| *** ricardocrudo has quit IRC | 00:42 | |
| *** artfwo has quit IRC | 02:40 | |
| *** brummer_ has joined #lv2 | 03:14 | |
| *** brummer_ has left #lv2 | 03:18 | |
| *** ddom has quit IRC | 03:24 | |
| *** ddom has joined #lv2 | 03:25 | |
| *** ddom has quit IRC | 04:55 | |
| *** bgola has quit IRC | 06:24 | |
| *** edogawa has joined #lv2 | 07:51 | |
| *** edogawa has quit IRC | 07:56 | |
| *** edogawa has joined #lv2 | 07:58 | |
| *** ventosus has joined #lv2 | 08:05 | |
| *** falktx|work has joined #lv2 | 09:03 | |
| *** sigma6 has joined #lv2 | 10:03 | |
| *** falktx|work has quit IRC | 10:24 | |
| *** ricardocrudo has joined #lv2 | 11:29 | |
| *** drobilla has quit IRC | 12:34 | |
| *** artfwo has joined #lv2 | 12:44 | |
| *** ddom has joined #lv2 | 14:23 | |
| *** NickSB2 has quit IRC | 14:42 | |
| *** deva has joined #lv2 | 15:41 | |
| *** rncbc has joined #lv2 | 16:50 | |
| *** sigma6 has quit IRC | 17:37 | |
| *** ventosus has quit IRC | 17:50 | |
| ssj71 | falktx`: jsky zita at1 does have latency about 1024 samples (but it varies slightly during operation) | 18:34 |
|---|---|---|
| falktx` | I don't remember seeing at1 code calling latency callbacks on jacl | 18:41 |
| falktx` | *jack | 18:41 |
| ssj71 | I don't think it reports it, but its there | 18:42 |
| falktx` | bad zita | 18:42 |
| * ssj71 is very familiar with that code | 18:42 | |
| ssj71 | jack programs can report latency? | 18:42 |
| ssj71 | I believe in the initialization you see that the read and write pointers of the buffer are offset by 1024 (at 44.1 or 48k) | 18:43 |
| falktx` | of course jack programs report latency | 18:46 |
| falktx` | not only do they report it, they must compensate for it | 18:46 |
| ssj71 | they must compensate? how can a program like at1 compensate for latency? | 18:48 |
| ssj71 | that would be non-causal :) | 18:48 |
| ssj71 | I understand how a DAW does latency compensation, but an standalone audio processing program I don't see how thats possible | 18:49 |
| falktx` | see the jack latency api then | 18:52 |
| ssj71 | RTFM! :) | 18:53 |
| * ssj71 goes to RTFM | 18:53 | |
| ssj71 | ah! they compensate for latency earlier in the chain | 18:54 |
| ssj71 | thats complicated though. I can see why none of the jack programs I've reviewed use that api | 18:55 |
| ssj71 | easier just to ignore it | 18:55 |
| rgareus | ssj71: if a jack app has two input ports AND the processing depends on the data of both ports beeing aligned. the app in question is supposed to delay one of the ports if needed or cope with it one some way. | 18:56 |
| ssj71 | ya, makes sense | 18:56 |
| ssj71 | perhaps I've mostly reviewed mono programs | 18:57 |
| rgareus | otherwise just don't do anything. the default operation (if there's no custom jack_latency callback in the app) is to just pass on the latency downstream | 18:57 |
| rgareus | or upstream (in case of playback latency) | 18:57 |
| ssj71 | gotcha | 18:58 |
| ssj71 | thats cool | 18:58 |
| ssj71 | no wonder everything works so well | 18:58 |
| rgareus | fons acknowledged that some of his apps don't to this properly yet (after 3 glasses of wine at LAC'15) and said he'll look into it | 18:59 |
| rgareus | I recall he sent an email to LAD a few month ago asking to test zita-ajbridge's latency reporting | 18:59 |
