Main page
[00:10] realityking (n=realityk@p57BDEBB5.dip.t-dialin.net) left #handbrake-dev.
[00:12] ritsuka (n=Ritsuka@87.19.240.219) left irc:
[00:25] [jbrjake] i split that off to tiki bar
[00:26] [eddyg] FYI I have modified the reader.c and Controller.mm to allow the same audio stream to be selected in multiple tracks. This is a stepping stone on the way to AC3/AAC together.
[00:26] [eddyg] Diff in a moment.
[00:26] *** s55 really would love just to make the develpoment forum developers only and have a custom error message telling users to post in a development subform for users to bithc
[00:27] [s55] and maybe a seperate one for keeping patches only
[00:27] [eddyg] http://pastebin.ca/905684
[00:28] [eddyg] So now you can have different mixdowns for the same audio track.
[00:28] [eddyg] in the same movie.
[00:28] [s55] thought you could do that already? or did I mis-understand that?
[00:29] [eddyg] No - you can't have the same input audio track going to two outputs at the moment
[00:29] [eddyg] It's one->one
[00:29] [eddyg] I've changed that to one->many
[00:29] [s55] thats kinda cool :)
[00:29] [eddyg] Which was one of the hurdles with having AAC/AC3 from one input track
[00:32] [eddyg] My ideal Mac implementation of the GUI would be to add an extra mixdown for AAC/AC3, but it's not easy.
[00:33] [eddyg] Hopefully soon though.
[00:33] [eddyg] So unless anyone has any issues I want to commit that existing patch right away - and then move on to other stuff.
[00:34] [eddyg] The reason for committing now is that it allows others to work on the same issue in parallel.
[00:34] [eddyg] They may not have been comfortable doing this initial change.
[00:36] [jbrjake] commit it, eddyg
[00:37] [eddyg] ta
[00:37] [jbrjake] not for others working in parallel, but because dynaflash and i have a few macosx/ changes to check in
[00:37] [jbrjake] unreleated
[00:37] [jbrjake] oh btw, did you see what i was asking you about earlier, with the vfr delay queue?
[00:37] [CIA-1] eddyg * r1266 /trunk/ (libhb/reader.c macosx/Controller.mm): Allow multiple output audio tracks from a single DVD audio track.
[00:38] [hbrbot] ::hb_trac:: Changeset [1266]: Allow multiple output audio tracks from a single DVD audio track. @ http://trac.handbrake.fr/changeset/1266
[00:38] [eddyg] Sorry - i didn't want to get sidetracked from the Ac3 thing, so I didn't even look. I couldn't reply at the time as I was in a different room just reading the irc logs.
[00:41] [jbrjake] k. it does need to be dealt with before 0.9.2, though.
[00:44] Dr3w (n=andrewmc@accendo.demon.co.uk) left irc: Client Quit
[00:49] [s55] is that a joke I see john?
[00:50] [s55] you actually joke?
[00:50] [s55] you have humor?
[00:50] [s55] humour*
[00:51] [jbrjake] i watch a lot of sitcoms, it rubs off ;p
[00:52] [jbrjake] anyway, i do have Issues with locking down the dev forum. as much frustration as people like gswhite cause me by polluting it with bullshit n00b hand-holding, we've gotten a lot of valuable contributions from people popping in there
[00:52] [jbrjake] as i recall, maurj, eddyg, and van all started off by commenting on dev posts before they ever submitted patches
[01:10] [s55] hence the custom message telling people to post in the subforum ;)
[01:11] alcuin (n=alcuin@p50886F3F.dip.t-dialin.net) joined #handbrake-dev.
[01:11] [s55] a subforum dedicated only for patches could be useful though? Saves with the clutter on the main develpoment forum which is ebtter suited for discussion
[01:15] [jbrjake] wtf, Ozy thinks he can notice a 50ms audio de-sync? is that even possible? i thought the low-end was 100ms to 150ms for very perceptive people.
[01:16] [eddyg] We only track the sync at 100ms
[01:16] [eddyg] So it could well be out by 50ms
[01:16] [eddyg] But who cares :)
[01:16] [jbrjake] right but i thought hb used 100ms because it's, like, half what a person could practically notice
[01:16] [eddyg] Unless you are anal about your sync, a gave up tuning my sync delay on the amp.
[01:17] [jbrjake] hehe
[01:17] [eddyg] I'm not sure, I'd have thought it would be due to sample size
[01:20] [eddyg] Guess what I have.
[01:20] [eddyg] A HB produced file with both AC3 and AAC
[01:20] [jbrjake] got it working?
