| *** gianMOD has joined #lv2 | 00:25 | |
| *** gianMOD has quit IRC | 00:29 | |
| *** HarryHaaren has quit IRC | 00:39 | |
| *** gianMOD has joined #lv2 | 01:26 | |
| * drobilla submits halfway respectable Ingen paper | 01:29 | |
| drobilla | Right. NEXT. | 01:29 |
|---|---|---|
| *** gianMOD has quit IRC | 01:32 | |
| *** gianMOD has joined #lv2 | 02:28 | |
| *** gianMOD has quit IRC | 02:33 | |
| drobilla | Here's a possibly crazy idea Re: the portable menus and file selector stuff: how about shelling out to a separate process? | 03:21 |
| drobilla | Voila, no more Gtk linking issues | 03:21 |
| drobilla | Zenity, basically, though I don't think it can do context menus. Doing hierarchical menus would require a fancier protocol | 03:23 |
| drobilla | and can't deal with system menu on OSX anyway, so I guess that's a non-starter | 03:23 |
| *** gianMOD has joined #lv2 | 03:29 | |
| *** gianMOD has quit IRC | 03:34 | |
| *** Magnus_RM has quit IRC | 03:41 | |
| *** ddom has quit IRC | 03:50 | |
| *** gianMOD has joined #lv2 | 04:30 | |
| *** gianMOD has quit IRC | 04:34 | |
| *** ricardocrudo_ has joined #lv2 | 04:59 | |
| *** drobilla has quit IRC | 05:01 | |
| *** ricardocrudo has quit IRC | 05:02 | |
| *** gianMOD has joined #lv2 | 05:30 | |
| *** gianMOD has quit IRC | 05:36 | |
| *** gianMOD has joined #lv2 | 06:33 | |
| *** burayon_ has joined #lv2 | 06:33 | |
| *** gianMOD has quit IRC | 06:38 | |
| *** burayon_ has quit IRC | 07:11 | |
| *** burayon_ has joined #lv2 | 07:19 | |
| *** gianMOD has joined #lv2 | 07:34 | |
| *** gianMOD has quit IRC | 07:38 | |
| *** drobilla has joined #lv2 | 08:05 | |
| *** burayon_ has quit IRC | 08:24 | |
| *** gianMOD has joined #lv2 | 08:34 | |
| *** gianMOD has quit IRC | 08:39 | |
| *** gianMOD has joined #lv2 | 09:24 | |
| *** ddom has joined #lv2 | 09:30 | |
| *** gianMOD has quit IRC | 09:34 | |
| *** falktx has joined #lv2 | 10:22 | |
| *** gianMOD has joined #lv2 | 10:35 | |
| *** gianMOD has quit IRC | 10:40 | |
| *** gianMOD has joined #lv2 | 10:40 | |
| *** gianMOD has quit IRC | 10:40 | |
| *** gianMOD has joined #lv2 | 10:41 | |
| *** deva has joined #lv2 | 12:01 | |
| *** HarryHaaren has joined #lv2 | 12:25 | |
| *** ricardocrudo_ has quit IRC | 12:35 | |
| deva | I'm trying to implement latency reporting in my plugin, so that the host can compensate. I so far managed to get the reporting hooked up, but the host doesn't seem to use the value for anything but showing it in the plugin window as a slider... Do I need to do anything special in order to have the host use the reported value for the compensation? | 12:41 |
| HarryHaaren | deva, you're marking the port with http://lv2plug.in/ns/lv2core/#reportsLatency? | 12:43 |
| *** ricardocrudo_ has joined #lv2 | 12:43 | |
| HarryHaaren | after that its up to the host to use the value IIUC - what host are you testing with? | 12:43 |
| deva | I used exactly that port type. I'm testing with Ardour | 12:44 |
| HarryHaaren | hmm ok. | 12:44 |
| HarryHaaren | deva, why does DG introduce latency? | 12:44 |
| deva | http://pastebin.com/MJa5xyg7 | 12:45 |
| deva | DG is about to become untight .... and in order to do that,. it needs to be able to play midi notes /before/ they are triggered | 12:45 |
| HarryHaaren | could add some props: lv2:portProperty lv2:reportsLatency, lv2:integer, pprops:notOnGUI; | 12:46 |
