--- Log opened Fri Mar 28 00:00:03 2008
--- Day changed Fri Mar 28 2008
00:00 < genete> I'll test it tomorrow.
00:03 < dooglus> me too
00:05 < genete> I need someone who sing me a beauty nursery rhyme close to my ear to get me slept
00:05 < genete> |O
00:05 < genete> one sheep.... two sheeps.... three sheeps....
00:05 < genete> ...
00:06 < genete> ... 98.2327.823 sheeps...
00:06 < genete> night! ;)
00:07 < factor> 99 bottles of beer on the wall
00:07 < genete> xD
00:07 -!- genete [n=Genete@84.122.42.180.dyn.user.ono.com] has quit ["Abandonando"]
00:07 < factor> take one down pass it around 98 bottles of beer on the wall
00:13 < dooglus> Tout ce que je voulais, c'etait un moment de bonheur!
00:15 < factor> sounds like a line from A Perfect Circle song.
00:15 < dooglus> I downloaded a video with burned-in French subtitles by mistake
00:15 < dooglus> that was one of them :)
00:20 < Yoyobuae> opps, didn't notice that bug on the patch, looking at it now
00:21 < Yoyobuae> hmm, I dooglus and I are trying to use the same thing for different purposes XD
00:22 < Yoyobuae> the valuedesc from a duck, it needs to point to the parameter controlled by the duck.  my patch checks wether the valuedesc points to a "Bline Vertex" convert
00:25 < Yoyobuae> but in Duckmatic::add_to_ducks(), in the code that create the ducks for blinepoints (inside "case ValueBase::TYPE_BLINEPOINT:"), sets the valuedesc to the blinepoint valuenode instead
00:25 < Yoyobuae> which in turn breaks my code XD
00:27 < Yoyobuae> I simply removed that for my code to work, but then the "Link blinepoint to bline" breaks XD
00:28 < factor> oh cool was thinking of that line "one moment of happiness" that you said googled youtube  and someone put the A Perfect Circle song to the Anime oh my goddess.
00:29 < factor> would like to do some tuts on anime when I get the animation stuff down.
00:35 < factor> afk be back to hack more on synfig
00:39 < Yoyobuae> factor: do you have any general ideas as to how animate anime characters in synfig?
00:42 < Yoyobuae> factor: I been trying to come up with ideas to do it myself, but still nothing that comes close to the simplicity/flexibility achieved by frame by frame animation
00:47 < Yoyobuae> I guess that it would be possible simply editing duck by duck, but then you might be eaten alive by swarms of angry ducks XD
00:50 < factor> nope
00:50 < factor> still have to get animation in general down
00:59 < Yoyobuae> oh, do you have experience using any other software? or synfig it the first? =)
00:59 < factor> just gimp or inkscape
00:59 < factor> no direct animation
00:59 < factor> blender but its 3d
01:00 < factor> love all the gfx progs on linux.
01:02 < factor> love the new ghost in the shell stuff.
01:02 < Yoyobuae> oh, i see
01:02 < factor> not much on remakes but they did a great job
01:03 < factor> but blender you can mix 2d and 3d
01:06 < Yoyobuae> i haven't seen blender myself
01:07 < Yoyobuae> just read about nodes on blender once out of curiosity =)
01:08 < factor> trying to find the ox animation 
01:08 < factor> good mix of 2d and 3d
01:17 < factor> cant find the rendered demo anywhere, sked in the blender channe;
01:18 < Yoyobuae> oh, that's too bad
01:27 < factor> http://www.youtube.com/watch?v=fsyfye_z9-g  video sequence editor
01:27 < factor> not as cool as the blacksmith one but shows the jest of it
01:35 < factor> but blender only manipulates the generated 2d stuff not really good at 2d but it can do it as well but that would be done in inkscape or synfig
01:35 < factor> then you just green screen the synfig animations to the blender enviroment.
