*** falktx has quit IRC | 00:24 | |
*** bgribble has joined #lv2 | 00:59 | |
*** bgribble has quit IRC | 01:27 | |
*** Yoda-- has quit IRC | 01:33 | |
*** Yoda-- has joined #lv2 | 01:33 | |
*** bgribble has joined #lv2 | 01:40 | |
*** ricardocrudo has quit IRC | 01:42 | |
*** Yoda-- has quit IRC | 01:42 | |
*** Yoda-- has joined #lv2 | 01:42 | |
*** ricardocrudo has joined #lv2 | 01:47 | |
*** bgribble has quit IRC | 02:01 | |
*** unclechu has quit IRC | 03:40 | |
*** edogawa has joined #lv2 | 05:22 | |
*** mlpug has joined #lv2 | 07:27 | |
*** mlpug has quit IRC | 07:45 | |
*** falktx has joined #lv2 | 08:11 | |
*** ddom has joined #lv2 | 08:46 | |
*** falktx has quit IRC | 08:48 | |
*** falktx has joined #lv2 | 10:10 | |
*** rncbc_jolla has joined #lv2 | 11:42 | |
*** rncbc_jolla has quit IRC | 13:14 | |
*** NickSB2 has quit IRC | 13:44 | |
*** ddom has quit IRC | 15:51 | |
wrl | question, are CV ports the only way of the host changing a port's value inside of a process() call? | 15:57 |
---|---|---|
wrl | there's no, say, event type for doing that for normal control ports | 15:57 |
wrl | ? | 15:57 |
*** NickSB2 has joined #lv2 | 16:01 | |
falktx | modifying input ports, no | 16:11 |
*** falktx has quit IRC | 16:11 | |
drobilla | wrl: lv2:ControlPort is just a float, it can't possibly contain an event | 16:12 |
drobilla | wrl: You could do this via events in a sequence port, but we don't have a standard event for it yet | 16:13 |
*** falktx has joined #lv2 | 16:15 | |
*** edogawa_ has joined #lv2 | 16:48 | |
*** edogawa has quit IRC | 16:51 | |
*** edogawa_ is now known as edogawa | 16:54 | |
wrl | drobilla: gotcha. | 18:28 |
drobilla | Well, we kinda sorta do. You could use the path extension stuff ala eg-sampler, but hosts don't support it anyway. | 18:32 |
drobilla | This may be a bit too clunky for numeric controls, and easy wrappers needed regardless, etc. | 18:33 |
*** mlpug has joined #lv2 | 18:33 | |
*** rncbc has joined #lv2 | 18:35 | |
*** Joeboy has quit IRC | 18:56 | |
*** falktx has quit IRC | 18:58 | |
*** falktx_ has joined #lv2 | 18:58 | |
*** Joeboy has joined #lv2 | 18:58 | |
wrl | drobilla: it's something that vst2.4 devs have complained about very often | 19:02 |
wrl | what hosts will do in lieu of being able to update parameters intra-process() (to keep automation in sync) is subdivide the block and set the params between calls | 19:03 |
wrl | i.e. blocksize of 1024, automation point halfway through the block, so call process() with 512 frames, set the params, then another 512 frame process() | 19:03 |
falktx_ | carla does that :) | 19:06 |
wrl | most of the good hosts do at this point tbh | 19:07 |
*** zth has joined #lv2 | 19:26 | |
drobilla | wrl: Yes, that's the alternative. | 19:31 |
*** Anchakor_ has quit IRC | 19:31 | |
drobilla | wrl: Single float control ports are utter shit | 19:31 |
drobilla | But it's what we inherited, for better or worse | 19:31 |
drobilla | Ingen recently actually implements control buffers internally as sequence buffers of float, and just connects ControlPorts and splits cycles appropriately | 19:32 |
drobilla | So you can (and I do) have plugins that actually take a sequence of float natively and can do sample accurate processing of them | 19:32 |
drobilla | Different than having multiple dispatch of a sort by putting all the control through a single port, but that's another way | 19:33 |
*** Anchakor_ has joined #lv2 | 19:50 | |
*** falktx__ has joined #lv2 | 20:06 | |
*** falktx_ has quit IRC | 20:09 | |
*** mlpug has quit IRC | 20:20 | |
*** zth has quit IRC | 20:20 | |
*** falktx__ is now known as falktx | 20:21 | |
*** HarryHaaren has joined #lv2 | 21:16 | |
*** edogawa has quit IRC | 21:32 | |
*** gian_ has joined #lv2 | 21:37 | |
*** HarryHaaren has quit IRC | 21:55 | |
*** HarryHaaren has joined #lv2 | 22:00 | |
*** gian_ has quit IRC | 22:12 | |
*** gian_ has joined #lv2 | 22:14 | |
*** rncbc has quit IRC | 22:32 | |
*** gian_ is now known as gianMOD | 22:32 | |
*** wumpus has quit IRC | 23:09 | |
*** HarryHaaren has quit IRC | 23:13 |
Generated by irclog2html.py 2.13.0 by Marius Gedminas - find it at mg.pov.lt!