| *** kaspar_e has joined #lv2 | 00:09 | |
| *** timbyr has quit IRC | 00:30 | |
| *** timbyr has joined #lv2 | 00:32 | |
| *** rncbc|AFK has quit IRC | 00:46 | |
| *** timbyr has quit IRC | 00:57 | |
| *** timbyr has joined #lv2 | 00:59 | |
| *** kaspar_e has quit IRC | 01:11 | |
| *** speps has quit IRC | 01:39 | |
| *** speps has joined #lv2 | 01:53 | |
| *** timbyr has quit IRC | 02:53 | |
| *** timbyr has joined #lv2 | 02:55 | |
| *** HarryHaaren has quit IRC | 04:03 | |
| *** falktx_ has quit IRC | 04:09 | |
| *** speps has quit IRC | 05:46 | |
| *** awilliams has quit IRC | 07:20 | |
| *** awilliams has joined #lv2 | 07:21 | |
| *** abique has joined #lv2 | 07:28 | |
| *** abique has quit IRC | 07:33 | |
| *** abique has joined #lv2 | 07:46 | |
| *** edogawa has joined #lv2 | 08:55 | |
| *** kaspar_e has joined #lv2 | 13:28 | |
| *** HarryHaaren has joined #lv2 | 13:41 | |
| *** falktx has joined #lv2 | 13:48 | |
| *** mlpug has joined #lv2 | 14:04 | |
| *** falktx_ has joined #lv2 | 15:29 | |
| *** falktx has quit IRC | 15:32 | |
| *** edogawa has quit IRC | 16:22 | |
| *** edogawa has joined #lv2 | 16:29 | |
| *** sigma6 has quit IRC | 16:48 | |
| *** sigma6 has joined #lv2 | 16:50 | |
| *** sigma6 has left #lv2 | 16:56 | |
| *** drobilla has joined #lv2 | 16:59 | |
| drobilla | Heh, apparently the standard practice in glut for grabbing the mouse is to warp it back into the window on every evet | 17:05 |
|---|---|---|
| drobilla | I wonder how much fun that'll be to do on win/osx | 17:06 |
| drobilla | (actual grabbing, that is) | 17:06 |
| wrl | are you going to have pugl act similarly? | 17:17 |
| drobilla | Warping? Unlikely, that's awful. | 17:19 |
| wrl | agreed | 17:20 |
| drobilla | Being able to grab is pretty crucial for many things like knobs and slider controls you drag, though, so API will be needed | 17:21 |
| wrl | yeah | 17:21 |
| falktx_ | grab and hide cursor? | 17:22 |
| wrl | i like the grab and hide cursor pattern | 17:22 |
| falktx_ | I like when knobs do that | 17:22 |
| drobilla | There is API for changing the cursor which can presumably be used for the hide bit, at least | 17:22 |
| drobilla | The tricky/annoying thing may be event masks, since it has no event types exposed at all | 17:23 |
| drobilla | Maybe "grab mouse everything ever" is good enough | 17:23 |
| falktx_ | I did that once for X11, there was an issue when the user releases the mouse | 17:23 |
| falktx_ | if the window behind the cursor is not the plugin, it jerks a bit | 17:24 |
| drobilla | The Gtk grab stuff seems to work well for me, but I'm not sure what it does internally | 17:24 |
| wrl | falktx_: couldn't you just rewarp the pointer back to the drag start after every motion notify? | 17:25 |
| falktx_ | wrl: don't remember the code exactly now, its better to have a fresh start | 17:25 |
| falktx_ | X11 API is confusing | 17:26 |
| wrl | x11 is a ghetto | 17:26 |
| wrl | wayland will save us all | 17:26 |
| wrl | i need to figure out why this happens, but like | 17:27 |
| wrl | i added a pugl backend to rutabaga and all the window resizes got super laggy compared to the xcb code i had already written | 17:27 |
| drobilla | The main problem with warping is the mouse position gets all wacky and preserving that would be really tedious (and probably error prone) | 17:30 |
| drobilla | I wonder if that's xlib vs xcb or something else. I was idly wondering what if any wins using xcb instead would yield. | 17:30 |