01:36 < Yoyobuae> oh ok
01:37 < Yoyobuae> i heard some discussions that it would be posible to combine 2d from synfig with 3d from blender, or something like that
01:38 < factor> oh yeah
01:38 < factor> would be nifty
01:39 < Yoyobuae> but I still feel like synfig is missing something, to make character animations better/easier
01:39 < factor> it has a ways to go, hopefully I can help out on that
01:39 < factor> bones
01:39 < factor> would rock for animation and lots of ducks
01:40 < Yoyobuae> yeah that would be good
01:41 < Yoyobuae> what I'm worried is that character will look stiff
01:41 < factor> depends on the tween effect used
01:41 < factor> http://www.youtube.com/watch?v=DjOLgLz_X-Y  blender green screen
01:44 < Yoyobuae> dunno, I was considering doing some sort of combination, between the classic method (drawing lots of frames) and the computer generated method (bones, interpolation, etc)
01:47 < Yoyobuae> nice green screen
01:49 < Yoyobuae> blender is quite amazing =)
01:49 < factor> it has came a long ways
01:52 < factor> kudoes to them and the team
01:53 < Yoyobuae> hopefully synfig can get there, someday =)
01:53 < factor> when I first used it it really sucked and made funny landscape shapes with it , that was about it
01:53 < factor> yup
01:54 < factor> I was working with moon-- cant remember the name of the 3d editor at the time it is now definct.
01:54 < factor> it was really good but disappeared
01:55 < factor> http://www.moonlight3d.eu/cms/
01:55 < factor> looks like it has some life again
01:56 < factor> I worked with it back in 98 -99
02:00 < Yoyobuae> hmm 3d is good.  too much of a good thing IMHO
02:00 < factor> just different styles
02:01 < Yoyobuae> yeah 2d techniques are being left behind, somewhat
02:02 < Yoyobuae> that's one of the reasons I started searching for a 2d animation package some time ago
02:02 < factor> I dont think so
02:02 < factor> I think both and other are still alive
02:03 < factor> you have 2d - 3d   2and 3d mix as well as abstract forms
02:03 < factor> but yeah lots of 3d on the movie sets these days
02:03 < factor> or 2d 3d mix
02:03 < factor> ghost in the shell movie was 2 /3 d mix
02:04 < factor> innocence
02:04 < Yoyobuae> i dont think anime wil ever be fully 3d
02:04 < Yoyobuae> it's nature wont allow it
02:04 -!- pixelgeek [n=chatzill@c-71-59-140-184.hsd1.or.comcast.net] has joined #synfig
02:05 < Yoyobuae> one thing I would really like to see someday, 2d vector graphics on games =D
02:06 < Yoyobuae> but for now it seems like an impossible dream XD
02:07 < factor> http://www.dreamworks.com/trailers/gits2/gits2_trlr_qt_240.mov    ghost trailer 2/3d mix
02:07 < factor> loved it by the way
02:11 < factor> but I love 3d and 2d toon , but 2d are on the top of my list 3d are below them
02:19 < factor> but right now you can do high level anime style stuff on synfig
02:19 < factor> just have to figure it out the synfig way
02:20 < factor> will get there
02:28 < Yoyobuae> i guess so, but as a programmer I'm also interested in improving synfig to make the way a bit easier, if possible =)
02:48 < pixelgeek> factor: /me <3 teh blender greenscreen
02:48 < pixelgeek> Check out post 42 - http://blenderartists.org/forum/showthread.php?t=89288
02:49 < pixelgeek> One of my 'if I had unlimited time and knowledge' ideas is to port paprmh's plugin to synfig
03:00 < pixelgeek> pabs3: If you have more people with problems installing under Windows, point them at the video walkthrough - I posted a link on the download page.  
03:47 -!- rubikcube [n=kvirc@dslc-082-082-086-095.pools.arcor-ip.net] has quit ["KVIrc 3.2.6 Anomalies"]
04:38 < pixelgeek> dooglus: looking at the render code, this caught my eye
04:39 < pixelgeek> 	// Change the random seed, so that each thread has a different one
04:39 < pixelgeek> 	srand(mythread*20+threads+time(0));
04:40 < pixelgeek> what ranges do mythread and threads have?  If threads is >20 won't this random seed be non-unique?
