IRC log for #riftuidev on 20141012

01:26.45Nosheigood bye
08:29.49*** join/#riftuidev borbra (~quassel@grm-128-39-147-128.studby.uia.no)
11:53.46*** join/#riftuidev jorian2 (5ed71ff6@gateway/web/freenode/ip.94.215.31.246)
11:53.58jorian2hello
11:54.33jorian2can someone help me with this rift addon i'm trying to make?
13:48.39*** join/#riftuidev Imhothar (~Imhothar@ppp-93-104-169-154.dynamic.mnet-online.de)
16:00.17*** join/#riftuidev Imhothar1 (~Adium@ppp-93-104-169-154.dynamic.mnet-online.de)
16:51.27seebsYou know, I think this jorian2 person is going to have to learn patience.
17:20.12Torhalseebs: IRC is from another era, whough. Everyone these days is used to "give it to me now."
19:26.07*** join/#riftuidev Noshei (~quassel@c-24-9-171-99.hsd1.co.comcast.net)
20:00.59*** join/#riftuidev ZorbaBeta (tocohl@maximilian.pavlovian.net)
20:01.54*** join/#riftuidev borbra (~quassel@grm-128-39-147-128.studby.uia.no)
20:02.13*** join/#riftuidev micgro42 (~Michael@f052155065.adsl.alicedsl.de)
20:11.43micgro42hi could you give me an example for Command.Console.Display() with html? 'Command.Console.Display("combat", true,<font color=#0000ff>"this text is blue", true)' gives me an error
20:14.00ZorbaBetagotta put the <font color> stuff inside the quotes
20:14.21ZorbaBetaand there may be more requirements - maybe <font color='#0000ff'>
20:14.56Imhothar1a closing </font> is probably advisable as well
20:16.03Imhothar1Command.Console.Display("combat", true,"<font color='#0000ff'>this text is blue</font>", true)
20:16.07micgro42mh, better, but 'Command.Console.Display("combat", true,"<font color=#0000ff>this text is blue</font>", true)' passes the startup, but gives me a generic error message on execution
20:16.31Imhothar1put the color value in single quotes
20:16.58micgro42yes the single quotes around the color helped
20:17.06micgro42thank you :)
20:17.07Imhothar1or use extended literals with [[]] then use can use double quotes
20:17.23Imhothar1[[<font color="#0000ff">this text is blue</font>]]
20:17.34ZorbaBetayeah, the parser is pretty strict
20:26.51NosheiZorbaBeta: it isn't using bash is it?
20:27.56ZorbaBetascaleform has its own html parser, and we're doing a pre-parse and validation step before that
20:30.01Nosheiyeah my comment was a joke
20:30.19Imhothar1so the answer is… yes?
20:30.24ZorbaBetagiven how much nastiness is in the engine, it wouldn't surprise me if it wasn't :V
20:30.39ZorbaBetalast week I actually had to implement a workaround for an nvidia driver bug >_<
20:30.55Imhothar1go oooon…
20:31.33ZorbaBetain SLI mode, nvidia takes a Clear call, used to clear colors from a buffer that was rendered to, as a sign that this buffer does not need to be shared among GPUs
20:31.34Imhothar1there's a shitload of driver bugs on androids
20:31.41ZorbaBetathe problem is that they don't test to see how much of the buffer was cleared
20:32.00Imhothar1oh
20:32.01ZorbaBetaand so if you're clearing only part of a buffer, expecting for the rest of the buffer to be preserved, they'll happily still discard remaining data
20:32.16Imhothar1yeah that does happen
20:32.37ZorbaBetayeah, check the clear bounds, guys :V
20:32.41Imhothar1they like to forget that glScissor and glViewport affect glClear
20:33.00ZorbaBetait's even worse than that - on dx, the Clear function just straight-up takes a rect bound
20:33.07Imhothar1heh
20:33.18Noshei./lost
20:33.29ZorbaBetaand it works great on non-SLI, note, so they're clearly paying attention to that parameter
20:33.31Imhothar1though in GL there is an extra glDiscard(EXT) call for that
20:33.32ZorbaBetajust not enough attention
20:33.51ZorbaBetanoshei, I say "clear this small part of this texture", nvidia says "OKEYDOKEY WE THREW AWAY THE ENTIRE TEXTURE JUST LIKE YOU WANTED"
20:34.04Nosheioh yeah i got that part
20:34.12Nosheijust not the rest of it after that
20:34.56ZorbaBetait's sort of understandable if a function relies on a lot of extra global state, which OGL is kind of notorious for, but when the function actually includes a parameter saying what it should do . . . come on guys. read the parameter. that is what it is there for.
20:35.02ZorbaBetawelp
20:35.12ZorbaBetaanyway sli shadows will work again in 3.0, I hope.
20:37.15Imhothar1currently an entire series of Adreno GPUs on androids have bugs in how the driver computes uniform offsets in arrays of structs
20:38.38ZorbaBetawonderful
20:38.52Imhothar1setting a single array value writes to the entire array
20:38.59Imhothar1at least that's what it looks like
20:39.11Imhothar1they guys at Qualcomm aren't really sure yet
20:39.22Imhothar1and I reported the bug half a yea ago
20:39.31Imhothar1s/yea/year
20:39.37Imhothar1ach
20:40.05Nosheiput a / on the end
20:40.16Imhothar1I know, I always forget it
20:40.46Nosheirunning through stillmoor with mobs is strange
20:40.47Imhothar1it's the number one error I make whenever I use sed
22:09.14*** join/#riftuidev ZorbaGama (tocohl@maximilian.pavlovian.net)
22:21.25*** part/#riftuidev micgro42 (~Michael@f052155065.adsl.alicedsl.de)
22:21.32*** join/#riftuidev micgro42 (~Michael@f052155065.adsl.alicedsl.de)
22:21.57*** part/#riftuidev micgro42 (~Michael@f052155065.adsl.alicedsl.de)
22:34.32*** join/#riftuidev seebs (~seebs@home.seebs.net)
22:34.33*** mode/#riftuidev [+o seebs] by ChanServ
22:50.46NosheiI kinda want to build a new unit frame addon, that uses canvas instead of a ton of masks.
22:54.38Nosheiand then I remember that I really have no clue how to contruct such a large program with as many parts as that would require.  At least not in an efficient and organized manner.
22:55.22Ivnedar_could you build it on top of gadgets?
22:56.46Nosheithats what I dont want to do, lot of bload in gadgets
22:57.00Nosheis/bload/bloat/
23:08.33Nosheiplus Ivnedar_ I want it to be super customizable.

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.