| rgareus | zita-at may be in the queueu | 18:59 |
| ssj71 | I suppose he may handle it in a different module. I've mostly looked at the retuner file | 19:01 |
| rgareus | https://github.com/x42/robtk/blob/master/jackwrap.c#L622 .. 666 is a simple LV2-host jack-client that uses a delayline to re-align inputs (and reports the latency of those) | 19:03 |
| *** HarryHaaren has joined #lv2 | 19:05 | |
| rgareus | in short: calculate the worst-case latency of all ports. and the best-case (of all ports). then delay ports as needed to align them to worst-case. report back the difference per port back to jack. | 19:05 |
| rgareus | HarryHaaren: long time now see. How's things going? | 19:06 |
| HarryHaaren | hey rgareus! Yeah been pretty busy with work things - have holidays now for 2 weeks until the 4th of Jan, so getting ready for a coding spree :D | 19:06 |
| HarryHaaren | how are things you side? | 19:06 |
| * HarryHaaren also got a new keyboard, which makes lovely new clicky-key sounds, and light action | 19:07 | |
| ssj71 | HarryHaaren: which keyboard (I've been lurking in the market for a new one) | 19:07 |
| ssj71 | ? | 19:07 |
| HarryHaaren | ssj71, got a pretty expensive one with mechanical keys, the "non-clicky" ones apparently, but they're still pretty loud when I'm using them. A button-basher apparently. | 19:08 |
| rgareus | HarryHaaren: computer keyboard or music-keyboard? | 19:08 |
| HarryHaaren | Its a DasKeyboard 4C something | 19:08 |
| HarryHaaren | computer :) | 19:08 |
| HarryHaaren | yes, a 4C "professional" (not the ultimate one, with absolutely no marks / letters on it..) | 19:08 |
| HarryHaaren | its a smaller one, without the numpad. So i can keep the mouse right beside me now | 19:09 |
| ssj71 | ah keyboard, not a keyboard :) I'm looking for a new midi controller | 19:09 |
| HarryHaaren | ah ok, umm in that case, I dunno what size are you looking for? I had an E-MU Xboard 61 for a while, they're pretty nice | 19:10 |
| ssj71 | some of the guys here at work get fancy ones, I'm not too picky. | 19:10 |
| HarryHaaren | semi-weighted, so pretty synth action | 19:10 |
| ssj71 | fancy qwerty keyboards I mean | 19:10 |
| ssj71 | sorry switching topics here :) | 19:10 |
| HarryHaaren | heh yep, this is my guilty pleasure, loud clicky fast typing with minimal effort | 19:10 |
| *** artfwo has quit IRC | 19:11 | |
| *** artfwo has joined #lv2 | 19:11 | |
| ssj71 | I think I just don't know any better, I use the low profile keyboards and like em | 19:12 |
| rgareus | HarryHaaren: holidays from doing what? are you still in sysadmin/network admin? | 19:15 |
| HarryHaaren | rgareus, software programming now, working for a company in Shannon (~20 minute drive from Limerick) doing network data-plane software | 19:16 |
| *** HarryHaaren has quit IRC | 19:27 | |
| *** NickSB2 has joined #lv2 | 19:33 | |
| *** ventosus has joined #lv2 | 19:41 | |
| *** artfwo has quit IRC | 20:17 | |
| *** artfwo has joined #lv2 | 20:31 | |
| *** ricardocrudo has quit IRC | 20:53 | |
| *** drobilla has joined #lv2 | 21:15 | |
| *** ventosus has left #lv2 | 21:18 | |
| *** NickSB2 has quit IRC | 22:34 | |
| *** edogawa has quit IRC | 22:45 | |
| *** deva has quit IRC | 23:10 | |
| *** ricardocrudo has joined #lv2 | 23:19 | |
Generated by irclog2html.py 2.13.0 by Marius Gedminas - find it at mg.pov.lt!