04:40 < pixelgeek> (Apologies if this is a stupid question)
05:34 -!- pabs3 [n=pabs@d122-105-78-204.per9.wa.optusnet.com.au] has quit [Connection timed out]
05:41 -!- pixelgeek [n=chatzill@c-71-59-140-184.hsd1.or.comcast.net] has quit [Read error: 113 (No route to host)]
06:15 -!- pabs3 [n=pabs@d122-105-78-204.per9.wa.optusnet.com.au] has joined #synfig
07:06 -!- LinuxO [i=wbzrigy@201.248.156.142] has joined #synfig
07:55 -!- ZanQdo [n=Daniel@201.201.2.22] has joined #synfig
09:09 -!- Zelgadis [n=zelgadis@87.103.169.190] has joined #synfig
09:13 < Zelgadis> genete: When I do compositing of shadows, I make them all the same color (i.e. black). And all shadow layers I make transperent, leaving Blend Method = Composite.
09:13 < Zelgadis> genete: Same for highlights, but their color is white
09:16 < Zelgadis> genete: Also I exporting and linking colors of shadow and highlights for all regions.
09:17 < Zelgadis> genete: Amount values I exporting too. So I can easily tweak the look of my image according to lightness. (Shadow color could be not black ;) )
09:20 < Zelgadis> factor: IMHO there is no common principle to do animation in anime-style. Each animated scene must be previously analyzed and planned to define better structure. Structure of sif-file do a great impact on animation process. That's why good animation with some (bad) structure cannot be achieved at all.
09:21 -!- Zelgadis [n=zelgadis@87.103.169.190] has quit ["Schastlivo!"]
09:26 < pabs3> http://people.imendio.com/richard/archives/2008/03/gtk_application.html
09:26 < pabs3> excellente
09:41 < dooglus> pixelgeek: I suspect darco thought we'd never use more than 20 threads, but you're right.  That 20 could be replaced with the number of threads to make the code better.
09:43 < dooglus> pabs3: did you hear back from the drive-by mac user who offered to help with packaging while I was away?
09:43 < pabs3> nope. sent him/her a mail, didn't get any reply yet
09:44 < pabs3> we've had a number of such people though
09:53 -!- pabs3 [n=pabs@d122-105-78-204.per9.wa.optusnet.com.au] has quit ["brb"]
09:56 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has quit [Read error: 110 (Connection timed out)]
10:00 < CIA-41> synfig: dooglus * r1929 / (7 files in 4 dirs): Accept 1927294: Patch from Gerald Young to allow intuitive editing of the BLinePoint ValueNode's "Amount" parameter by dragging the duck in the workarea.
10:24 -!- pabs3 [n=pabs@d122-105-78-204.per9.wa.optusnet.com.au] has joined #synfig
10:26 -!- ZanQdo [n=Daniel@201.201.2.22] has quit ["Adios"]
10:33 < dooglus> Yoyobuae: I think you've confused the BLineCalcVertex's "loop" sub-param with the BLine's "loop" flag - they're different things
10:34 < dooglus> Yoyobuae: the sub-param determines what happens if "Amount" is outside the range 0-1 (whether it's truncated or looped)
10:34 < dooglus> and the flag on the BLine determines whether the bline is open or closed
10:47 < pabs3> dooglus: how do you interact with Yoyobuae's patch?
10:48 < pabs3> hmm, not a very well phrased question
10:55 < pabs3> I mean what UI changes does it bring?
11:02 -!- artfwo [n=art-deve@87.103.169.152] has joined #synfig
11:10 < pabs3> hi artfwo :)
11:10 < artfwo> hello!
11:11 < pabs3> still 3 days left to participate in monster month :) http://synfig.org/forums/viewtopic.php?t=51
11:12 < pabs3> artfwo: is it too late to sync synfig 0.61.08 to hardy?
