jfs

Sat 23:03:38 - * Joins: interactii (Yakushi_Ka@68-191-74-135.dhcp.oxfr.ma.charter.com)
Sat 23:03:39 - <@equinox> lol
Sat 23:03:43 - <interactii> equi!
Sat 23:03:44 - <@equinox> what are you trying to do
Sat 23:03:45 - <@jfs> i think i actually do know the real problem and the solution
Sat 23:03:55 - <@equinox> inty!
Sat 23:04:00 - <@jfs> but i can't get a clear view of an approach to use implementing it
Sat 23:04:01 - <interactii> can you send me the code for your fax fay patch
Sat 23:04:15 - <interactii> :D
Sat 23:04:16 - <@equinox> i don't even know if i have it anymore ^^;
Sat 23:04:39 - <interactii> See if you can find it for me :D
Sat 23:05:04 - * @equinox wakes his 40 GB disk from poweroff
Sat 23:05:26 - <interactii> Thank you ^____^
Sat 23:05:35 - <@jfs> actually interactii, consider whether you really do need it :)
Sat 23:05:49 - <interactii> Why wouldn't I?
Sat 23:05:53 - <@jfs> as RoRo correctly showed on the aegisub forum, you don't need it in most cases
Sat 23:06:04 - <interactii> orly? link me
Sat 23:06:11 - <@jfs> if you have to typeset a sign with a "strange" rotation
Sat 23:06:13 - <interactii> I mean I'm sure it can be replicated
Sat 23:06:20 - <interactii> but at tremendous labor
Sat 23:06:23 - <@jfs> calculate the vanishing point for the lines around the sign
Sat 23:06:28 - <@jfs> and place the \org there
Sat 23:06:30 - <interactii> the fax and fay are much easier
Sat 23:06:35 - <@equinox> uhm
Sat 23:06:44 - <@jfs> then you can \pos and rotate it correctly
Sat 23:06:46 - <@equinox> jfs: the whole point of fax and fay is that there is no vanishing point
Sat 23:06:50 - <@jfs> i know
Sat 23:06:58 - <@jfs> but it's usually used for cases like there
Sat 23:07:00 - <@jfs> these
Sat 23:07:03 - <interactii> yes but doing those calulations
Sat 23:07:11 - <interactii> is somewhat complex./
Sat 23:07:19 - <@jfs> it could be completely automated
Sat 23:07:54 - <@jfs> perhaps even the rotation-calculations
Sat 23:08:21 - <interactii> sure. But :D
Sat 23:09:23 - <@equinox> found the patch.
Sat 23:09:31 - <interactii> Yay!
Sat 23:10:44 - <@equinox> http://celeste.diac24.net/textsub_fax_fay.patch
Sat 23:10:46 - <@equinox> against 2.23.
Sat 23:12:09 - <interactii> Thank you very much ^_^
Sat 23:12:56 - <@equinox> now if my found my guliverkli checkout, i could create an updated version ^^;
Sat 23:14:26 - <interactii> :D
Sat 23:17:07 - <@equinox> grr, stupid svn
Sat 23:20:20 - <Kanzaki_Urumi> hum, fax?
Sat 23:20:58 - <@jfs> fax/fay are x/y shear
Sat 23:21:04 - <Kanzaki_Urumi> oh
Sat 23:21:13 - <Kanzaki_Urumi> semi-cute
Sat 23:21:37 - <@TheFluff> jfs: vanishing point of the lines around the sign?
Sat 23:21:50 - <@TheFluff> please explain for someone who doesn't grok 3d
Sat 23:22:00 - <@TheFluff> +geometry
Sat 23:22:23 - <@jfs> the traditional pen+paper method of drawing 3d images
Sat 23:22:30 - <@jfs> you decide on a vanishing point
Sat 23:22:57 - <@jfs> parallel lines going away from you meet at the vanishing point in the drawing
Sat 23:23:06 - <@equinox> inty: don't apply the patch on SVN, it horribly misapplies
Sat 23:24:09 - <interactii> Yea, its just so I know the techiuqie you used :D
Sat 23:24:40 - <@TheFluff> jfs: ah
Sat 23:25:51 - <Kanzaki_Urumi> the most useful change would def. be changinge bool fBlur to float fBlur
Sat 23:25:54 - <Kanzaki_Urumi> -e
Sat 23:26:06 - <@jfs> blur edges?
Sat 23:26:13 - <@jfs> and slow down things even more? :)
Sat 23:26:26 - <Kanzaki_Urumi> wont slow much:)
Sat 23:26:49 - <interactii> mmm
Sat 23:26:51 - <interactii> floating blur.
Sat 23:26:54 - <interactii> that would rock my socks.
Sat 23:26:55 - <@jfs> i'm not even sure if \be is actually implemented as a traditional blur or not
Sat 23:27:02 - <Kanzaki_Urumi> no idea:)
Sat 23:27:20 - <Kanzaki_Urumi> i wish there was a proper way to have shadows alpha-ed out by main/edge too
Sat 23:27:33 - <Kanzaki_Urumi> i'm forced to resort to doing shadow fonts because of that:)
Sat 23:30:01 - <@equinox> ouch.