| drobilla | (As for Wayland, now way I'm getting involved in the display server mess until the dust settles) | 17:31 |
| wrl | drobilla: xcb if nothing else is considerably more verbose | 17:32 |
| wrl | and to do any GL stuff xlib needs to be involved if only slightly | 17:32 |
| drobilla | Meh to that | 17:32 |
| * falktx_ blames glx | 17:33 | |
| wrl | yeah | 17:33 |
| wrl | falktx_: yep! | 17:33 |
| wrl | it's all glx. | 17:33 |
| falktx_ | wayland will be great to sort this out | 17:33 |
| falktx_ | they have to :) | 17:33 |
| wrl | wayland is already pretty cool | 17:34 |
| wrl | it's going to be a critical mass thing imo | 17:34 |
| * drobilla is definitely not getting involved in that mess until the dust completely settles | 17:35 | |
| drobilla | Though I guess in the case of pugl the amount of platform code is relatively small and totally isolated, so tinkering shouldn't be that big a deal | 17:35 |
| drobilla | Speaking of rutabega, two things: | 17:35 |
| drobilla | 1) Last I tried, compilation failed due to -Werror. Having that on by default is slightly insane. | 17:36 |
| drobilla | 2) Can't update: fatal: reference is not a tree: 7efa265e68d55ad551f6e8b364df30c309ddc815 | 17:36 |
| drobilla | Unable to checkout '7efa265e68d55ad551f6e8b364df30c309ddc815' in submodule path 'third-party/libuv' | 17:36 |
| wrl | ah yeah, give me a sec | 17:36 |
| falktx_ | I hope plugin UIs won't break because of wayland. Xembed over XWayland sounds painful | 17:36 |
| wrl | drobilla: submodules are a pain in the ass | 17:36 |
| wrl | but subtree is even worse | 17:36 |
| wrl | ugh i need to keep working on that pull request too | 17:37 |
| drobilla | Yeah. Recent minor work on Jack is my first experience with them | 17:38 |
| drobilla | I don't look forward to migrating my many sub-projects to git... | 17:38 |
| wrl | drobilla: ah, i did some fucked up rebasing in my libuv mirror, sec | 17:39 |
| drobilla | They probably will. Such is the cost of exposing the lowest level thing, but it's pretty much either that or toolkit | 17:39 |
| drobilla | Then again it's not like we're the only people embedding anything, so maybe it will magically work | 17:40 |
| *** falktx_ has quit IRC | 17:43 | |
| * drobilla would still like to tear the drawing stuff from the event stuff, but it's not a particularly good use of his time | 17:44 | |
| wrl | damn, for some reason the compiler is now completely ignoring -fvisibility=hidden | 17:45 |
| wrl | the hell | 17:46 |
| wrl | hah | 17:53 |
| wrl | well, clang pays attention to it, gcc doesn't though | 17:54 |
| *** mlpug has quit IRC | 17:54 | |
| wrl | drobilla: git pull, should be alright now | 17:56 |
| drobilla | We could really use a standard that allows build systems to map a pkg-config name to a package name, and thus offer to automagically install | 17:58 |
| drobilla | wrl: death by warning | 17:59 |
| wrl | drobilla: can you pastebin the output? | 17:59 |
| drobilla | http://pastebin.com/q1aJYBas | 17:59 |
| *** mlpug has joined #lv2 | 17:59 | |
| wrl | danke sehr | 17:59 |
| wrl | oh yeah what are your cflags again? | 18:00 |
| drobilla | -O2 -march=corei7-avx -fomit-frame-pointer -DNDEBUG -ftree-vectorize | 18:00 |
| wrl | oh yeah that's the issue, you compile with -DNDEBUG | 18:00 |
| wrl | and the examples have a bunch of dumb shit in assert()s | 18:00 |
| wrl | which i should have fixed already and the warnings in this case are actually big deals | 18:01 |
| drobilla | Right, builds with debuggey flags | 18:02 |