[01:20] [jbrjake] awesome
[01:20] [eddyg] Not sure yet - haven't played it on the ATV.
[01:21] [s55] you have to be paying some serious attention to the movie to notice >200ms
[01:23] [jbrjake] 0.9.3 is going to be awesome, i can't wait to encode main audio in ac3 and commentaries in aac.
[01:24] [eddyg] By golly - it works
[01:25] [eddyg] I'm using a mixdown as the selection method.
[01:25] [eddyg] Which means you lose an audio track
[01:25] [eddyg] in the GUI.
[01:25] [eddyg] But that's easily changed within the GUI to select two tracks.
[01:25] [eddyg] But personally I like it as-is.
[01:25] [eddyg] I can select AC3 only
[01:25] [eddyg] AAC
[01:25] [eddyg] or both
[01:25] [eddyg] in either order
[01:26] [eddyg] Maybe we need to add extra audio tracks to the GUI.
[01:26] *** s55 glares at eddyg
[01:26] [jbrjake] not happening for 0.9.2
[01:26] [eddyg] But for now it works, which is the important thing
[01:26] [jbrjake] dynaflash and i have talked about it, major audio track upheaval waits for 0.9.3
[01:27] [eddyg] I'll send out the diffs.
[01:27] [eddyg] No upheaval required in this commit.
[01:27] [s55] The windows GUI only JUST got 2 audio tracks
[01:27] [s55] although, It only has the mixdown option on 1
[01:27] [jbrjake] god don't even talk about the cli
[01:27] [eddyg] Same for the Mac GUI
[01:27] [jbrjake] it will be hell to do this right for the cli
[01:28] [jbrjake] getopt hell
[01:28] [eddyg] this was the easiest way to get it working.
[01:28] [jbrjake] and probably custom string parsing, which imo is the task c is least suited for :/
[01:28] [eddyg] Now that this is done we can look at the GUI and whether we really need more audio streams when one is AC3.
[01:29] [eddyg] So with this diff, when you select AAC on one and AC3 on the other you are then out of audio tracks
[01:29] [jbrjake] that won't fly
[01:29] [eddyg] But it is easy within the GUI to frig this and generate two actual tracks from the one selected in the GUI.
[01:30] [jbrjake] dynaflash last night in pm: "as well, I dont like disabling that extra track for atv encodes since currently the macgui only has two"
[01:30] [eddyg] Do we want people to be able to select AC3 only, without an accompanying AAC
[01:30] [jbrjake] yes
[01:31] [jbrjake] but there needs to be an easy way for the presets to make sure the appletv has both and in the correct order (aac first)
[01:31] [eddyg] mmm.. thats two mixdowns then, AC3, AC3+DPLII
[01:31] [eddyg] Two new ones that is.
[01:31] [eddyg] Actually the AAC is second
[01:32] [eddyg] Sorry
[01:32] [eddyg] AAC is ffirst
[01:32] [eddyg] But it probably doesn't make any difference.
[01:32] [eddyg] Let me push this change out - it doesn't do any damage - just isn't the be all and end all.
[01:32] [jbrjake] what about what dynaflash and i suggested yesterday? when you're in H.264/AC3 mode you only have one "mixdown" popup item: ac3 pass-thru. add another for ac3+dpl2.
[01:33] [eddyg] Try what I have here, and see what you think then.
[01:33] s55 (n=Scott@62-31-67-171.cable.ubr01.livi.blueyonder.co.uk) left irc: "Leaving"
[01:33] [eddyg] In the meantime I want to try a couple of things.
[01:33] [jbrjake] k
[01:34] [eddyg] http://pastebin.ca/905764
[01:35] [eddyg] note that I've left the track disable in, ATV ignored it anyway
[01:36] [jbrjake] i believe it's necessary
[01:37] *** jbrjake can't test right now, trying out ac3-in-m4v and vfr on an hdtv encode.
[01:43] [eddyg] Give me 20mins and I'll have a variation on that.
[01:43] [jbrjake] ok, yeah, now i see what you mean about it being easy to simplify the gui presentation for 0.9.2
[01:50] cdaiger|laptop (n=cdaiger@c-68-80-83-55.hsd1.de.comcast.net) joined #handbrake-dev.
[01:50] #handbrake-dev: mode change '+v cdaiger|laptop' by BrakeBot!n=BrakeBot@rocky.lapsus.org
[02:12] [eddyg] OK done - I need to commit this change as I'm being hassled to spend time with the family (It's Saturday). So here it comes.. and it's pretty good.. I think you'll be happy.