Sat 23:32:55 - <@equinox> http://celeste.diac24.net/textsub/
Sat 23:33:03 - <@equinox> updated version for current SVN
Sat 23:34:53 - * Quits: Fictive-san (~fictive-s@216-224-151.0502.adsl.tele2.no) (QUIT:  bleh.)
Sat 23:36:52 - <Jeroi> ?
Sat 23:37:10 - <@equinox> actually. *moves that stuff to the asa page*
Sat 23:38:31 - <Kanzaki_Urumi> is libass going anywhere btw?
Sat 23:38:57 - <@amz> hmm
Sat 23:39:01 - <@amz> where's my MSVC DVD
Sat 23:39:16 - <@equinox> DW: apparently, yes
Sat 23:39:34 - <Kanzaki_Urumi> it was still missing a few key features last i heard
Sat 23:39:44 - <Kanzaki_Urumi> and it def. needs a dshow/avs port:)
Sat 23:39:45 - <@equinox> no idea
Sat 23:40:18 - <@equinox> actually. *compiles and tests*
Sat 23:41:39 - <@amz> if it builds on MSVC++, an avs port is easy
Sat 23:41:51 - <@amz> but that's a big if
Sat 23:42:10 - <@amz> it's probably not pure ANSI-C, or using c99 extensions
Sat 23:42:13 - <@equinox> yay. the libass CVS on sf.net is empty
Sat 23:42:20 - <@equinox> its probably still in the mplayer SVN
Sat 23:42:44 - <@equinox> or has no public repo like libssa
Sat 23:43:24 - <@equinox> configure.ac:1: error: Autoconf version 2.60a or higher is required
Sat 23:43:28 - <@equinox> hooray.
Sat 23:43:49 - <@equinox> and it ships non-autogen'ed
Sat 23:45:30 - <@jfs> hmm... i think i might have it nailed now...
Sat 23:45:32 - <@jfs> almost :)
Sat 23:46:20 - <@amz> hmm
Sat 23:46:25 - <@amz> should I install Microsoft SQL Server?
Sat 23:46:30 - <@equinox> EEEK
Sat 23:46:35 - <@jfs> for what?
Sat 23:46:39 - <@amz> dunno
Sat 23:46:44 - <@amz> just in case I need an SQL server one day
Sat 23:46:47 - <@amz> yea, wtf, nevermind that.
Sat 23:47:19 - <@amz> only did C++, C# and assorted tools
Sat 23:47:30 - <@equinox> C#?
Sat 23:47:35 - <@amz> C#.
Sat 23:47:41 - <@amz> I want to learn it someday.
Sat 23:47:48 - <@jfs> i also installed VB.NET for some reason...
Sat 23:47:51 - <@equinox> wtf.
Sat 23:47:58 - <@equinox> autoconf (GNU Autoconf) 2.60
Sat 23:48:00 - <@amz> yea, VB can go die a horrible death >_>
Sat 23:48:24 - <Kanzaki_Urumi> 2.60a!
Sat 23:48:25 - <Kanzaki_Urumi> :)
Sat 23:48:26 - <@equinox> GRAH
Sat 23:48:27 - <@jfs> vb.net is at least more sensible than vb 6
Sat 23:48:28 - <@equinox> yeah.
Sat 23:48:31 - <@equinox> and guess what
Sat 23:48:34 - <@jfs> the vb 6 object model is... *pukes*
Sat 23:48:38 - * Joins: D4RK-PH0ENiX (~I.Love.NH@host-86-106-219-248.moldtelecom.md)
Sat 23:48:40 - <@equinox> 2.60a is only available on alpha.gnu.org
Sat 23:50:14 - <@amz> so it needs an alpha autoconf to install?
Sat 23:50:17 - <@amz> how nice of it >_>
Sat 23:50:18 - <@equinox> yes.
Sat 23:50:24 - <@amz> can you have both installed?
Sat 23:50:31 - <Kanzaki_Urumi> maybe you can modify it?
Sat 23:50:33 - <@equinox> with hacks, yes.
Sat 23:50:43 - <Kanzaki_Urumi> did you try removing the check for autoconf?
Sat 23:50:43 - <@equinox> i have some stale memory of trying and failing with 2.59b
Sat 23:50:48 - <@equinox> uh
Sat 23:50:56 - <Kanzaki_Urumi> it might work:)
Sat 23:51:00 - <@equinox> DW: autoconf is the set of macros that *generates* the checks ;)
Sat 23:51:26 - <Kanzaki_Urumi> just remove whatever fails it then;)
Sat 23:51:37 - <@equinox> already updated to 2.60b
Sat 23:51:42 - <Kanzaki_Urumi> haha:)
Sat 23:51:46 - <@equinox> take that, bitch, i got a newer version! :PPP
Sat 23:51:50 - <@equinox> *g*
Sat 23:52:18 - <Kanzaki_Urumi> configure.ac:1: error: Autoconf version 2.60a or lower is required
Sat 23:52:19 - <@equinox> that'll be the day when i do something a software tells me to :D
Sat 23:52:26 - <Kanzaki_Urumi> ;)
Sat 23:52:28 - <@equinox> haha
Sat 23:52:51 - <@equinox> it compiles fine.