| * drobilla just nukes user CFLAGS entirely if configure --debug is given | 18:02 | |
| wrl | cabbage patch breaks with ndebug tho, let me see what's up | 18:03 |
| wrl | (same issue) | 18:03 |
| drobilla | There appears to be no text in anything | 18:03 |
| drobilla | Otherwise, works | 18:03 |
| wrl | oh wow, hm. | 18:04 |
| wrl | send a screenshot? | 18:04 |
| wrl | also, should build fine with NDEBUG now | 18:04 |
| drobilla | Do I really need to make a screenshot to prove there's no text anywhere? :) | 18:04 |
| wrl | okay, okay ;) | 18:04 |
| wrl | let me think about why that might be | 18:04 |
| wrl | no text in anything, huh. | 18:05 |
| wrl | hm | 18:05 |
| wrl | so, in cabbage_patch | 18:05 |
| drobilla | cabbage_patch or test, no text. modules and knobs, respectively, no labels anywhere | 18:05 |
| wrl | in cabbage_patch, is there space for text, and it's just not there, or is there not any space for it? | 18:06 |
| drobilla | oooooooohhhhhhhh fine | 18:06 |
| wrl | haha | 18:07 |
| drobilla | (also apps don't close when the window is closed) | 18:07 |
| wrl | i guess it wouldn't be a problem loading the font, because at the moment that generally just makes it segfault | 18:07 |
| drobilla | I like the potential of this project + pugl for plugin GUIs anyway | 18:08 |
| wrl | yeah that's kind of the aim | 18:08 |
| drobilla | I guess Robin also has that Gtk-API-like one too | 18:08 |
| wrl | yeah, yeah | 18:08 |
| wrl | drobilla: what WM, and also can you pastebin me your glxinfo? | 18:08 |
| drobilla | So many of the pre-existing ones try really hard to look like Windows 95 in GL, it's weird | 18:08 |
| drobilla | http://drobilla.net/files/cabbage_patch.png | 18:08 |
| wrl | okay, probably a GL issue | 18:09 |
| drobilla | http://pastebin.com/p4X9geMB | 18:09 |
| wrl | font looks like it's being loaded fine, since the bounds are being factored into the layout | 18:09 |
| drobilla | In awesome | 18:09 |
| wrl | hm | 18:10 |
| wrl | okay you're running a version of mesa about one major release behind me, but have almost the exact same card | 18:10 |
| wrl | let me test on my old old old machine, see if i broke anything recently | 18:11 |
| * drobilla supposes he will have to abandon his stubborn hatred of knobs on screens if he wants to build a control panel builder thingie | 18:15 | |
| wrl | haha! | 18:16 |
| wrl | well | 18:16 |
| wrl | rtb_fanslider? | 18:16 |
| wrl | (it's not there, i haven't written it) | 18:16 |
| drobilla | Ganv effectively uses fansliders without the fan | 18:17 |
| drobilla | AFAIK fan isn't really nicely doable without requiring compositing | 18:17 |
| wrl | yeah | 18:17 |
| drobilla | Whatever, not very important. | 18:18 |
| *** falktx_ has joined #lv2 | 18:23 | |
| drobilla | falktx_: http://dev.drobilla.net/ticket/952 | 18:24 |
| LAbot | Title: #952 (suil fails to build on OSX when qt4 is installed) – drobillad (at dev.drobilla.net) | 18:24 |
| drobilla | falktx_: I don't understand this. You mean it works with Qt5, or you just did that to make it not use Qt at all? | 18:24 |
| drobilla | rgareus: Can we turn that part off per-channel by any chance? | 18:24 |
| falktx_ | drobilla: I don't have qt5, so it stkipped qt build | 18:25 |
| wrl | drobilla: i need to sort dinner out, but i want to see if we can figure out what's up with the text rendering. things are kosher on my fairly-old test system | 18:25 |
| drobilla | falktx_: Looks like a check for QtX11 something is required, though, oddly, it seems qx11embed_x11.h does exist | 18:25 |
