Wednesday, November 29, 2006

gna 2.0

based on the feedback, i made several changes to the algorithm. my first step was to create a funtion to map the points each volume was responding to. in the process of doing this, i found a bug in the previous algorithm which took me a while to fix. i also rewrote the placement function, so the archipelago generated is much more 'linear'. in appearence, the current algorithm is very similar to the previous, but the inner working have some major revisions.

i also played around with using metaballs. i didnt like the forms it was generating, so i put that exploration on hold. i may come back to it.

i also looked into animating the points. i posted on the formz forum for some help. the functions if need will not be available until formz 6.1 is released. i was told that they were hoping to release this upgrade by january. so i may be able to explore that further if the new version is released.

the next step is to have the size/form of each volume also respond to the conditions. i think that i have a good backbone in place, so now i can really begin to experiment.

here are some images of the current algorithm:







here is a .txt file of the algorithm

1 comment:

rael said...

jason, perhaps it metaballz were too blobby, you could try metaformz, which you could apply to your existing rectangles so that they can repel and attract each other, if that is somthing you want to explore on that front.

i think there are two critcal ways to communicate your project for the final. the first is through animation - much like the techniques you used at midterm. the second is suggesting what these giant structures are/do when they are placed in the physical/virtual landscape.