Sat 23:53:48 - <@jfs> wow... now i'm puzzled :(
Sat 23:54:15 - <Kanzaki_Urumi> it compiles:/ this is sad
Sat 23:54:34 - <@jfs> i figured out what to do about during-script dialogs
Sat 23:54:46 - <Kanzaki_Urumi> on a side note... did gabest do anything in the past 2 months?
Sat 23:54:51 - <@jfs> but the seeming simpler (?) case of pre-config dialogs for export is still unsolved
Sat 23:54:51 - <@equinox> 5383 lines of code... i somehow feel assed for my 7249 lines for asa, which don't even make it work...
Sat 23:55:06 - <Kanzaki_Urumi> blame freetype
Sat 23:55:08 - <Kanzaki_Urumi> or actually
Sat 23:55:10 - <Kanzaki_Urumi> blame movax
Sat 23:55:10 - <Kanzaki_Urumi> better
Sat 23:55:15 - <@equinox> libass uses freetype too -_-
Sat 23:55:24 - <Kanzaki_Urumi> i know:)
Sat 23:55:24 - <@equinox> and yeah, its movax's fault anyway
Sat 23:55:41 - <Kanzaki_Urumi> but the freetype reference is about as clear as shouyu
Sat 23:55:44 - <@equinox> let's check what it supports
Sat 23:56:07 - <@jfs> how about doing some benchmarking? :)
Sat 23:56:15 - <Kanzaki_Urumi> that too:)
Sat 23:56:32 - <@equinox> later. conformance first :D
Sat 23:56:52 - <@jfs> i think libass still has the sizing problems though
Sat 23:56:58 - <@jfs> the same as automation does :P
Sat 23:57:13 - <@jfs> (yeah, automation in libass compliant but not vsfilter compliant ;)
Sat 23:57:36 - <@equinox> its not libass and auto having the issues
Sat 23:57:42 - <@equinox> its vsfilter.
Sat 23:57:51 - <@equinox> and that's no joke or sarcasm or whatever
Sat 23:59:16 - <@TheFluff> bug-for-bug compatible
Sat 23:59:50 - <@jfs> how about just redefining what's "right"? :/
Sat 23:59:58 - <@jfs> and make everyone use the "right" thing
Sun 00:00:10 - <@jfs> eg. put a libass-based thingy into CCCP
Sun 00:00:11 - * Quits: @equinox (~equinox@88.134.102.206) (Connection reset by peer)
Sun 00:00:19 - <@TheFluff> heh
Sun 00:00:36 - * Joins: equinox (~equinox@88.134.102.206)
Sun 00:00:36 - * ChanServ sets mode: +o equinox
Sun 00:00:39 - <@TheFluff> trust the fansubbers not to adapt
Sun 00:01:21 - <@equinox> haha
Sun 00:01:30 - <@equinox> the included testing app screws up the PNG image it is supplied
Sun 00:03:15 - <@equinox> http://asa.diac24.net/libass/
Sun 00:03:36 - <@jfs> lol?
Sun 00:03:41 - <@equinox> don't ask me.
Sun 00:04:24 - <@equinox> maybe it doesn't support B/W pngs
Sun 00:04:33 - <@equinox>   Colorspace: Gray
Sun 00:04:34 - <@equinox>   Channel depth:
Sun 00:04:34 - <@equinox>     Gray: 1-bit
Sun 00:04:50 - <@jfs> doesn't it use libpng?
Sun 00:04:55 - <@equinox> it does.
Sun 00:04:57 - * Joins: demi` (demi@201.79.247.98)
Sun 00:05:01 - <@equinox> which means the programmer fails it.
Sun 00:05:06 - <@jfs> i think libpng resamples everything to 24 or 32 bit on load... or can do it
Sun 00:05:14 - <@equinox> yes, its a single line of code.
Sun 00:06:31 - <Kanzaki_Urumi> well whatever:)
Sun 00:06:34 - <Kanzaki_Urumi> just use a standard png:)
Sun 00:07:44 - <@jfs> hmm i'm out for some time... need to stock up on snacks :s
Sun 00:09:13 - <@equinox> uh
Sun 00:09:31 - <Kanzaki_Urumi> ?
Sun 00:09:35 - <@equinox> it screws color images too
Sun 00:09:40 - <Kanzaki_Urumi> fun:)
Sun 00:10:04 - <Kanzaki_Urumi> but wait... isnt it supposed to do an alpha directly on the pic?
