| *** timbyr_ has quit IRC | 01:06 | |
| *** timbyr has joined #lv2 | 01:08 | |
| *** Gethiox has quit IRC | 01:27 | |
| *** Gethiox has joined #lv2 | 01:44 | |
| *** drobilla has quit IRC | 01:55 | |
| *** drobilla has joined #lv2 | 01:58 | |
| *** NickSB2 has joined #lv2 | 02:16 | |
| *** LAbot has joined #lv2 | 02:52 | |
| *** LAbot` has quit IRC | 02:53 | |
| drobilla | For real-time support, out- | 03:33 |
|---|---|---|
| drobilla | put buffer space is made available by the host before | 03:33 |
| drobilla | calling process(). The default behaviour is for | 03:33 |
| drobilla | output buffers to be the same size as input buffers of | 03:33 |
| drobilla | the same type, but plugins that require more space | 03:33 |
| drobilla | can request larger output buffers ahead of time. | 03:33 |
| drobilla | Anchakor: ^ | 03:33 |
| *** HarryHaaren has quit IRC | 03:38 | |
| drobilla | rgareus: http://drobilla.net/ns/sweetSpot :) | 03:58 |
| *** falktx has joined #lv2 | 04:04 | |
| *** shanipribadi has quit IRC | 04:13 | |
| *** shanipribadi has joined #lv2 | 04:16 | |
| *** shanipribadi has quit IRC | 04:17 | |
| *** shanipribadi has joined #lv2 | 04:19 | |
| *** shanipribadi has quit IRC | 04:32 | |
| *** shanipribadi has joined #lv2 | 04:34 | |
| *** shanipribadi has quit IRC | 04:38 | |
| *** shanipribadi has joined #lv2 | 04:39 | |
| *** shanipribadi has quit IRC | 04:49 | |
| *** shanipribadi has joined #lv2 | 04:55 | |
| *** shanipribadi has quit IRC | 05:06 | |
| *** shanipribadi has joined #lv2 | 05:09 | |
| *** shanipribadi has joined #lv2 | 05:10 | |
| *** gabrbedd has joined #lv2 | 05:13 | |
| *** shanipribadi has quit IRC | 05:15 | |
| drobilla | rgareus: I went to edit my submission but there is no option to upload another pdf... | 05:41 |
| drobilla | rgareus: oh, n/m. Completely separate option from 'Edit Submission' | 05:41 |
| *** falktx_ has joined #lv2 | 06:33 | |
| *** awilliam1 has joined #lv2 | 06:34 | |
| *** awilliam1 has quit IRC | 06:34 | |
| *** awilliam1 has joined #lv2 | 06:34 | |
| *** falktx has quit IRC | 06:38 | |
| *** drobilla has quit IRC | 06:38 | |
| *** awilliams has quit IRC | 06:38 | |
| *** Anchakor_ has quit IRC | 06:38 | |
| *** Anchakor_ has joined #lv2 | 06:39 | |
| *** awilliam1 is now known as awilliams | 06:39 | |
| *** drobilla has joined #lv2 | 06:50 | |
| *** falktx_ has quit IRC | 07:03 | |
| *** falktx has joined #lv2 | 07:19 | |
| *** abique has joined #lv2 | 07:41 | |
| *** timbyr has quit IRC | 08:10 | |
| *** timbyr has joined #lv2 | 08:12 | |
| *** abique has quit IRC | 08:21 | |
| *** abique has joined #lv2 | 08:22 | |
| *** timbyr has quit IRC | 08:34 | |
| *** timbyr has joined #lv2 | 08:36 | |
| *** NickSB2 has quit IRC | 09:23 | |
| Anchakor | drobilla: yeah that is good | 10:08 |
| *** Anchakor_ has quit IRC | 10:49 | |
| *** NickSB has quit IRC | 11:51 | |
| *** Joeboy_ is now known as Joeboy | 11:52 | |
| *** NickSB has joined #lv2 | 12:29 | |
| *** falktx has quit IRC | 12:32 | |
| *** falktx has joined #lv2 | 13:30 | |
| *** falktx_ has joined #lv2 | 15:23 | |
| *** falktx has quit IRC | 15:23 | |
| *** Anchakor_ has joined #lv2 | 15:30 | |
| * drobilla sighs | 15:39 | |
| drobilla | Participating on a *%^#& mailing list is not a very high price to pay to get to influence the direction of the APIs you use. Honestly. | 15:39 |
| Anchakor | who is complaining this time? :) | 15:47 |
| drobilla | Me? Hardly unusual :P | 16:03 |
| drobilla | Oh, I get it. No, not forum trolls this time, actually useful discussion, but in private means there's no record or community involvement, and really feeds the "one man show" image of the public record which I really do not like | 16:04 |
| *** abique has quit IRC | 16:06 | |
| *** abique has joined #lv2 | 16:07 | |
| drobilla | Whatever, pretty common, I just find it super weird. Like, hidden chains of people where Bob tod Alice told ............ about a post I made on the mailing list, finally Zeus gets around to starting a private conversation about it. | 16:07 |