11:12 < artfwo> wow, that's some cool artwork for sure
11:12 < pabs3> (I see you are in #ubuntu-motu)
11:12 < pabs3> indeed :)
11:13 < artfwo> I think it is late indeed, though I am not an ubuntu developer (yet :) )
11:13 < pabs3> oh, bummer
11:19 < dooglus> pabs3: draw two blines.  select them both.  select a duck at one end of one of them.  right-click somewhere on the bezier curve of the other bline (between ducks, not on a duck), "link to bline"
11:19 < dooglus> pabs3: that will join the end of the first bline to the 2nd bline
11:20 < dooglus> pabs3: Yoyobuae's patch then allows you to drag that joined duck along the bline it's joined to, constraining it to stay on the bline
11:20 < dooglus> pabs3: I just committed a couple of fixes to make it work better
11:23 < pabs3> ah, cool
11:27 < CIA-41> synfig: dooglus * r1930 /synfig-core/trunk/src/synfig/valuenode_bline.cpp: Use 0 for the index if the closest point is on the segment which closes the loop. This keeps the final 'amount' value in the range 0..1.
11:27 < CIA-41> synfig: dooglus * r1931 /synfig-studio/trunk/src/gtkmm/ (canvasview.cpp renderer_ducks.cpp): When looking for the closest point on a bline to a duck, use the bline's loop flag rather than the BLineCalcVertex's loop sub-param, since this is what tells us whether the bline is looped or not.
11:32 < pabs3> thats a really cool feature addition!
11:35 < dooglus> yes, it's nice isn't it :)
11:35 < pabs3> indeed
11:38 < pabs3> is there a way to convert the tangent for the attached bline point to a right angle against the bline that the point is attached to?
11:42 -!- Zelgadis [n=zelgadis@87.103.169.190] has joined #synfig
11:42 < Zelgadis> hi, artfwo ! :)
11:42 < artfwo> hello :)
11:42 < Zelgadis> what brings you here?
11:42 < Zelgadis> ;)
11:43 < dooglus> pabs3: I was just wondering about that.  there is, yes, but I'm not sure how easy it is to do
11:43 < artfwo> Zelgadis: well, it's you who told me, that #synfig is where something great is always happening (in the free software and animation sense)
11:43 < factor> sounds like a cool fix
11:43 < Zelgadis> artfwo: aha :)
11:44 < Zelgadis> artfwo: glad to see you here
11:44 < dooglus> pabs3: ah, I remember: instead of selecting just the end point of the bline before linking, select the end point *and* the tangent of that point (and the width duck too, if you like)
11:44 < pabs3> aha, cool
11:44 < dooglus> then 'link to bline' and the tangent (and width) will be linked to the tangent (& width) of the bline
11:45 < Zelgadis> artfwo: are you not first time here today?
11:46 < Zelgadis> artfwo: ah... got it... pabs3 saw you on #ubuntu-motu
11:47 < dooglus> Zelgadis: artfwo has a very similar IP address to you
11:47 < pabs3> dooglus: ah, hmm. the tangent seems to point along the bline and always in one direction
11:47 < Zelgadis> dooglus: cool! Yoyobuae's patch is accepted! Now it's my turn to finish research with circular references! (only examples left)
11:47 < Zelgadis> dooglus: ^_^
11:48 < artfwo> I guess I'm using the same ISP as Zelgadis
11:48 < factor> nifty will have to recompile to see it
11:48 < dooglus> pabs3: 2 things: 1) while dragging the vertex along the bline, the tangent appears to be fixed - but when you release, it updates to point along the bline
11:48 < Zelgadis> artfwo: I guess we are from the same town :)
11:49 < dooglus> pabs3: 2) open up the vertex parameter, and open the 'bline tangent' node.  you'll see an 'offset' param.  set it to 90 degrees to get your perpendicular tangent
11:49  * Zelgadis updating synfig
11:50 < dooglus> I'll see if I can get the tangent's direction to change dynamically as the point is dragged
11:51 < pabs3> dooglus: awesomeness. just need a way to manipulate offset in the gui using the tangent duck 
11:52 < Zelgadis> pabs3: great idea!
11:53 < Zelgadis> if we linked tangent then when we manipulating it we changing amount parameter
11:53 < Zelgadis> looks logical
11:53 < pabs3> I do love synfig's linking stuff (convert stuff too once I learn about it)
11:53 -!- jcome [n=jcome@123.115.5.150] has joined #synfig
11:53 < Zelgadis> linking is the main synfig power
11:57 -!- jcome [n=jcome@123.115.5.150] has quit [Client Quit]
12:01 < Zelgadis> gotta go.