Sun 00:10:19 - <@equinox> ARGH
Sun 00:10:25 - <@equinox> it doesn't read the input image at all
Sun 00:10:39 - <@jfs> heh, explains? :)
Sun 00:10:45 - <@equinox> kinda. -___-
Sun 00:11:16 - <Kanzaki_Urumi> hahhahahah
Sun 00:11:43 - <Kanzaki_Urumi> if you know libpng, should be just a min to hardcode it in
Sun 00:11:51 - <@equinox> yeah
Sun 00:11:52 - <@equinox> but
Sun 00:11:53 - <@equinox> screw it.
Sun 00:12:01 - * @equinox just tests stuff as it is
Sun 00:12:05 - <Kanzaki_Urumi> ya:)
Sun 00:13:39 - <@equinox> uh
Sun 00:13:52 - <@equinox> i think that guy got the color codes wrong.
Sun 00:14:01 - <@equinox> it was &HBBGGRR&, right?
Sun 00:14:06 - <Kanzaki_Urumi> yeah
Sun 00:14:12 - <Kanzaki_Urumi> he did RRGGBB?
Sun 00:14:14 - <@equinox> haha.
Sun 00:14:17 - <@equinox> yes. FAIL.
Sun 00:14:29 - <Kanzaki_Urumi> that's easy to fix:)
Sun 00:14:35 - <@equinox> anyway. FAIL. :P
Sun 00:14:36 - <Kanzaki_Urumi> about 20 chars in sed or so:)
Sun 00:14:55 - <Kanzaki_Urumi> now please go on, i'm more interested into the actual stuff;)
Sun 00:15:37 - * Quits: Andy|away (UPP@HIDDEN-7715A1DC.user.veloxzone.com.br) (Ping timeout)
Sun 00:16:04 - <@equinox> yah yah -_-
Sun 00:16:11 - <niblifar> ooo.
Sun 00:16:14 - <niblifar> Hi.
Sun 00:17:34 - <@equinox> err
Sun 00:17:40 - <@equinox> i think he got \an wrong too
Sun 00:18:09 - <@equinox> ah, no.
Sun 00:18:14 - <Kanzaki_Urumi> :)
Sun 00:18:25 - <@equinox> been too long since i TSed :(
Sun 00:18:32 - <Kanzaki_Urumi> wait!
Sun 00:18:38 - <Kanzaki_Urumi> seeing how BBGGRR got inverted
Sun 00:18:41 - <Kanzaki_Urumi> maybe it's a GIRL
Sun 00:19:02 - <Kanzaki_Urumi> no sane ssa/ass user would invert it, other than a girl:)
Sun 00:19:03 - <@equinox> . . .
Sun 00:19:07 - <Kanzaki_Urumi> okay :D
Sun 00:19:32 - <@TheFluff> wtf
Sun 00:19:55 - <@equinox> hmm
Sun 00:20:01 - <@equinox> \bord affects \shad in libass
Sun 00:20:35 - <Kanzaki_Urumi> that's actually not such a bad idea:)
Sun 00:20:47 - <@equinox> but its not compatible.
Sun 00:21:09 - <Kanzaki_Urumi> it's minor;) i mean... no sane one relies on that anyway:)
Sun 00:21:21 - * Kanzaki_Urumi just cares about functionalities being present
Sun 00:21:59 - * Quits: luizhpga (luizhpga@HIDDEN-27E360E9.user.veloxzone.com.br) (QUIT:  :: Restarting :: ( Ultra Protection Project ))
Sun 00:22:46 - <@equinox> \be is present
Sun 00:22:52 - <Kanzaki_Urumi> that's good
Sun 00:22:55 - <Kanzaki_Urumi> now
Sun 00:23:06 - <Kanzaki_Urumi> what about all the \org \frz \fry \frx
Sun 00:24:08 - <@equinox> 45° should be / right?
Sun 00:24:28 - <Kanzaki_Urumi> isnt it \
Sun 00:24:37 - <@equinox> try it?
Sun 00:24:40 - <Kanzaki_Urumi> okay
Sun 00:25:07 - <@equinox> neither frx nor fry are present
Sun 00:25:35 - <@amz> the manual has pics showing it
Sun 00:25:53 - <Kanzaki_Urumi> it's / indeed
Sun 00:26:01 - <@equinox> ok, frz working then
Sun 00:26:28 - <@equinox> \org working too
Sun 00:26:50 - <@amz> what are you doing?
Sun 00:26:57 - <Kanzaki_Urumi> a test of libass
Sun 00:27:05 - <@amz> ah
Sun 00:27:12 - <Kanzaki_Urumi> equinox compiled it and is testing the lib's support/compatibility
Sun 00:27:21 - <@equinox> fsc* working too
Sun 00:27:39 - <Kanzaki_Urumi> what abgout fry frx?:)
Sun 00:27:47 - <@equinox> <@equinox> neither frx nor fry are present
Sun 00:27:52 - <Kanzaki_Urumi> oh ok
Sun 00:28:01 - <Kanzaki_Urumi> i inverted you with amz:)
Sun 00:28:13 - <Jeroi> equinox?