| drobilla | wrl: k. I'll probably stick around all day even though I shouldn't | 18:26 |
| drobilla | falktx_: Can't really sort this out without OSX around | 18:26 |
| falktx_ | drobilla: np, I understand | 18:26 |
| drobilla | falktx_: Speaking of which, what installer did you use to get a vbox running? | 18:26 |
| falktx_ | drobilla: suil doesn't support cocoa yet anyway | 18:26 |
| drobilla | ../src/gtk2_in_qt4.cpp:21:22: error: gdk/gdkx.h: No such file or directory | 18:26 |
| drobilla | I guess checking for that would do | 18:26 |
| falktx_ | drobilla: I use iDeneb 10.5.8 | 18:26 |
| falktx_ | in latest VirtualBox | 18:26 |
| falktx_ | the ISO i have here is no longer online though | 18:27 |
| drobilla | falktx_: No, but it does work on OSX if X11 is around, which AFAIK is still how Ardour rolls | 18:27 |
| falktx_ | but macports qt4-mac is built without x11 | 18:27 |
| drobilla | Sure. Being actually X11 dependent raises the question of whether having qt_in_gtk nad vice versa is even worth it, really | 18:29 |
| drobilla | Gtk in theory has an API for this, but plug/socket only actually works on X11 too, I think | 18:29 |
| falktx_ | having gtk paint on a rgb buffer might be a better choice | 18:29 |
| drobilla | Doesn't sound fun | 18:32 |
| drobilla | I just meant using the same basic "get system window thing" based wrapping for every combo | 18:32 |
| drobilla | instead of tk1_in_tk2 for every combo, a get_win_for_tk1, and embed_win_in_tk2 | 18:33 |
| rgareus | drobilla: ? the bot url-title? | 18:33 |
| drobilla | rgareus: yeah | 18:34 |
| rgareus | drobilla: mmh. not sure. maybe. I'll have a look | 18:34 |
| drobilla | rgareus: Seems kinda noisy and pointless to me, but not really a big deal either | 18:34 |
| rgareus | drobilla: meanwhile, you should mention in the topic that this channel is publicly archived | 18:35 |
| rgareus | drobilla: http://linuxaudio.org/irc/lv2/ | 18:35 |
| LAbot | Title: #lv2 IRC logs | Linuxaudio.org (at linuxaudio.org) | 18:35 |
| rgareus | drobilla: according to the doc, the 'titleSnarfer' config is channel-specific. So now I only need to find out how to set it per channel .. | 18:39 |
| *** ChanServ sets mode: +o drobilla | 18:40 | |
| * rgareus should have chosen a simpler bot | 18:41 | |
| *** drobilla changes topic to "LV2 - http://lv2plug.in - Logged at http://linuxaudio.org/irc/lv2/" | 18:42 | |
| drobilla | Big brother is watching you ;) | 18:42 |
| *** drobilla sets mode: -o drobilla | 18:42 | |
| rgareus | test: http://lv2plug.in/ | 18:45 |
| rgareus | drobilla: seems to work. | 18:45 |
| drobilla | rgareus: Thanks | 18:47 |
| drobilla | I guess the two papers in LAC2013 are probably the only published LV2 things currently | 19:05 |
| falktx_ | drobilla: you're going to lac? | 19:09 |
| drobilla | falktx_: Well, I'm probably going to submit a paper anyway | 19:10 |
| falktx_ | deadline is 27th, right? | 19:10 |
| drobilla | Despite really having to work on other things... | 19:10 |
| drobilla | falktx_: yeah | 19:10 |
| falktx_ | I want to submit a paper as well, but I'm most likely to not be present :( | 19:10 |
| drobilla | I figure general atom-idea paper this year, then Ingen next year, since the client/server aspect of it heavily builds on these ideas | 19:11 |
| drobilla | Yeah, well.. hopefully the grad office will pay for most of my flight... otherwise, I dunno :) | 19:11 |
| drobilla | This bloody long citation style makes line length/filling brutal | 19:13 |
