i made this patches for an outdoor projection on a house. we will use mice and wacoms to draw the elements live onto it using 2 beams.
it uses dynamic patching for the strokes and a little branching algorythm for the trees.
after a tree is created it is snaped with snap2tex, and aferwards the same tree is created again with other random vals.

hope you like it
greets
sebastian

i used wintablet to connect 1 wacom with pd (wich workes great) on a XP machine with 3 nvidia cards and 4 beams (4000 Lumen each).
i will develop my drawing tools further, its great to involve non-nerds into the vj-session.
the video is quite bad, i was too lazy for propper documentation as usual

Great work!

for the drawing along the cursors track
i use some dynamic patching
a point P is drawn at current curserposition.
when the cursor is moved away by a specifyed lenght
a new point P1 and with it, a new patch is generated automatically with the coordinates of P and P1 (new position) as creation arguments . inside that patch, a line joins P + P1 in gem.
i hope this is understandable.

the trees is a generative algorythm i made up.
its one tree, spreading at different positions - grows, and then gets [snaped to tex]
just to grow somewhere else with new random values.
its really so many things in there...

. . . i was also wondering where you did the mapping/ did you map the building at all or just arrange the four projection surfaces to tile. i heard pd was not that great at projection mapping, v4 being far superior . . .

about v4 being far more superior in anythig than pd...
i dont think so.
it really depends on the tools, and knowledge you have.
both are nice frameworks for generative visuals. i dont think there is anything in vvvv that couldnt be done with pd, while the other way around...i dont think you can go so deep in vvvv (without compiling new stuff)
i did an installation with a big screen projection over 8 videobeamers with pd and it worked great. it ran fast and very stable. the meso guys had to fight much more than me to get that equipment into grip with vvvv.
and..what you need for projection mapping? mainly some squares witch corners you can drag with the mouse, to fit it to something and use it as a videoplane . so i really dont see why pd + gem should be not so good. maybe because there are not so many tools readymade and one would have to think a little. i guess the real reason why anybody feels that vvvv is more superior: because there is a big community doing this stuff with it. i think ive proved that gem + pd is as powerfull and able with my work here. but look at the feedback...its not that too many people are interested in it. pd remains audio, while vvvv is known for good video stuff. but this has nothing to do with the software, just with the people that use it.

I would really like to understand how the dynamic patching works.