12:01 < Zelgadis> bye!
12:01 -!- Zelgadis [n=zelgadis@87.103.169.190] has quit ["Schastlivo!"]
12:15 -!- mwiriadi [n=mwiriadi@124-169-89-77.dyn.iinet.net.au] has quit [Remote closed the connection]
12:18  * factor just hacks around on synfig no real purpose except to see how it works.
12:19 -!- mwiriadi [n=mwiriadi@124-169-89-77.dyn.iinet.net.au] has joined #synfig
12:24 -!- jcome [n=jcome@61.49.255.103] has joined #synfig
12:26 < factor> is anyone going to be doing anything with fmod
12:28 < pabs3> since we can't distribute binaries linked to it (GPL violation), it is unlikely anyone will work on it
12:28 -!- Zelgadis [n=zelgadis@87.103.169.190] has joined #synfig
12:29 < factor> i agree
12:29 < pabs3> we should pick gstreamer or openal or have output plugins for the different systems that are available
12:29 < factor> yup
12:31 < Zelgadis> ha-ha! "Following a BLine" tutorial will be much shorter from now! 
12:34 < Zelgadis> dooglus: somehow I can't link to Bline layer Offset/Origin
12:35 < dooglus> Zelgadis: what to?
12:36 < Zelgadis> dooglus: I selecting green layer offset duck, then right-cliking on bline and there is no "Link to Bline" option
12:37 < Zelgadis> SVN1931
12:37 < dooglus> I see it here
12:38 < dooglus> you need to select the bline, right?
12:38 < Zelgadis> dooglus: oh, my mistake, forgot ot select bline
12:38 < Zelgadis> sorry
12:41 < Zelgadis> dooglus: It would be nice to have possibility to link Amount duck of Rotate layer to bline
12:41 -!- evilkillerfiggin [i=524481b6@gateway/web/ajax/mibbit.com/x-f35b61eb2e55f4df] has joined #synfig
12:43 -!- Zelgadis [n=zelgadis@87.103.169.190] has quit ["Schastlivo!"]
12:47 -!- jcome [n=jcome@61.49.255.103] has quit ["暂离"]
13:05 < factor> out
13:05 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has left #synfig ["Ex-Chat"]
13:11 < dooglus> Zelgadis asked "It would be nice to have possibility to link Amount duck of Rotate layer to bline" - I just added that
13:12 < dooglus> if you select the offset duck of an encapsulation layer and the angle duck of a rotate layer that controls the rotation of that encap layer at the same time, then right-click a selected bline and 'link to bline', then the encap layer will follow the bline and rotate with it too
13:12 < dooglus> r1932
13:16 < CIA-41> synfig: dooglus * r1932 /synfig-studio/trunk/src/synfigapp/actions/valuedescblinelink.cpp: Allow the linking of angle layer parameters to blines too.
13:29 -!- artfwo [n=art-deve@87.103.169.152] has quit ["Leaving."]
13:42 -!- xerakko [n=Miguel@debian/developer/xerakko] has quit ["Ex-Chat"]
14:15 -!- AkhIL [n=AkhIL@90.188.201.182] has quit [Read error: 104 (Connection reset by peer)]
14:17 -!- AkhIL [n=AkhIL@90.188.201.182] has joined #synfig
14:35 -!- factor [n=Factor-@32.145.114.75] has joined #synfig
14:51 -!- Zelgadis [n=zelgadis@87.103.169.190] has joined #synfig
14:51 < factor> lo zegadis
14:51 < Zelgadis> hello, factor ^_^
14:52 < Zelgadis> dooglus: many thanks! ^_^
14:56 < factor> have to wait till i get home to mess with synfig :(
15:46 -!- jcome [n=jcome@123.115.5.77] has joined #synfig
15:50 -!- genete [n=Genete@84.122.55.133.dyn.user.ono.com] has joined #synfig
15:50 < dooglus> seen this?  https://www.photoshop.com/express/index.html  a "free" version of PhotoShop online
15:52 < genete> hi dooglus 
15:53 < genete> my flash player version doesn't support that page
15:53 < genete> I have to update
15:55 < dooglus> genete: your version of ubuntu is out of its support period now anyway I think?