| drobilla | Writing this makes me strongly inclined to nuke atom:Blank and atom:Resource. It looks weird. | 19:50 |
| *** abique has quit IRC | 19:57 | |
| drobilla | TIL erlang calls them "terms" | 20:09 |
| drobilla | and "atom", like X11, is essentially a symbol | 20:09 |
| drobilla | Perhaps an unwise overload, though I think "symbol" is much clearer than "atom" for that | 20:09 |
| wrl | agreed | 20:20 |
| drobilla | 'item' would have been a good ext name | 20:25 |
| drobilla | that that implies "not collection" as much as atom does | 20:25 |
| drobilla | whatever, I guess | 20:25 |
| *** rncbc has joined #lv2 | 20:32 | |
| wrl | yeah, it's almost like the way that a LV2 atom is currently used is more as a collection of atoms | 20:35 |
| drobilla | Well, there are both. Some are primitives, some are collections | 20:36 |
| drobilla | String is a variable sized kinda-primitive which mucks up the concepts a bit | 20:36 |
| *** HarryHaaren has quit IRC | 20:39 | |
| *** mlpug has quit IRC | 20:46 | |
| * drobilla fills forge.h up with a bunch of deprecated warning pragma filth | 20:53 | |
| drobilla | sigh. | 20:53 |
| drobilla | Pretty stupid to be making deprecated warnings about code inside deprecated functions, gcc | 20:53 |
| falktx_ | drobilla: btw, this is an odd request... but I'm forcing myself to use more strict gcc flags and lv2 headers fail on them | 20:56 |
| falktx_ | my full list is this: | 20:56 |
| falktx_ | -Wall -Wextra -Wcast-qual -Wconversion -Wlogical-op -Werror | 20:57 |
| drobilla | There are const problems in some of the atom stuff | 20:57 |
| falktx_ | not just const | 20:57 |
| falktx_ | there are some size issues as well | 20:57 |
| drobilla | -Wcast-qual will cause warnings | 20:57 |
| drobilla | like, alignment? | 20:57 |
| falktx_ | mostly things like this: | 20:58 |
| falktx_ | uint32_t someFunc() { return sizeof(int); } | 20:58 |
| falktx_ | it's very picky yes, but I find that good to spot little mistakes in my code | 20:58 |
| drobilla | (-Werror is insane) | 20:58 |
| drobilla | Werror doesn't make anything more picky, it just makes builds fail | 20:58 |
| drobilla | Fair enough if that's your thing, but don't have it on by default for users | 20:59 |
| drobilla | I will check what is missing from my --ultra-strict flags | 20:59 |
| falktx_ | yes, this is only for the developmen git branch | 20:59 |
| falktx_ | production code should never have -Werror | 20:59 |
| falktx_ | specially since some distros add their own -W stuff | 21:00 |
| drobilla | Unfortunately -Wcast-qual and C is very problematic. C sucks at const-correct | 21:00 |
| drobilla | The standard library isn't even const correct | 21:00 |
| drobilla | If/when I make successor versions of those utility headers I can fork the entire API in 2 and add _c suffixes to const things or something | 21:00 |
| wrl | -Wextra is insane | 21:01 |
| * drobilla already broke other code by trying to conservatively "fix" as much of these as he could | 21:01 | |
| drobilla | wrl: All my code is -Wextra clean | 21:01 |
| wrl | heh, damn | 21:01 |
| wrl | i'm mostly running afoul of missing field initializers, seems | 21:01 |
| drobilla | falktx_: It's -Wconversion that I don't use | 21:02 |
| falktx_ | ok, I'd be nice to have that sorted | 21:02 |
| drobilla | falktx_: We'll see, but cast-qual can't be without breaking the API | 21:03 |
| falktx_ | drobilla: regarding const, I set all lv2-atom stuff to use consts. then in c++ I can const_cast<type*>() when I need | 21:03 |