| drobilla | I guess that means the ML post sort of kinda did part of it's job a little bit, but it's still really fucked up from my perspective :) | 16:07 |
| drobilla | or: Spammers ruin everything, film at eleven. | 16:08 |
| Anchakor | I don't know which ML post in particular you mean | 16:11 |
| Anchakor | but this social phenomenon is classic :) | 16:11 |
| drobilla | I should just CC the ML I suppose | 16:12 |
| drobilla | Aside from the mostly inconsequential fact that I find it weird, the real reason that's a problem is there's no public record about the "why" | 16:12 |
| wrl | do it | 16:12 |
| drobilla | Which is often a very useful thing to have | 16:12 |
| drobilla | Also increases how much you are all fucked if I get hit by a bus :) | 16:13 |
| * falktx_ is very confused right now | 16:13 | |
| falktx_ | drobilla: btw.... is it possible to deprecate instance-access usage like done with atom blank? | 16:15 |
| falktx_ | the instance-access doesn't have any types or structs, so I'm not sure if possible | 16:15 |
| drobilla | falktx_: It's not really deprecated, so much as was never suggested for abuse in the first place | 16:16 |
| falktx_ | well, I'd love to have it deprecated | 16:16 |
| drobilla | I don't know why the hell I made instance-access and data-access separate extensions. I was a little extension happy at the time. | 16:16 |
| drobilla | falktx_: Why? Just don't support it. | 16:16 |
| drobilla | I mean, I hate it on principle, so that sounds appealing personally, but it doesn't seem appropriate. | 16:17 |
| falktx_ | hehe, I'm not as crazy as you | 16:17 |
| falktx_ | well, I am , in a different way. I want *all* plugins :) | 16:18 |
| drobilla | Fool's errand. Not all plugins are suited for all host. This is a universal truth. | 16:18 |
| drobilla | Those extensions serve two purposes: 1) visualisation and such, which can gracefully degrade 2) kludge for easy porting and getting things to work as LV2 | 16:19 |
| drobilla | If you can't or don't want to support it, then don't. Problem solved. | 16:21 |
| drobilla | falktx_: I thought a bunch of your JUCE ports depended on it. That went away? | 16:21 |
| falktx_ | that is still the case | 16:21 |
| drobilla | ... Yet you want it deprecated | 16:21 |
| falktx_ | yes | 16:21 |
| drobilla | and you're not as crazy as me | 16:21 |
| drobilla | okay ;) | 16:21 |
| falktx_ | it's mostly for toolkit UIs + instance-data | 16:21 |
| falktx_ | like say gtk2+instance or qt4+instance | 16:21 |
| falktx_ | I really hate those | 16:21 |
| drobilla | I don't see how it is related to toolkit at all | 16:22 |
| wrl | imo there's value in instance-ui for extremely complex UIs | 16:22 |
| falktx_ | because toolkit mixup = crashes | 16:22 |
| falktx_ | for juce it's no issue because it's a static lib | 16:22 |
| drobilla | Anyway, I'll be pissing of plenty of people in the next little while by changing things with actual utility. | 16:22 |
| wrl | for example, how would you implement melodyne only with ports? | 16:22 |
| drobilla | So, no, I won't be doing even more of it just for shits and giggles. | 16:23 |
| drobilla | falktx_: I still don't see how acess to the plugin instance is related to that at all. | 16:23 |
| falktx_ | wrl: you don't have only ports. you have real ui<->communication now | 16:23 |
| wrl | ah yeah, via atoms right? | 16:23 |
| falktx_ | drobilla: I like to run my UIs in a separate process. with that I can't | 16:23 |
| falktx_ | drobilla: and because they rely on some toolkit, showing them might be a bad thing at times | 16:23 |
| drobilla | falktx_: Well, okay. Funny position from someone who contributed so many plugins like that to the world... | 16:24 |
| falktx_ | wrl: yes. eg-scope is a good example of this working afaik | 16:24 |
| falktx_ | drobilla: again, I'm ok with juce because it's not a system shared lib toolkit | 16:24 |