| deva | So I'm trying to add latency of say 200samples, and then position the sample itself at 0 offset in order to be 200 samples early, or offset 400 in order to be 200 samples late | 12:46 |
| deva | What does the lv2:integer do? the port data itself is a float... | 12:47 |
| HarryHaaren | sure - i hope there's gonna be an "off" switch for this untight-ness? Live usage would be better if there's no latency | 12:47 |
| badosu | deva: basically you say that your scales points are mostly integers | 12:47 |
| HarryHaaren | lv2:integer just says the data should be used as int | 12:47 |
| deva | ok | 12:47 |
| deva | yes, there will be an "off" button | 12:47 |
| HarryHaaren | awesome ;) | 12:47 |
| badosu | deva: same thing as lv2:toggle, you say that it is 0.0 or 1.0 | 12:47 |
| badosu | deva: see lv2:toggle for boolean | 12:48 |
| HarryHaaren | deva, i haven't used reportsLatency (since Live is my gig, i don't want any :) .. falktx has some experience, and I know he's worked on Carla support for latency too... perhaps he has an idea. | 12:48 |
| *** ricardocrudo_ has quit IRC | 12:48 | |
| deva | How do the input/output port callbacks line up? | 12:49 |
| falktx | deva: there's 2 ways of showing latency in lv2. the old way and new way | 12:49 |
| falktx | the best thing is to use both, ensuring it works on all hosts | 12:49 |
| deva | falktx, I have so far only found one way (the one described by Harry above) | 12:50 |
| deva | Is that the new or the old way? | 12:50 |
| falktx | for latency, you need: | 12:50 |
| falktx | a lv2:OutputPort, lv2:ControlPort ; | 12:50 |
| falktx | lv2:designation < LV2_CORE__latency > ; | 12:51 |
| falktx | lv2:portProperty lv2:reportsLatency, lv2:integer ; | 12:51 |
| falktx | replace LV2_CORE__latency for its actual string value | 12:51 |
| falktx | pprops:notOnGUI should not be needed | 12:52 |
| deva | is the "designation" the new way? | 12:53 |
| falktx | yes | 12:54 |
| deva | So, like this: http://pastebin.com/EQi5Xujk ? | 12:55 |
| falktx | lv2:portProperty looks wrong | 12:55 |
| falktx | the first "p" is lowercase | 12:55 |
| deva | ooh... ok | 12:55 |
| falktx | maybe it's a good idea to run sord-validate on those ttl files | 12:56 |
| falktx | deva: http://lv2plug.in/pages/validating-lv2-data.html | 12:57 |
| deva | falktx, I think I'll add that to my unit test suite ;) | 12:58 |
| deva | Is the ports in "sync" when run is called? or do the input ports recieve data that are "letency" samples before the output samples? | 13:00 |
| HarryHaaren | deva, ah that's what you mean by "callbacks line up". yes they're in sync, same as a normal run() without any latency | 13:02 |
| HarryHaaren | the idea is that the host adds a delay line to all other "tracks" of processing, equalling out total delay in samples. | 13:03 |
| HarryHaaren | so if I report a plugin with 250 samps latency, and you have 150 for DG, you output at 150 as normal, and the host should add a 100 sample delay to DG in order for the signals to be in sync again | 13:04 |
| HarryHaaren | deva, make sense? | 13:04 |
| deva | yes, that makes sense... I just can't seem to get it to work like that... | 13:04 |
| HarryHaaren | you mean that Ardour won't work like that? | 13:05 |
| HarryHaaren | (ie, what do you mean by "it" :) | 13:05 |
| deva | In my engine, when I receive a midi note, i start playing a sample | 13:05 |
| falktx | very few hosts implement latency | 13:06 |