15:56 < genete> maybe but I'm updating using a tarball
15:57 < genete> what's the version of a working flash player in linux? 
15:57 < genete> for that webpage
16:01 < pabs3> version 9 works here
16:02 < pabs3> I have 9,0,115,0
16:02 < dooglus> Shockwave Flash 9.0 r115 here
16:02 < dooglus> :)
16:03 < genete> :) I've just installed the 9.0 r115 using a tarball :)
16:05 < factor> i really wish gnash worked up to par
16:10 < genete> a little limited but quite cool photoshop express
16:10  * Zelgadis will never exchange GIMP for photoshop
16:10 < AkhIL> is it support 16 or 32 bits per channel?
16:11 < factor> 16 or 32 bit per ip :)
16:12 < genete> Yoyobuae!!!! your patch is amazing!!!
16:12 < genete> dooglus: good work too!!!
16:12 < Zelgadis> Yeah, like it too ^_^
16:14 < Zelgadis> dooglus also did some work to Rotate layer and now "Following a Bline" is a piece of cake ^_^
16:14 < genete> but, before the patch I remember to select the Vertex parameter (the Bline point) and the small red square marked the duck. No it seems not to work :/
16:15 < factor> dho i have to wait till later to check it out
16:15 < Zelgadis> noticed that too
16:15 < genete> Zelgadis: it should be documented (the rotate layer thing) 
16:15 < Zelgadis> genete: sorry, busy with http://synfig.org/Linking_to_Blines_-_Equations
16:16 < genete> Zelgadis: NP. I just suggest it for the TODO list :)
16:17 < Zelgadis> genete: you can link Amount duck of rotate layer to bline too - that's the case
16:17 < Zelgadis> see today's logs
16:17 < dooglus> Zelgadis: I didn't touch the Rotate layer - just every Angle layer param can now be linked
16:17 < genete> ah I see
16:17 < Zelgadis> dooglus: well it's just way of speak :)
16:19 -!- pabs3 [n=pabs@d122-105-78-204.per9.wa.optusnet.com.au] has quit ["zzz"]
16:24 < dooglus> genete: I suspect Yoyobuae's patch stopped the red squares from appearing.  Hopefully he can fix it too.
16:25 < genete> telling that the patch is absolutely amazing, I have one question
16:26 < genete> if the bline that holds the vertex is looped how do you do to move the vertex from 0.9 to 0.1 by the shortest way?
16:26 < genete> it always go from 0.9 to 0.1 passing by 0.5 and not by 1.0
16:27 < genete> it is something similar to angles.
16:27 < dooglus> genete: yes.  do you think it should watch how the duck is moved, like it watches the rotation of angle ducks?
16:28 < genete> well I'm not sure of that
16:28 < genete> anyway if that is requierd you can manually modify the amount parameter and place a 1.1 instead of a 0.1 and the duck moves by the shortest way
16:29 < dooglus> yes
16:29 < genete> for some tasks it is good to go by the shortest way and for other don't
16:31 < dooglus> notice how once you have linked the vertex to the bline, you can delete the bline's layer and it stays linked?
16:35 < genete> YUP
16:35 < genete> it is like export the bline but with no name :)
16:44 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has joined #synfig
16:45 < Yoyobuae> I'm glad you guys are liking the patch
16:48 < dooglus> Yoyobuae: did you see the comments about the disappearing red squares?
16:50 < Yoyobuae> dooglus: i noticed that too, it happens because the vertex duck now has the valuedesc of the vertex valuenode rather than the blinepoint valuenode
16:50 < Yoyobuae> the red square appears if you expand and select the vertex vector valuenode
16:51 < Yoyobuae> the changes in duckmatic.cpp cause it
16:54 < dooglus> right
16:55 < dooglus> the red square is useful though, to immediately know which vertex is which without having to open it
16:55 < dooglus> is there any way we can go back to the previous duckmatic.cpp and still have your patch work?