[02:13] [CIA-1] eddyg * r1267 /trunk/ (7 files in 2 dirs): Allow mac gui to select AC-3 + AAC at the same time, still needs CLI work for the same.
[02:14] [eddyg] I haven't checked what the ATV does with 4 audio tracks, 2 AAC and 2 AC3.
[02:17] [hbrbot] ::hb_trac:: Changeset [1267]: Allow mac gui to select AC-3 + AAC at the same time, still needs CLI work & @ http://trac.handbrake.fr/changeset/1267
[02:19] [eddyg] Should all work fine with the Queue as well.
[02:20] [jbrjake] awesome
[02:20] [eddyg] As for the CLI, it will require the extra mix-down options, and the track insertion
[02:20] [eddyg] as per changes in Controller.mm
[02:28] *** jbrjake will wire up the cli as soon as his encode finishes
[02:42] [eddyg] Note that on the Mac GUI the AC-3 option may be better implemented as a tick box next to the mix-down (allowing AC-3 + Mono for example). However what we have here will work for 99% of folks.
[02:45] alcuin (n=alcuin@p50886F3F.dip.t-dialin.net) left #handbrake-dev.
[03:06] dynaflash (n=dynaflas@c-24-118-100-117.hsd1.mn.comcast.net) joined #handbrake-dev.
[03:07] #handbrake-dev: mode change '+o dynaflash' by HandBot!n=HandBot@213.246.52.12
[03:20] [dynaflash] Hot damn eddyg :)
[03:21] [NightStorm] anoyne tried a ac3+aac file on iphone yet? :)
[03:22] [dynaflash] wow, this is the shit
[03:24] [dynaflash] gonna be a long night once the kids go to bed ;)
[03:26] [NightStorm] lol
[03:26] [NightStorm] im compiling now
[03:27] [dynaflash] eddyg *doesnt* suck :P
[03:30] Cavalicious (n=Cavalici@adsl-70-254-91-105.dsl.rcsntx.swbell.net) joined #handbrake-dev.
[03:31] #handbrake-dev: mode change '+v Cavalicious' by jbrjake!n=jbrjake@214-243.187-72.tampabay.res.rr.com
[03:37] [Cavalicious] thanks
[04:01] [NightStorm] good news everyone, files with ac3+aac are playable on the iphone!
[04:06] NightStorm (n=NightSto@cblmdm72-241-69-59.buckeyecom.net) left irc: Read error: 104 (Connection reset by peer)
[04:06] NightStorm (n=NightSto@cblmdm72-241-69-59.buckeyecom.net) joined #handbrake-dev.
[04:07] [jbrjake] NightStorm: no way, that kicks ass
[04:07] [jbrjake] this is going to be huge.
[04:07] [jbrjake] i wonder if they play on regular ipods?
[04:09] [Cavalicious] hell...forgot to check the trac. I damn glad I ony encoded 2 movies so far
[04:19] [saintdev] jbrjake: so qt dies with all sources that have >= 13 bframes and/or b-pyramid?
[04:20] #handbrake-dev: mode change '+v NightStorm' by jbrjake!n=jbrjake@214-243.187-72.tampabay.res.rr.com
[04:20] [NightStorm] also dynaflash the dpl2+ac3 selection does not apply through a preset
[04:20] [NightStorm] ie. it resets to dpl2
[04:21] [NightStorm] thanks jbrjake
[04:22] [jbrjake] saintdev: huh? no... qt seems to be struggling to the point of dying with >= 3 bframes with b-pyramid. the >13 thing i was talking about earlier today was just a render queue not getting flushed.
[04:22] [saintdev] oh ok
[04:26] [NightStorm] has the ac3 code been tested when it is less than 5.1
[04:26] [NightStorm] ?
[04:27] [saintdev] could someone test http://saintdevelopment.com/files/test2.new-tc.mp4 in quicktime
[04:29] [NightStorm] very jittery
[04:29] [NightStorm] audio seems fine
[04:30] [saintdev] yeah video is what i'm testing
[04:30] [NightStorm] using qt w/o perian
[04:31] [saintdev] that's what i needed, thanks
[04:31] [NightStorm] np
[04:31] [NightStorm] kinda trippy to watch
[04:31] [NightStorm] lol
[04:32] [jbrjake] NightStorm: i've tested with stereo ac3
[04:33] [NightStorm] ok just noticed a disc i queued up was 5.0
[04:41] NightStorm (n=NightSto@cblmdm72-241-69-59.buckeyecom.net) left irc: Read error: 104 (Connection reset by peer)
[04:42] NightStorm (n=NightSto@cblmdm72-241-69-59.buckeyecom.net) joined #handbrake-dev.