| wrl | interesting, i'll check that out. | 16:25 |
| drobilla | There's a big bold warning in the docs nobody reads about use not being recommended, and plugins not relying on it being supported. Feel free to point at it. | 16:25 |
| drobilla | falktx_: Well, this is hyper specific to your particular host and this particular platform and only tangentially related to what the actual extensions do. | 16:25 |
| Anchakor | wrl: read the work-in-progress paper about atoms drobilla pasted here yesterday or so ;) | 16:26 |
| drobilla | They weren't deprecated all these years when you bastards were pissing me off with Ingen. Touché :P | 16:26 |
| falktx_ | :P | 16:27 |
| wrl | Anchakor: yeah it's just a different way of thinking | 16:27 |
| drobilla | falktx_: Tangible reasons why things inherently CAN'T work are infinitely more compelling than me sticking some preprocessor directives somewhere. | 16:27 |
| drobilla | By all means, don't support it. I actively encourage you not to. | 16:28 |
| wrl | though for an interesting take on ui/dsp separation there have been a few VST3 threads on KVR recently where some fairly reputable devs weigh in on it | 16:28 |
| *** HarryHaaren has joined #lv2 | 16:28 | |
| Anchakor | will VST3 still be patented and closed source? | 16:29 |
| * drobilla highly doubts it's very interesting | 16:29 | |
| wrl | well it's not patented | 16:29 |
| wrl | but VST3 has this separation of UI and DSP baked in as well | 16:29 |
| wrl | and devs do *not* like it | 16:29 |
| drobilla | blah blah I'm so smart and I need direct access so I can do such and such blah blah | 16:29 |
| wrl | drobilla: dude, that attitude. | 16:29 |
| drobilla | meanwhile 99.99999% of developers fuck it up and make flaky shit plugins that cause dropouts when you click in the UI | 16:30 |
| drobilla | wrl: I'm sorry, did I give the impression that I had warm and fuzzy feelings about the VST community? ;) | 16:30 |
| wrl | it's not that, i just don't like this attitude that i get from a lot of the linux folks that there's nothing to be learned from or appreciated about the larger music software community | 16:31 |
| wrl | i.e. "the commercial/closed-source guys" | 16:31 |
| drobilla | Fine. I've heard it all before, and the arguments are of a very common pattern. | 16:31 |
| Anchakor | wrl: anyway I'd like if you linked the threads here, at least for reference | 16:32 |
| wrl | Anchakor: sure, sec | 16:32 |
| wrl | the big thread: http://www.kvraudio.com/forum/viewtopic.php?f=33&t=390814 | 16:33 |
| wrl | urs from u-he shares some insight in here, let me dig some posts out | 16:33 |
| drobilla | It's simply the wrong way to do UIs for real-time. | 16:34 |
| drobilla | Not even plugin specific. | 16:35 |
| wrl | drobilla: yeah absolutely. if anything, i'm not disagreeing with you, just unhappy with the dismissiveness. | 16:35 |
| drobilla | It's an absolute nightmare once you go down the shared state + locking path, which is precisely where everyone will go if you don't provide a better alternative | 16:35 |
| drobilla | wrl: I'm an asshole. Film at eleven. | 16:35 |
| drobilla | Much of it comes from just getting constantly rained on with hate due to this project, so forgive me if I'm pretty dismissive of whiny developers. | 16:36 |
| wrl | nah it's cool, i'll just push back a bit ;) | 16:36 |
| drobilla | Especially proprietary ones who contribute nothing back whatsoever. | 16:36 |
| falktx_ | that "AdmiralQuality" guy is always ranting about everything | 16:36 |
| wrl | takes some gumption and stubbornness to lead a project, i can appreciate it | 16:37 |
| drobilla | Frankly you're lucky this card carrying FSF member didn't AGPL3 the whole thing and support libraries. Fuck 'em. ;) | 16:37 |
| drobilla | There is 0% love in my portability efforts and 100% "destroy them from the inside". | 16:37 |
| wrl | destroy who, lv2? | 16:38 |
| drobilla | VST, for one. Proprietary music software in general. | 16:39 |