16:56 < Yoyobuae> yeah, it could get the vertex valuedesc from the blinepoint valuedesc (or maybe the vertex valuenode directly)
16:56 < factor> when you have lots of ducks too
16:59 < dooglus> I don't think valuenodes can store valuedescs - because valuenodes are a core class, and know nothing about studio classes?
17:00 < Yoyobuae> i meant that getting the valuenode FROM the valuedesc
17:00 < dooglus> right, ok
17:01 < Yoyobuae> also, do you know that converting the "Bline" to "Dynamic List" breaks everything? XD
17:02 < dooglus> converting anything to anything seems to lose any details for that something, doesn't it?
17:02 < Yoyobuae> yes, but it seems weird to me that the interface is specifically made for a single valuenode type
17:03 -!- Zelgadis [n=zelgadis@87.103.169.190] has quit ["Schastlivo!"]
17:06 < Yoyobuae> dooglus: in renderer_ducks.cpp:line 312, we could check if the valuedesc parent is selected
17:06 < genete> Zelgadis: for when you come back: your equatin system to solve circular references is always determined? I mean the matrix is always invertable?
17:07 < Yoyobuae> dooglus: this way the valuedesc can still point to the vertex, but the red rectangle would show up when selecting both, the parent blinepoint or the vertex itself
17:09 < genete> Zelgadis: what if you want to link two vetices at the same time and they produce (both) the circular reference?
17:13 < genete> oh guys! I've created a link to bline with circular reference and not crash!!!
17:14 < factor> a plus
17:15 < dooglus> genete: how?
17:16 < genete> selecting two different blines/ducks when llinking the final one
17:17 < genete> http://i85.photobucket.com/albums/k74/Genete/synfig/Pantallazo-circular_reference2sifzA.png
17:17 < genete> that's after remove the extra bline
17:18 < dooglus> Yoyobuae: I don't understand your patch much, or any of the duck code really
17:18 -!- jcome [n=jcome@123.115.5.77] has quit ["暂离"]
17:18 < dooglus> why do you call restrict_blinevertex_duck() so many times?  is it needed for the tangents too?
17:19 < factor> sounds like it nereds some docs
17:19 < Yoyobuae> it's needed for vertexes, tangents, beziers (they're made from ducks), and their origin duck's, and their orgin's origin duck, ..... etc
17:20 < genete> http://www.darthfurby.com/genete/synfig/circular_reference.sifz
17:20 < genete> check it out
17:20 < genete> remove the black outline and it is done
17:23 < genete> oh! that's not a circular reference case! 
17:23 < dooglus> genete: that's not circular, no :)
17:23 < genete> I was dreaming
17:23 < dooglus> green -> (blue & red); red -> blue
17:23 < genete> ;)
17:23 < dooglus> blue -> nothing :)
17:23 < genete> true
17:24 < Yoyobuae> but then can't we just avoid circular references entirely?
17:24 < genete> Zelgadis is working on that
17:25 < Yoyobuae> no, i mean do things like what you just did
17:25 < dooglus> Yoyobuae: the user might make a circular reference accidentally
17:26 < Yoyobuae> then we warn him, and he can do things some other way XD
17:26 < dooglus> sure
17:26 < dooglus> we currently don't know it's happened (until we run out of stack space due to infinite recursion)
17:27 < Yoyobuae> bleh, we need a spanning tree algorithm XD
17:28 < dooglus> we have one - the Node tree has a "last modified" field that can be used
17:33 < dooglus> Yoyobuae: are you looking at the 'red square' thing?  I'm not sure what your solution involves still
17:33 < Yoyobuae> would this be ok: selecting a blinepoint highlights all of it's ducks with red squares (vertex, tangents, width). selecting each parameter highlights each one individually
17:34 < dooglus> that would be better, I think, yes
17:34 < dooglus> makes more sense, don't you think?
17:34 < Yoyobuae> it was accidental, I intended it to highlight the vertex duck only XD
17:35 < dooglus> I think the use of "vertex" to mean 2 different things should stop too, don't you?
17:36 < dooglus> "Vertices" contains "Vertex 001" which contains "Vertex", "Width" and "Tangent"?