Sun 00:28:13 - <@amz> lavc video provider still needs a lot of work, though
Sun 00:28:17 - <@amz> e.g. making mkv work
Sun 00:28:19 - <Jeroi> what are you doing?
Sun 00:28:25 - <Jeroi> what is libass?
Sun 00:28:32 - <@amz> nothing you should be concerned about
Sun 00:28:33 - <@equinox> mplayer's ass renderer
Sun 00:28:39 - <@TheFluff> lol
Sun 00:28:48 - <@TheFluff> I was just about to kickban him with what amz said as reason
Sun 00:29:06 - <@equinox> rect-\clip working too
Sun 00:29:24 - <Kanzaki_Urumi> cool
Sun 00:29:37 - <Kanzaki_Urumi> so just frx/fry and inverted colors so far:)
Sun 00:29:45 - <Kanzaki_Urumi> frx/fry are pretty much essential though:/
Sun 00:29:46 - <@amz> eh
Sun 00:29:50 - <@amz> they got the colour order wrong?
Sun 00:29:51 - <@amz> -_-
Sun 00:29:52 - <Kanzaki_Urumi> yeah
Sun 00:29:54 - <@equinox> yes.
Sun 00:29:55 - <Kanzaki_Urumi> RRGGBB
Sun 00:30:03 - <@amz> 21:18:32 <Kanzaki_Urumi> maybe it's a GIRL
Sun 00:30:03 - <@amz> 21:18:53 <Kanzaki_Urumi> no sane ssa/ass user would invert it, other than a girl:)
Sun 00:30:05 - <@amz> lol
Sun 00:30:07 - <@movix> well a top of the RRGGBB to you too sir
Sun 00:30:09 - * Quits: demi` (demi@201.79.247.98) (Ping timeout)
Sun 00:30:17 - * Kanzaki_Urumi blames movix again
Sun 00:30:29 - * @movix makes a BBRRGG 
Sun 00:30:33 - * @movix makes a BBRRGG noise*
Sun 00:30:53 - * Kanzaki_Urumi burries movix in salt
Sun 00:31:09 - * Joins: demi` (demi@201.79.218.35)
Sun 00:31:12 - <@jfs> how about \k and variations?
Sun 00:31:13 - <@equinox> \move working
Sun 00:31:29 - <Kanzaki_Urumi> \t on colors!
Sun 00:31:36 - <Kanzaki_Urumi> beri important
Sun 00:31:53 - <@amz> it's funny that \frx and y aren't working
Sun 00:31:59 - <@amz> they aren't hard to implement, the vsfilter way
Sun 00:32:01 - <@equinox> frx and fry are stupid
Sun 00:32:08 - <@amz> iirc, the code that does their transform is really simple
Sun 00:32:12 - <@equinox> yes.
Sun 00:32:23 - <Kanzaki_Urumi> equinox, if you have shear code, then yes:)
Sun 00:32:36 - <@equinox> it uses a rotation point with coordinates in the 10000s
Sun 00:32:37 - <Kanzaki_Urumi> shear+fscx/fscy that is
Sun 00:32:42 - <@equinox> well
Sun 00:32:48 - <@amz> yea, ASS should support shearing by default >:(
Sun 00:32:53 - <@equinox> a \fmatrix(xx,xy,yx,yy) would be neat
Sun 00:33:03 - <Kanzaki_Urumi> haha, yeah:) but dont dream;)
Sun 00:33:05 - <@amz> but polygon positioning specified in AS5 is more powerful
Sun 00:33:17 - <@amz> have you read the AS5 "specs", DW?
Sun 00:33:48 - <@equinox> \k and \K working.
Sun 00:33:53 - <Kanzaki_Urumi> i saw a few things, but only part of it
Sun 00:34:06 - <Kanzaki_Urumi> and as long as it doesnt have a windows&*nix renderer, i dont care about it:(
Sun 00:34:16 - <Kanzaki_Urumi> equinox, do fonts in script work?
Sun 00:34:26 - <@equinox> let me test \t first
Sun 00:34:28 - <@amz> they don't even work in vsfilter...
Sun 00:34:30 - <@amz> so, meh.
Sun 00:34:53 - <@amz> and mkvmerge is broken there
Sun 00:34:56 - <@amz> it only merges some fonts
Sun 00:35:05 - <@amz> I forgot how exactly the bug was manifest
Sun 00:35:14 - <Kanzaki_Urumi> they do work in vsfilter o.O just that they tend to break most of the time:)
Sun 00:35:37 - * Quits: ender (~ender@84.52.165.220) (QUIT:   A computer program will always do what you tell it to, and seldom what you want it to.)
Sun 00:35:38 - <@amz> well, hmm, I don't remember what were the results of my tests
Sun 00:35:45 - <@equinox> \1c&HFF0000&\t(\1c&H0000FF&) works
Sun 00:35:53 - <@equinox> (aside from reversing the colors)
Sun 00:36:10 - <@amz> I still can't believe that they managed to reverse the colors.