| wrl | yeah | 16:39 |
| drobilla | Especially plugin APIs with licensing (and maybe even patent) tricks that are actively hostile to free software. | 16:39 |
| wrl | well, imo, what lv2 would need to start wooing people would be some "killer plugin" or feature that can really only be realized with lv2 | 16:39 |
| wrl | there was this situation where the melodyne guys wanted really deep integration into a host (presonus's studio one iirc) and they had to hack up their own VST extension to support it | 16:40 |
| wrl | essentially they wanted access to the raw audio data on the timeline | 16:40 |
| drobilla | Well, network transparent UIs is such a feature for some... | 16:40 |
| wrl | situations like that would be, imo, where lv2 could shine | 16:40 |
| wrl | oh yeah absolutely | 16:40 |
| drobilla | Whiiiich many others will hate you for doing. | 16:41 |
| wrl | i'm of the mind that the option is good | 16:42 |
| wrl | or rather that having the option is the better choice | 16:42 |
| drobilla | Anyway, clearly I am ridiculously stressed the fuck out and working at all on this crap is an ultra bad idea. So you're welcome. | 16:42 |
| drobilla | Almost as bad of an idea as being on IRC right now. ttyl :) | 16:42 |
| wrl | the vibe that i get from the vst3 threads on kvr, honestly, is not "fuck ui/dsp separation" or "fuck this async message bus stuff" it's "don't tell me how to write my code" | 16:42 |
| wrl | alright dude, take it easy | 16:42 |
| *** drobilla has quit IRC | 16:44 | |
| wrl | :/ | 16:46 |
| wrl | c'mon. | 16:46 |
| Anchakor | the urs guy says interesting stuff: "Hehehe, I really like AAX. Don't care much about the DSP stuff, but the SDK is simple, well strcutured and the object communication is great. It fulfills the promise that AU made, without the bloat. " | 16:52 |
| Anchakor | "If there was ever gping to be another discussion about a general plug-in standard, I think tehy should take a close look at AAX. | 16:52 |
| Anchakor | (without the PACE signing stunt, goes without saying)" | 16:52 |
| wrl | would be interesting to see what the rack extension API looks like as well | 16:54 |
| wrl | unfortunately that's even more locked down :/ | 16:54 |
| falktx_ | rewire is locked down but can work | 16:56 |
| falktx_ | I'm 25% there | 16:56 |
| *** abique has quit IRC | 17:18 | |
| *** HarryHaaren has quit IRC | 17:20 | |
| *** NickSB2 has joined #lv2 | 17:20 | |
| wrl | oh huh, didn't know you could use urn:s as lv2 identifiers | 17:34 |
| falktx_ | they are valid uris | 17:35 |
| falktx_ | file:/// is one too | 17:35 |
| falktx_ | which means root | 17:35 |
| *** Anchakor has quit IRC | 17:46 | |
| *** Anchakor1 has joined #lv2 | 17:46 | |
| Anchakor1 | yeah but you run risk of non-uniqueness | 17:46 |
| falktx_ | urn:author:name is usually unique enough | 17:47 |
| Anchakor1 | which is the point we are are using URIs in the first place | 17:47 |
| *** Anchakor1 is now known as Anchakor | 17:48 | |
| Anchakor | many people have the same names/nicks | 17:48 |
| falktx_ | urn:distrho:3bandeq | 17:49 |
| falktx_ | I doubt anyone will use that | 17:49 |
| falktx_ | but I do prefer real URLs anyway | 17:50 |
| Anchakor | you can place your documentation at the http addresses | 17:54 |
| rgareus | FWIW, all LV2 identifiers are URIs. There's a nice euler diagram at https://en.wikipedia.org/wiki/URI | 18:34 |
| rgareus | you could even use mailto:my.cool@spam.org as ID and refer to plugin-documentation via ISBN, urn:isbn:0-123-666888-4 :) | 18:37 |
| falktx_ | hmmm | 18:38 |
| falktx_ | jalv.gtk maito:drobilla@drobilla.com :D | 18:38 |
| rgareus | at least there's no shell-escape chars in there. | 18:40 |
| *** awilliams has quit IRC | 19:38 | |
| *** mlpug has joined #lv2 | 20:10 | |
| *** mlpug has quit IRC | 21:02 | |
| *** awilliams has joined #lv2 | 22:47 | |
| *** awilliams has quit IRC | 23:12 | |
Generated by irclog2html.py 2.13.0 by Marius Gedminas - find it at mg.pov.lt!