17:37 < Yoyobuae> "Bline Point 001" seems more appropiate, for bline points that is
17:38 < Yoyobuae> or  "Point 001" if shorter is better
17:40 -!- alvarae [i=44dea9fd@gateway/web/ajax/mibbit.com/x-5a6e790b2b78c326] has joined #synfig
17:43 -!- alvarae [i=44dea9fd@gateway/web/ajax/mibbit.com/x-5a6e790b2b78c326] has quit [Client Quit]
17:44 -!- Zelgadis [n=zelgadis@87.103.169.190] has joined #synfig
17:44 < Zelgadis> genete: I'm back by request :)
17:47 < genete> Zelgadis: welcome back :)
17:49 < Zelgadis> dooglus: in respond to: "17:26 < dooglus> we currently don't know it's happened"
17:49 < Zelgadis> dooglus: we can predict when it's happened
17:52 < Zelgadis> agree with Yoyobuae about "Point 001"
17:56 < factor> out
17:56 -!- factor [n=Factor-@32.145.114.75] has quit ["mIRGGI meni puis"]
17:57 < genete> see you on monday... nice weekend :)
17:57 -!- genete [n=Genete@84.122.55.133.dyn.user.ono.com] has quit ["Abandonando"]
17:59 < Zelgadis> genete: I don't know if matrix is always invertable. If it isn't then there is no solution and vertex cannot be placed on a specified place on bline
17:59 -!- evilkillerfiggin [i=524481b6@gateway/web/ajax/mibbit.com/x-f35b61eb2e55f4df] has quit ["http://www.mibbit.com ajax IRC Client"]
18:00 < Zelgadis> genete: I can't see what could be wrong with linking two verticles producing circular reference. Can you be more specific? Maybe some example?
18:00 -!- Zelgadis [n=zelgadis@87.103.169.190] has quit ["Schastlivo!"]
18:06 < Yoyobuae> dooglus: added a small patch to the tracker, to fix the red squares issue =)
18:16 -!- omry [n=omry@bzq-84-108-20-56.cablep.bezeqint.net] has quit [Remote closed the connection]
18:28 -!- omry|away [n=omry@bzq-84-108-20-56.cablep.bezeqint.net] has joined #synfig
18:28 -!- omry|away [n=omry@bzq-84-108-20-56.cablep.bezeqint.net] has quit [Read error: 104 (Connection reset by peer)]
18:28 -!- omry [n=omry@bzq-84-108-20-56.cablep.bezeqint.net] has joined #synfig
18:32 -!- xerakko [n=Miguel@debian/developer/xerakko] has joined #synfig
19:02 -!- _Elk [n=_Elk@ti0110a340-0355.bb.online.no] has joined #synfig
19:13 -!- factor [n=Factor-@32.145.225.3] has joined #synfig
19:27 -!- xerakko [n=Miguel@debian/developer/xerakko] has quit ["Ex-Chat"]
20:13 < CIA-41> synfig: dooglus * r1933 /synfig-studio/trunk/src/gtkmm/workarea.cpp: Fix "zoom to fit" so that it centers the fitted canvas on the screen, and always zooms back to the previous zoom level and position if clicked a second time.
20:13 < CIA-41> synfig: dooglus * r1934 /synfig-studio/trunk/src/gtkmm/renderer_ducks.cpp: Accept 1927294 (a further patch from Gerald Young): the previous patch stopped the red square around ducks from showing up when individual blinepoints were selected in the parameter panel.
21:32 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has quit [Read error: 104 (Connection reset by peer)]
22:08 -!- xerakko [n=Miguel@debian/developer/xerakko] has joined #synfig
22:16 < CIA-41> synfig: dooglus * r1935 /synfig-studio/trunk/src/gtkmm/canvasview.cpp: Allow a BLineCalcTangent's tangent duck to be dragged and have it edit the 'offset' sub-parameter.
22:17 -!- factor [n=Factor-@32.145.225.3] has quit [Read error: 110 (Connection timed out)]
23:55 -!- LinuxO [i=wbzrigy@201.248.156.142] has quit [Read error: 110 (Connection timed out)]
--- Log closed Sat Mar 29 00:00:42 2008