Sun 00:36:22 - <@amz> do they work in styles, at least?
Sun 00:36:27 - <@equinox> no.
Sun 00:36:31 - <@equinox> you know what's worse?
Sun 00:36:43 - <@equinox> the horde of mplayer users watching softsubs and wondering why they are so ugly :D
Sun 00:36:55 - <@amz> -____-
Sun 00:36:59 - <@amz> did you fill a bug report?
Sun 00:37:03 - <@equinox> not yet.
Sun 00:37:05 - <@amz> It can't possibly be any hard to fix
Sun 00:37:12 - <@equinox> yeah.
Sun 00:37:20 - <niblifar> how recent is that build?
Sun 00:37:26 - <@equinox> 0.9.1
Sun 00:37:31 - <@equinox> tarball, no SVN
Sun 00:37:35 - <niblifar> Ah.
Sun 00:37:38 - <@jfs> but making such an error imo shows that the implementor has never actually used ASS...
Sun 00:37:38 - <Kanzaki_Urumi> [00:35:41] <@equinox> \1c&HFF0000&\t(\1c&H0000FF&) works => sounds awesome?:)
Sun 00:37:44 - <Kanzaki_Urumi> now the question is
Sun 00:37:49 - <Kanzaki_Urumi> is it as slow as vsfilter on that:)
Sun 00:37:59 - <@equinox> heh.
Sun 00:38:03 - <@amz> \t(\alpha) on vsfilter = zomgslow
Sun 00:38:13 - <@equinox> actually, lemme test alpha
Sun 00:38:18 - <Kanzaki_Urumi> \t(\anythingcolororalpharelated) = zomgslow
Sun 00:38:23 - <Kanzaki_Urumi> and zomgslow=at least 10 to 20x
Sun 00:38:26 - <Kanzaki_Urumi> (slower)
Sun 00:38:46 - <@equinox> hmm
Sun 00:38:59 - <@equinox> there's some blending bug there definitively
Sun 00:39:22 - <@equinox> (with plain borders, no animations in use)
Sun 00:39:32 - <Kanzaki_Urumi> no proper alpha-ing?
Sun 00:39:32 - <@amz> ?
Sun 00:39:34 - <@amz> like what?
Sun 00:39:39 - <@amz> It's not like vsfilter itself does it properly
Sun 00:39:42 - <@equinox> one sec
Sun 00:40:00 - <niblifar> Hmm, that build is recent as of Nov. 3.
Sun 00:40:11 - <@equinox> http://asa.diac24.net/libass/blendbug.png
Sun 00:40:26 - * Joins: luizhpga (luizhpga@HIDDEN-27E360E9.user.veloxzone.com.br)
Sun 00:40:26 - <@amz> eeew
Sun 00:40:37 - <@amz> are you trying it on aegisub?
Sun 00:40:42 - <@equinox> no.
Sun 00:40:45 - <Kanzaki_Urumi> it's not that bad o.O
Sun 00:40:47 - <@amz> have you?
Sun 00:40:51 - <@equinox> no.
Sun 00:40:56 - <Kanzaki_Urumi> unless... it is not supposed to look like that at all?
Sun 00:40:58 - <@amz> ok, let me know if it works
Sun 00:41:01 - <@amz> azzy's patch, that is
Sun 00:41:27 - <@equinox> DW: it is supposed to look like this, just the thin extra borders in the light gray border are off
Sun 00:42:06 - <@equinox> border and character probably have a little alpha inbetween them
Sun 00:47:27 - <@equinox> no font embedding apparently.
Sun 00:47:42 - <@jfs> [font] sections?
Sun 00:47:47 - <@equinox> yep.
Sun 00:48:04 - <@jfs> that's also rarely used afaik?
Sun 00:48:09 - <Kanzaki_Urumi> http://www.addictinggames.com/theidiottest.html
Sun 00:48:13 - <Kanzaki_Urumi> warning, frustratingly stupid
Sun 00:48:55 - <@equinox> is that flash or why don't i see anything?
Sun 00:49:05 - <Kanzaki_Urumi> yes
Sun 00:49:05 - <Kanzaki_Urumi> flash
Sun 00:49:21 - <@equinox> yay. saved from stupidness again.
Sun 00:49:41 - <@equinox> anything else i shall test?
Sun 00:50:22 - <@amz> yay, lost at the "lightest blue rectangle" thing
Sun 00:50:25 - <@amz> >_>
Sun 00:50:27 - * @amz won't try again
Sun 00:50:42 - <@equinox> [ass] **MSGL_V**: fontname: Estrangelo Edessa
Sun 00:50:42 - <@equinox> [ass] **MSGL_V**: font: 106326 bytes encoded data
Sun 00:50:43 - <@equinox> hmm
Sun 00:50:44 - <@equinox> strange
Sun 00:50:55 - <@equinox> ass] **MSGL_WARN**: fontconfig: selected font family is not the requested one: 'Arial' != 'Estrangelo Edessa'
Sun 00:50:56 - <@equinox> [ass] **MSGL_V**: fontconfig_select: (Estrangelo Edessa, 80, 0) -> /home/equinox/.fonts/arial.ttf, 0
Sun 00:51:59 - <Kanzaki_Urumi> boom .-.
