4900:d5e7:a68e:a8d4:46f)
09:51:12speachyI'm not against it going back in, even in its original form, but it's not someting we can use as a "feature" in today's era.
09:52:05speachyI"m reminded of some aftermarket car stereos I'd seen a while back, that sported a 2-ish-inch display on the face, claiming to support video playback.
10:04:15 Join benjaoming [0] (~benjaomin@
10:11:37speachyRe: newer GCC, it looks like we're stuck on 4.9 for android[-based] targets, not just the actual (mostly irrelevant) android builds but also the ibasso ports that rely on android userspace.
10:11:54speachywe do have some patches to bring android up to a newer state though
10:13:19speachyI did manage to get a newer toolchain put together for the other hosted targets, but we're stuck with old glibc+libraries and that makes me think we might just be better off staying on the existing 4.9 toolchains as they work and the needs of the platform preclude moving to anything truly modern.
10:14:38speachyI haven't done a new native arm toolchain yet. But on the xduoo x3, the main binary shrunk by ~10KB (~640K->~630K).
10:15:44speachyturns out our hosted toolchains arlready had c++ enabled, incidently. :)
10:20:33 Quit vup (Ping timeout: 252 seconds)
10:25:31paulcarrotyspeachy: do not remove video playback, core feature matters.
10:54:33 Join vup [0] (~~~~@
10:56:00***Saving seen data "./dancer.seen"
11:06:41 Quit michaelni (Quit: Leaving)
11:07:04 Join michaelni [0] (
11:18:08 Quit massiveH (Quit: Leaving)
11:24:56 Quit vup (Ping timeout: 260 seconds)
11:38:32speachy... let's see if my improvements to the toolchain dev script still work with the old stuff..
11:38:41sporkremoving it certainly helped the discussion
11:41:42speachyspork: indeed! It took a year for anyone to comment on the gerrit patch.
11:42:49speachy...oh, there are some "complications" involved in trying to get a freestanding libstdc++. Figures.
11:45:49speachyto work around that it'll probably require a two-stage bootstrap, like we have to do for the hosted targets.
11:58:58 Join vup [0] (~~~~@
12:38:45speachyI'm not worried about toolchain-induced bugs on linux-userspace-targeted builds. Only question will be is if there's a feature we want that simply doesn't exist or is buggy in gcc 4.9.
12:39:02speachyLTO is one example where newer is (much) better.
12:42:20rb-bluebotBuild Server message: New build round started. Revision 726673c638, 303 builds, 8 clients.
12:51:23speachyexisting toolchains still build cleanly with these changes but they bring some nice quality-of-life improvements to the script
12:56:02***Saving seen data "./dancer.seen"
13:02:43rb-bluebotBuild Server message: Build round completed after 1225 seconds.
13:02:45rb-bluebotBuild Server message: Revision 726673c638 result: All green
13:52:12 Join lebellium [0] (
13:55:13speachywoo, the autobuilder page finally shrunk down.
14:18:16speachygerrit upgraded to latest point release. yay, security holes.
14:56:04***No seen item changed, no save performed.
15:03:31 Quit Natch (Ping timeout: 250 seconds)
15:06:45 Join Natch [0] (
15:26:07 Quit hexadecagram (Quit: ZNC 1.8.2 -
15:35:38 Join hexadecagram [0] (~acc@user/hexadecagram)
16:56:07***Saving seen data "./dancer.seen"
17:19:33 Quit vup (Ping timeout: 265 seconds)
17:23:08speachyoh, in other news, the main rockbox server is going to require some downtime for OS upgrades.
17:23:19speachyvery overdue. :(
17:36:53 Quit hexadecagram (Quit: ZNC 1.8.2 -
17:38:12speachyI don't expect any issues with most things, but the PHP-based stuff could present with some problems that need fixing rapidly.
17:38:29 Join S|h|a|w|n [0] (~shawn156@user/shawn/x-4432647)
17:44:53 Join hexadecagram [0] (~acc@user/hexadecagram)
17:55:59 Quit hexadecagram (Remote host closed the connection)
18:08:12 Join hexadecagram [0] (~acc@user/hexadecagram)
18:44:25 Quit lebellium (Quit: Leaving)
18:45:27 Join vup [0] (~~~~@
18:56:11***Saving seen data "./dancer.seen"
20:14:48 Join dconrad [0] (~dconrad@
20:15:09 Quit CH23 (Quit: Connection closed)
20:15:43 Join CH23 [0] (~CH23@revspace/participant/ch23)
20:28:58 Quit CH23 (Quit: Connection closed)
20:56:14***Saving seen data "./dancer.seen"
21:16:18 Quit hexadecagram (*.net *.split)
21:16:18 Quit Natch (*.net *.split)
21:16:18 Quit tchan (*.net *.split)
21:16:18 Quit mendel_munkis (*.net *.split)
21:16:18 Quit speachy (*.net *.split)
21:16:18 Quit rogeliodh (*.net *.split)
21:16:18 Quit hombrelaser (*.net *.split)
21:16:18 Quit funman (*.net *.split)
21:16:40 Join speachy [0] (~speachy@
21:16:51 Join rogeliodh [0] (
21:16:55 Quit speachy (Changing host)
21:16:55 Join speachy [0] (~speachy@rockbox/developer/speachy)
21:16:55Mode"#rockbox +v speachy" by ChanServ (
21:16:57 Join tchan [0] (
21:17:48 Join hombrelaser [0] (~my@user/hombrelaser)
21:18:15 Join hexadecagram [0] (~acc@user/hexadecagram)
21:21:23 Join funman [0] (
21:21:39 Join Natch [0] (
21:22:53othello7I would like to note, I used mpegplayer on a regular basis. I've heard it said that "it was never more than a tech demo" but I used for everything from watching videos on the go, to playing a music videos to drum to. about 1/4 of my ipod's storage space was dedicated to mpg files. I created a simple bash script to automatically re-encode the videos to fit perfectly on my ipod and it didn't even take long to do so on my 11 ye
21:23:44othello7even when I used an ipod 4g grayscale instead of the ipod video I have now, it was amazing to be able to *play videos*, something that you couldn't even do with default OS
21:24:17othello7is there any reason for removing it other than it is hard to maintain and "nobody uses it"?
21:33:39 Join mendel_munkis [0] (
21:43:57 Quit dconrad (Remote host closed the connection)
21:45:37 Join captainkewlllll [0] (
21:48:14 Quit captainkewlllll (Client Quit)
21:54:07 Join dconrad [0] (~dconrad@
21:58:13 Quit dconrad (Ping timeout: 250 seconds)
22:12:08 Join dconrad [0] (~dconrad@
22:35:29 Quit dconrad ()
22:56:18***Saving seen data "./dancer.seen"
23:02:14 Quit rogeliodh (Quit: Ping timeout (120 seconds))
23:22:32 Join rogeliodh [0] (

Previous day | Next day