| *** drobilla has quit IRC | 00:06 | |
| *** Magnus_RM has quit IRC | 00:06 | |
| *** falktx has joined #lv2 | 01:27 | |
| *** uncle-j_j has quit IRC | 01:46 | |
| *** falktx has quit IRC | 02:20 | |
| *** drobilla has joined #lv2 | 03:42 | |
| *** LAbot` has joined #lv2 | 04:55 | |
| *** aombk3 has quit IRC | 05:02 | |
| *** LAbot has quit IRC | 05:02 | |
| *** coderkevin has quit IRC | 05:03 | |
| *** coderkevin has joined #lv2 | 05:14 | |
| *** edogawa has joined #lv2 | 08:36 | |
| *** uncle-j_j has joined #lv2 | 08:57 | |
| *** artfwo has joined #lv2 | 10:39 | |
| *** falktx has joined #lv2 | 11:13 | |
| *** edogawa has quit IRC | 11:44 | |
| *** rncbc has joined #lv2 | 12:13 | |
| *** sigma6 has joined #lv2 | 12:44 | |
| *** aombk3 has joined #lv2 | 12:59 | |
| *** aombk2 has quit IRC | 13:02 | |
| *** falktx has quit IRC | 13:06 | |
| *** falktx has joined #lv2 | 13:52 | |
| falktx | rgareus: setbfree doesn't kill pending notes on reactivation | 14:36 |
|---|---|---|
| falktx | rgareus: I just got some stuck notes. tried deactivate and activate. the note is still being played :( | 14:36 |
| falktx | or does ingen not call deactivate/activate when the plugin is reenabled? | 14:36 |
| rgareus | send a panic | 14:37 |
| rgareus | setBfree ignores activate/deactivate | 14:38 |
| rgareus | JACK does not support an activate/deactivate mechanism. and the core or setBfree is a jack-app (with a small LV2 wrapper) | 14:39 |
| falktx | errr | 14:39 |
| falktx | can't you just add a check for deactivate? it would be pretty easy | 14:40 |
| falktx | just because it's a jack app in its core doesn't mean the lv2 version has to follow strictly | 14:40 |
| falktx | you're making the lv2 version less useful | 14:40 |
| rgareus | activate/deactivate are optional are they not? | 14:40 |
| falktx | yes | 14:40 |
| falktx | not needed on all cases | 14:41 |
| falktx | setbfree is a good candidate for activate (all synths are) | 14:41 |
| rgareus | send an all-notes-off (CC 123) or all-sound-off (CC 120) | 14:41 |
| rgareus | no need to invent another standard | 14:41 |
| rgareus | there's already a spec for this that all synths follow | 14:41 |
| falktx | tf | 14:41 |
| rgareus | or should follow if they're midi synthhs | 14:41 |
| falktx | why are we going backwards again? | 14:41 |
| falktx | they're not midi synths! | 14:42 |
| falktx | they are plugins | 14:42 |
| falktx | they use use midi for notes because there's no better spec yet | 14:42 |
| rgareus | well, setBfree is first and foremost a MIDI synth. | 14:42 |
| falktx | which limits it in a very strong way | 14:42 |
| rgareus | it can be a plugin, it can be a jack app, it can be file-renderer.. | 14:43 |
| falktx | plugins should reset internal state when de/activate is called | 14:43 |
| rgareus | I can add a deactivate -> send panic to internal synth | 14:44 |
| falktx | it's the activate one | 14:45 |
| falktx | "the plugin instance MUST reset all state information dependent on the history of the plugin instance" | 14:45 |
| falktx | from http://lv2plug.in/doc/html/group__lv2core.html#ad34d59efaed8faaf8a05a4c375f62415 | 14:45 |
| rgareus | "except for any data locations provided by connect_port()" | 14:46 |
| rgareus | does that include MIDI? | 14:46 |
| falktx | data locations | 14:46 |
| falktx | is midi a pointer? | 14:46 |
| rgareus | depende whom you ask. in LV2 yes. it's a pointer to some [Atom] buffers | 14:47 |
| rgareus | it's not POD | 14:47 |
| rgareus | but it also affects internal state | 14:47 |
| rgareus | setBfree also sends its current state to external controllers. | 14:48 |
| rgareus | anyway, generating a "all notes off" on activate is easy. | 14:48 |
| falktx | I don't know why it always requires a long discussion for something so simple... :/ | 14:49 |
| rgareus | it won't be effective during activate() but the first run() after activate.. | 14:49 |
| falktx | of course | 14:50 |
| rgareus | falktx: because subtle changes can break things in deep ways | 14:50 |
| falktx | some things are rather obvious in nature | 14:50 |
| falktx | no need to question everything | 14:50 |
| falktx | rgareus: convo.lv2 performance should be good right? | 14:56 |
| rgareus | falktx: to me MIDI-panic is a lot more obvious for a MIDI-synth. I don't think of setBfree as a plugin. | 14:57 |
| rgareus | falktx: convo.lv2 performance depdens a lot on the IR length | 14:57 |
| rgareus | falktx: it uses sync (non-threaded) zita-convolver. and for long (think 5+sec) IRs zita-convolver performs best when threaded and adds latency. | 14:58 |
| falktx | well, time for a test | 14:59 |
| *** deva has joined #lv2 | 15:04 | |
| falktx | the hell, I build it with static fftw but it's complaining about missing fftw symbols :( | 15:04 |
| falktx | this is why I like -Wl,--no-undefined | 15:05 |
| falktx | ah, problem with zita being static | 15:08 |
| rgareus | convol.lv2 ? I don't think I've added that warning there. | 15:10 |
| falktx | you didn't. I did | 15:10 |
| falktx | just now to test what it was missing | 15:11 |
| falktx | damn linux doesn't check if all symbols are available by default | 15:11 |
| rgareus | indeed it's zita-convolver using fftw (not convo.lv2 directly) | 15:12 |
| rgareus | anyway setBfree activate call pushed to git | 15:12 |
| rgareus | falktx: in different news. ardour-nightlies since today come with fftw-3.3.5 (git) and call fftwf_make_planner_thread_safe() | 15:13 |
| falktx | thanks | 15:16 |
| falktx | sadly I can't test convo.lv2 on MOD right now | 15:16 |
| falktx | ingen doesn't seem to support file loading ? | 15:16 |
| falktx | at least not when connecting remotely | 15:17 |
| falktx | but I even copied the same files to /tmp to make sure it matched on both systems... :( | 15:17 |
| rgareus | state files? | 15:26 |
| falktx | I need to try if presets can work for now | 15:26 |
| *** edogawa has joined #lv2 | 17:43 | |
| *** ricardocrudo has joined #lv2 | 18:03 | |
| *** drobilla has quit IRC | 18:35 | |
| *** drobilla has joined #lv2 | 19:00 | |
| *** falktx has quit IRC | 19:05 | |
| *** deva has quit IRC | 19:08 | |
| *** ricardocrudo has quit IRC | 19:44 | |
| *** falktx has joined #lv2 | 19:55 | |
| *** drobilla has joined #lv2 | 20:12 | |
| *** Magnus_RM has joined #lv2 | 21:10 | |
| *** rncbc has quit IRC | 21:16 | |
| *** uncle-j_j has quit IRC | 22:40 | |
| *** edogawa has quit IRC | 22:42 | |
| *** falktx has quit IRC | 23:47 | |
Generated by irclog2html.py 2.13.0 by Marius Gedminas - find it at mg.pov.lt!