| deva | if I have latency, I place the offset of that sample at current time + latency | 13:06 |
| deva | If Ardour compensates for the reported latency, it /shou;d/ result in my sample being played in sync with the rest of the tracks in my project (guitar etc...) | 13:06 |
| deva | But it doesn't | 13:07 |
| HarryHaaren | deva, right ok. It sounds like time to hang out in #ardour, and ask if latency compensation is expected to work correctly in your version of Ardour.. and if so, test with another plugin, and compare TTL / details of implementation. | 13:09 |
| HarryHaaren | again, sorry, i have no experience in coding this stuff - i know some of the x42 plugins have latency compensation, which I assume works with Ardour given Robin works on them both :) | 13:10 |
| deva | Yeah, I guess #lv2 is not the place for this particular question ;) | 13:10 |
| HarryHaaren | well yes and no :) | 13:10 |
| *** gianMOD has quit IRC | 13:14 | |
| *** HarryHaaren has quit IRC | 13:18 | |
| *** HarryHaaren has joined #lv2 | 13:18 | |
| deva | Aah... it seems I accidentalle positioned the sample using the offset value twice in my own code... now everything works like a charm. | 13:28 |
| deva | Thanks for your helps guys | 13:28 |
| HarryHaaren | y/w, glad ya have it working! | 13:34 |
| *** gianMOD has joined #lv2 | 14:28 | |
| *** NickSB2 has quit IRC | 14:29 | |
| *** gianMOD has quit IRC | 15:15 | |
| *** gianMOD has joined #lv2 | 15:20 | |
| *** gianMOD has quit IRC | 15:36 | |
| *** gianMOD_ has joined #lv2 | 15:38 | |
| *** gianMOD_ has quit IRC | 15:38 | |
| *** HarryHaaren has quit IRC | 16:20 | |
| *** NickSB2 has joined #lv2 | 17:14 | |
| *** ricardocrudo has joined #lv2 | 18:27 | |
| *** ricardocrudo_ has joined #lv2 | 18:33 | |
| *** ricardocrudo__ has joined #lv2 | 18:36 | |
| *** ricardocrudo has quit IRC | 18:36 | |
| *** ricardocrudo_ has quit IRC | 18:38 | |
| *** ricardocrudo_ has joined #lv2 | 18:38 | |
| *** ricardocrudo__ has quit IRC | 18:41 | |
| *** ricardocrudo_ has quit IRC | 18:43 | |
| *** gianMOD has joined #lv2 | 18:46 | |
| *** gianMOD has quit IRC | 18:54 | |
| *** ricardocrudo has joined #lv2 | 18:59 | |
| *** ricardocrudo has quit IRC | 19:30 | |
| *** ricardocrudo has joined #lv2 | 19:41 | |
| *** deva has quit IRC | 19:41 | |
| *** ricardocrudo_ has joined #lv2 | 20:06 | |
| *** ricardocrudo__ has joined #lv2 | 20:09 | |
| *** ricardocrudo has quit IRC | 20:10 | |
| *** ricardocrudo_ has quit IRC | 20:11 | |
| *** gianMOD has joined #lv2 | 20:26 | |
| *** rncbc has joined #lv2 | 20:50 | |
| *** HarryHaaren has joined #lv2 | 21:06 | |
| *** falktx has quit IRC | 21:57 | |
| *** gianMOD has quit IRC | 22:07 | |
| *** rncbc has quit IRC | 22:36 | |
| *** edogawa has joined #lv2 | 22:45 | |
| *** ricardocrudo__ has quit IRC | 22:46 | |
| *** ricardocrudo__ has joined #lv2 | 22:58 | |
| *** gianMOD has joined #lv2 | 23:07 | |
| *** HarryHaaren has quit IRC | 23:10 | |
| *** gianMOD has quit IRC | 23:12 | |
| *** edogawa has quit IRC | 23:12 | |
| *** ricardocrudo__ has quit IRC | 23:13 | |
| *** ricardocrudo has joined #lv2 | 23:18 | |
| *** ricardocrudo_ has joined #lv2 | 23:21 | |
| *** ricardocrudo has quit IRC | 23:23 | |
| *** gianMOD has joined #lv2 | 23:37 | |
| *** gianMOD has quit IRC | 23:42 | |
| *** ddom has quit IRC | 23:46 | |
Generated by irclog2html.py 2.13.0 by Marius Gedminas - find it at mg.pov.lt!