Sun 00:52:35 - <Kanzaki_Urumi> http://www.zeronews-fr.com/flash/capnzappy.php
Sun 00:53:01 - * @jfs passed the test
Sun 00:53:15 - <Kaverin> heh
Sun 00:53:30 - <Kaverin> when it wanted you to press the smallest red button, I never noticed one in the upper right, so I failed the first time
Sun 00:53:43 - <Kaverin> but then I passed it all the second time
Sun 00:54:03 - <@jfs> equinox how does it handle \pos, in a way so it works with \t ?
Sun 00:54:08 - <@jfs> how about \t(\clip) ?
Sun 00:54:30 - <@jfs> \t with 2, 3 and 4 parameters
Sun 00:56:02 - <@jfs> http://www.edinburghonline.co.uk/info/7621/ <- libass
Sun 00:56:19 - <@jfs> is a fashion shop in edinburgh
Sun 00:57:58 - <@equinox> \t(\pos()) executes the contained \pos instead of animating it
Sun 00:58:18 - <@jfs> hmm, how about multiple \pos in a line?
Sun 00:58:35 - <@jfs> {\pos(10,10)}foo{\pos(100,200)}bar
Sun 00:59:07 - * Quits: iVoRy (~iVoRy@ppp-124.120.94.246.revip2.asianet.co.th) (Ping timeout)
Sun 00:59:42 - <@equinox> \t(\clip()) - OK
Sun 01:00:00 - <@equinox> \t(x,\clip()) - FAIL
Sun 01:00:37 - <@equinox> \t(x,x,\clip()) - OK
Sun 01:00:58 - <@equinox> \t(x,x,x,\clip()) - FAIL
Sun 01:01:04 - <@equinox> apparently it doesn't support the accel value
Sun 01:02:01 - <@equinox> yep.
Sun 01:02:03 - <@equinox> no accel value.
Sun 01:02:11 - <@jfs> ... why not >_>
Sun 01:02:15 - <@jfs> it should be real simple
Sun 01:02:22 - <@equinox> don't ask me.
Sun 01:02:25 - <@equinox> asa supports it. :P
Sun 01:02:44 - <@equinox> asa supports most of the complicated stuff, but fails at the easy :D
Sun 01:03:01 - <@jfs> how about \frz90\kf
Sun 01:03:02 - * Quits: @equinox (~equinox@88.134.102.206) (Connection reset by peer)
Sun 01:03:06 - <@jfs> >_>
Sun 01:03:09 - <@jfs> STOP THAT PEER
Sun 01:03:29 - * Joins: equinox (~equinox@88.134.102.206)
Sun 01:03:29 - * ChanServ sets mode: +o equinox
Sun 01:03:31 - <@equinox> the last of multiple \pos overrides the previous
Sun 01:03:34 - <@jfs> how about \frz90\kf
Sun 01:03:38 - <@equinox> and ChatSociety likes to reset me
Sun 01:03:42 - <@jfs> which direction does it fill in?
Sun 01:03:56 - <@amz> http://uncyclopedia.org/wiki/Peer
Sun 01:04:06 - <@jfs> (vsfilter always does strict left to right, independant of rotation)
Sun 01:04:20 - * Joins: iVoRy (~iVoRy@ppp-124.120.94.246.revip2.asianet.co.th)
Sun 01:04:23 - <@equinox> haha
Sun 01:04:28 - <@equinox> character-wise left to right
Sun 01:05:00 - <@equinox> http://asa.diac24.net/libass/frz90-K.png
Sun 01:05:07 - <@amz> * ProducePete Quit (Read error: Connection reset by peer)
Sun 01:05:08 - <@amz> * ProducePete has joined #hehe
Sun 01:05:08 - <@amz> * ProducePete Quit (Read error: Connection reset by peer)
Sun 01:05:08 - <@amz> * ProducePete has joined #hehe
Sun 01:05:08 - <@amz> * ProducePete Quit (Read error: Connection reset by peer)
Sun 01:05:08 - <@amz> * ProducePete has joined #hehe
Sun 01:05:08 - <@amz> <4rni3> you're going from p2p haha
Sun 01:05:08 - <@amz> <ProducePete> stfu
Sun 01:05:08 - <@amz> * ProducePete Quit (Read error: Connection reset by peer)
Sun 01:05:08 - <@amz> lol
Sun 01:05:48 - <@equinox> ("\K2000 abc \K2000 def" in that frz90-K.png) 
Sun 01:08:00 - <Kanzaki_Urumi> hahahah
Sun 01:08:22 - <Kanzaki_Urumi> equinox, SOOOOO
Sun 01:08:25 - <Kanzaki_Urumi> when do we get asa
Sun 01:08:29 - <Kanzaki_Urumi> and an avs filter
Sun 01:08:33 - <Kanzaki_Urumi> and a dshow filter
Sun 01:08:44 - <Kanzaki_Urumi> you can ignore vdub, no one cares;)
Sun 01:08:54 - <@jfs> http://uncyclopedia.org/wiki/Alt-F4
Sun 01:09:00 - <@equinox> asa actually has avs support ^^;
Sun 01:09:14 - <Kanzaki_Urumi> cool:)
Sun 01:09:18 - <Kanzaki_Urumi> so what does it fail?