| *** speps has joined #lv2 | 21:11 | |
| drobilla | falktx_: Most of the problem is iterators. Needs proper C++ iterator interface anyway, so that will make it go away for C++ at least | 21:12 |
| drobilla | The rest will have to remain as-is indefinitely. util2.h or perhaps iterator.h can replace those parts | 21:13 |
| *** unclechu has joined #lv2 | 21:26 | |
| *** Gethiox3 has joined #lv2 | 21:27 | |
| unclechu | hi guys, please tell me how I can convert db to coefficient? | 21:27 |
| drobilla | unclechu: There is a macro for that in the amp example | 21:28 |
| drobilla | /** Define a macro for converting a gain in dB to a coefficient */ | 21:29 |
| drobilla | #define DB_CO(g) ((g) > -90.0f ? powf(10.0f, (g) * 0.05f) : 0.0f) | 21:29 |
| *** Gethiox2 has quit IRC | 21:29 | |
| unclechu | drobilla but for stereo I must use half of this value? | 21:29 |
| drobilla | unclechu: That really depends on context | 21:30 |
| *** HarryHaaren has joined #lv2 | 21:31 | |
| unclechu | drobilla ok, thanks | 21:31 |
| drobilla | unclechu: y/w | 21:31 |
| unclechu | drobilla ?? | 21:31 |
| unclechu | drobilla aw, "you're welcome" i guess) | 21:35 |
| rgareus | is 'math-functions.lv2' some legacy thing or just some obscure plugin? Does anyone know who maintains it? | 21:49 |
| falktx_ | lars luthman? maybe | 21:49 |
| rgareus | it produces around 30 or so error: /usr/lib64/lv2/math-functions.lv2/math-functions.ttl:XXX name ends with `.' | 21:50 |
| falktx_ | old calf used to have such plugins too | 21:50 |
| *** HarryHaaren has quit IRC | 21:50 | |
| rgareus | (just seen that on a debug message https://community.ardour.org/node/7962) | 21:51 |
| rgareus | there's sord_validate. use it or loose it. | 21:51 |
| *** HarryHaaren has joined #lv2 | 21:51 | |
| falktx_ | never used it | 21:52 |
| falktx_ | never got it to work either | 21:52 |
| falktx_ | rgareus: ok, got that plugin in my ubuntu install. it's ll-plugins-lv2 package | 21:52 |
| falktx_ | ie, http://ll-plugins.nongnu.org/ | 21:52 |
| drobilla | I think that's a new serd error (caused by the new Turtle grammar allowing dots in things). | 21:52 |
| drobilla | Because breaking widely used grammars is a super fucking great fucking, W3C | 21:53 |
| drobilla | idea* even, heh | 21:53 |
| drobilla | Hm. I don't know if that's valid and a serd error, or invalid. | 21:54 |
| falktx_ | non-compilable text files will always be like that | 21:54 |
| drobilla | It's very hard to get right, ambiguous mess. | 21:54 |
| falktx_ | web browsers can tell... | 21:54 |
| drobilla | Web browsers dug their own grave with the initial stupid philosophy of "tolerant" parsing | 21:55 |
| drobilla | Web people in general don't give a damn about implementation burden whatsoever, it's baked into the culture at this point. Tens of thousands of lines of filthy error-prone garbage to work around every possible way an input can be outright wrong is best practice, there. | 21:56 |
| drobilla | rapper parses it, so it may be my fault | 21:57 |
| rgareus | we need to introduce the "certified LV2 developer" exam :-) | 21:58 |
| * drobilla gets mad even thinking about the RDF working group | 21:58 | |
| * falktx_ wants a lv2-ttl-certified banner on his plugins now | 21:58 | |
| Anchakor | don't worry I'll make RDF alternative this summer :) | 21:58 |
| drobilla | However, regardless, best practice in ttl itself is to put a space before your delimiters, which avoids this | 21:59 |