[04:56] [saintdev] :O we're generating negative durations in muxmp4, which don't quite fit in an unsigned, and thus result in very large durations
[05:00] [jbrjake] err where?
[05:00] [jbrjake] van was the last to change that
[05:00] [jbrjake] he thought he'd covered the longest
[05:01] [jbrjake] maybe he did too well ;
[05:02] [saintdev] when duration is calculated.
[05:02] Paul__ (n=chatzill@70-59-113-206.mpls.qwest.net) left irc: Client Quit
[05:02] [saintdev] some of the time bias comes out to be negative (about 2 long) making the result negative.
[05:04] [jbrjake] http://trac.handbrake.fr/changeset/1224 ?
[05:05] [saintdev] yep that's the one
[05:05] [saintdev] doesn't really seem to be affecting most players, though
[05:10] [saintdev] http://pastebin.ca/905947 "orig_duration" is the way things are currently done.
[05:19] [saintdev] d'oh forgot I changed the output timecodes from x264 in that build
[05:19] [saintdev] *ignore the above*
[05:22] [saintdev] i do think it's still a possibility with unmodified timecodes, however.
[05:26] [eddyg] dynaflash: I wasn't sure what to do with the bitrates in the GUI when you have mixed AAC/AC-3. I think I got it right - i.e. leave them on, but ignore those values and use the actual AC-3 bitrate when calculating the overall audio bitrate. But I couldn't guarantee it.
[05:27] [eddyg] The other worry is the sample rate, I guess that must be kept at 48
[05:27] [eddyg] if using AC-3
[05:28] [eddyg] travistex: should also check the queue operation - where the additional audio tracks are added in Controller.mm may not suite the queue.
[05:33] [eddyg] jbrjake: That vfr change in http://pastebin.ca/905764 looks safe, basically bypassing the delay queue if vfr is not on (but it pushing and then popping immediately.
[05:40] [eddyg] FYI. MetaX tags the AAC/AC3 Movies fine, as expected.
[06:35] *** Cavalicious wonders how he is going to redo LOTR EE combined with 2 tracks. Surely going to bust the QT/ATV 4gb limit
[07:09] #handbrake-dev: mode change '+v NightStorm' by jbrjake!n=jbrjake@214-243.187-72.tampabay.res.rr.com
[08:36] BradleyS (n=BradleyS@cpe-204-210-180-47.neo.res.rr.com) joined #handbrake-dev.
[08:36] #handbrake-dev: mode change '+v BradleyS' by HandBot!n=HandBot@213.246.52.12
[08:58] [dynaflash] NightStorm: k. will fix the presets, dont worry
[08:59] [dynaflash] eddyg: ten four on the bitrates. Its the same way six channel and dpl2 are handles currently . uses the higher bitrate.
[09:00] [dynaflash] Also, heard somewhere that > 4 gb files now work on atv. Someone should give it a shot
[09:41] BradleyS (n=BradleyS@cpe-204-210-180-47.neo.res.rr.com) left irc:
[09:53] ritsuka (n=Ritsuka@87.19.242.176) joined #handbrake-dev.
[09:53] #handbrake-dev: mode change '+v ritsuka' by BrakeBot!n=BrakeBot@rocky.lapsus.org
[10:49] ritsuka (n=Ritsuka@87.19.242.176) left irc:
[10:58] ritsuka (n=Ritsuka@87.19.242.176) joined #handbrake-dev.
[10:58] #handbrake-dev: mode change '+v ritsuka' by HandBot!n=HandBot@213.246.52.12
[13:07] jbrjake_ (n=jbrjake@214-243.187-72.tampabay.res.rr.com) joined #handbrake-dev.
[13:08] #handbrake-dev: mode change '+o jbrjake_' by BrakeBot!n=BrakeBot@rocky.lapsus.org
[13:16] jbrjake (n=jbrjake@214-243.187-72.tampabay.res.rr.com) left irc: Read error: 110 (Connection timed out)
[13:48] sr55 (n=Scott@62-31-67-171.cable.ubr01.livi.blueyonder.co.uk) joined #handbrake-dev.
[13:48] #handbrake-dev: mode change '+o sr55' by HandBot!n=HandBot@213.246.52.12
back next →
1 2 3 4
up
Generated by logs2html module for eggdrop v.2.3.4
Find latest version at http://sourceforge.net/projects/logs2html or http://shmupsik.osetia.org