Sun 01:09:32 - <@equinox> it has no positioning / wrapping code at all.
Sun 01:09:34 - <@equinox> like
Sun 01:09:37 - <@equinox> everything is at (0,0)
Sun 01:09:57 - <Kanzaki_Urumi> yeah i remember reading
Sun 01:10:00 - <Kanzaki_Urumi> that it has no wrapping
Sun 01:10:41 - <@equinox> i really need to sketch out an algorithm for it on paper
Sun 01:10:48 - <@equinox> or play with it in python.
Sun 01:11:12 - <Kanzaki_Urumi> does frx/frz/fry work in asa?
Sun 01:11:19 - <@equinox> frz only.
Sun 01:11:35 - <@equinox> http://asa.diac24.net/Compatibility
Sun 01:11:54 - <Kanzaki_Urumi> development status STALLED
Sun 01:12:03 - <@equinox> yes.
Sun 01:12:11 - <@equinox> last change was in july
Sun 01:12:20 - <Kanzaki_Urumi> collision detection => useless;) ignore it
Sun 01:12:28 - <@equinox> well
Sun 01:12:32 - <@equinox> you *need* collision detection
Sun 01:12:42 - <Kanzaki_Urumi> what for? it's gay:)
Sun 01:12:49 - <@equinox> two lines on-screen *does* happen from time to time...
Sun 01:12:50 - <Kanzaki_Urumi> just force positioning;)
Sun 01:13:08 - <Kanzaki_Urumi> PlayRes handling => broken:/?
Sun 01:13:24 - <@equinox> yeah, it just assumes 1:1 pixel mapping for now
Sun 01:13:26 - <Kanzaki_Urumi> \be missing is evil though! get it in:(
Sun 01:13:41 - <Kanzaki_Urumi> even if it's just as a simple convolution matrix:)
Sun 01:13:45 - <@equinox> implement me a fast algorithm :P
Sun 01:13:48 - <Kanzaki_Urumi> (aka non compatible but whatever:))
Sun 01:13:57 - <Kanzaki_Urumi> well just a fucking blur, that's enough:)
Sun 01:14:07 - <Kanzaki_Urumi> having to do diamonds to do blurs is painful enough:)
Sun 01:14:16 - <@equinox> implement me a fast algorithm :PP
Sun 01:14:35 - <Kanzaki_Urumi> [[1;1;1][1;1;1][1;1;1]]/9
Sun 01:14:35 - <@equinox> i accept C, C++ and x86 & AMD64 assembler :P
Sun 01:14:37 - <Kanzaki_Urumi> there you go:)
Sun 01:15:06 - <Kanzaki_Urumi> no \k:(
Sun 01:15:17 - * Joins: massi-san (~massive@dslb-084-059-032-124.pools.arcor-ip.net)
Sun 01:15:25 - <Kanzaki_Urumi> what about \shad?
Sun 01:15:35 - <@equinox> just didn't do that yet
Sun 01:15:49 - <@equinox> but it'll be stencil-blended
Sun 01:15:58 - <Kanzaki_Urumi> goooooooooood!
Sun 01:16:05 - <Kanzaki_Urumi> that's exactly what i was saying earlier:)
Sun 01:16:10 - <Kanzaki_Urumi> (that vsfilter needs)
Sun 01:16:36 - <Kanzaki_Urumi> need dshow though :)
Sun 01:16:36 - <@equinox> well, it was kind of stupid to make bord stencil-blended and shad not, so :D
Sun 01:16:48 - <@equinox> movax said he was looking into that
Sun 01:16:49 - <Kanzaki_Urumi> well i blame gabest
Sun 01:16:53 - <Kanzaki_Urumi> actually
Sun 01:16:55 - <Kanzaki_Urumi> i blame movax
Sun 01:16:56 - <@equinox> so it really is movax' fault this time :D
Sun 01:16:59 - <Kanzaki_Urumi> :D
Sun 01:17:23 - * Quits: mASSIVe (~massive@dslb-084-059-033-128.pools.arcor-ip.net) (Ping timeout)
Sun 01:19:35 - * @movix really is working >.>
Sun 01:19:48 - <@movix> little does my company know they pay me overtime to work on anime-related things...
Sun 01:19:49 - * @movix cackles
Sun 01:20:09 - <niblifar> How sinister.