| falktx_ | rgareus: do you think it's safe to update to latest a3 right now? | 21:59 |
| rgareus | falktx_: there was a relase just today. | 21:59 |
| falktx_ | rgareus: I'm speaking of that one | 22:00 |
| rgareus | falktx_: it's definitley better than 3.5.143 (which had some semi-serious issues. e.g. potential buffer x-talk) | 22:01 |
| wrl | drobilla: hey, still around? | 22:05 |
| drobilla | Anchakor: My only real grip with the data model is the "language OR datatype" thing | 22:05 |
| drobilla | Anchakor: Syntax-wise Turtle and friends are WAY too hard to parse and full of arbitrary/redundant/inappropriate SPARQL bullshit now, though | 22:05 |
| drobilla | wrl: ... | 22:06 |
| wrl | drobilla: alright cool | 22:06 |
| wrl | drobilla: if you've got a sec i want to see if we can work through some stuff wrt the rutabaga text | 22:06 |
| drobilla | wrl: Stack is a bit deep at the moment | 22:07 |
| wrl | alright, no worries | 22:07 |
| Anchakor | drobilla: well there are other inelegant things, and the issue of branding and choice of language and audience focus of the specs, but I suppose you care mainly about syntaxes (which I don't) | 22:09 |
| drobilla | Anchakor: Your hyper extremist dismissal of anything human friendly frankly makes me pretty dubious I'll like what you cook up ;) | 22:09 |
| drobilla | (like, numbers for all IDs) | 22:10 |
| drobilla | Anchakor: Branding/language/audience is the main reason the semweb is a failure though, that's for sure | 22:10 |
| drobilla | json.org is how you sell a model | 22:10 |
| drobilla | The W3C is like a farce of exactly how to turn everybody completely off of one | 22:11 |
| drobilla | Well, that and RDF/XML, which is a wonderful example of how syntax matters | 22:12 |
| Anchakor | syntax matters in short term | 22:12 |
| Anchakor | in long term, not so much (hopefully :)) | 22:13 |
| drobilla | Short term, like: "this is ugly, I'm not using it" | 22:13 |
| drobilla | Which makes long term irrelevant, because you're dead | 22:13 |
| Anchakor | but lets just agree that it matters :) | 22:13 |
| Anchakor | yep | 22:13 |
| drobilla | That said, model being distinct from syntax is certainly important anyway, but if your standard one is ugly, you're doomed | 22:14 |
| falktx_ | rgareus: got 30 secs? | 22:14 |
| drobilla | wrl: I can try things, but I can't go through the glacial process of having a dialog while manually tinkering with things | 22:14 |
| * drobilla is already like 8 frames down from what he should be doing | 22:15 | |
| wrl | drobilla: don't sweat it too much. was basically just going to pastebin you some shader code and have you swap out the text shader a few times | 22:15 |
| rgareus | falktx_: yes | 22:17 |
| * drobilla discovers he is, infact, using blank node IDs in Ingen | 22:18 | |
| drobilla | FFFFFFFFFFFFFFFUUUUUUUUUUUUu | 22:18 |
| drobilla | Though this is problematic anyway because you can't always round-trip serialise that, if it's abbreviated | 22:19 |
| drobilla | Needs a property. Clearest name for a sequence number property? | 22:27 |
| drobilla | e.g. I send a GET, and receive a reply, both having a matching number means the reply is for that get | 22:28 |
| drobilla | I guess sequenceNumber is like TCP, but a bit wordy | 22:30 |
| *** HarryHaaren_ has joined #lv2 | 22:36 | |
| *** HarryHaaren has quit IRC | 22:38 | |
| *** HarryHaaren_ is now known as HarryHaaren | 22:39 | |
| *** edogawa has quit IRC | 23:37 | |
Generated by irclog2html.py 2.13.0 by Marius Gedminas - find it at mg.pov.lt!
