Single icicle with air bubbles, rendered in 4.5min.
I do not know what to think about this prototype yet.
I am happy with the irregular shapes formed by the bubbles, but not  
completely satisfied with the distribution.

Real icicles are formed in layers, which I could do by (I think) a method  
similar to the way I test if the air bubbles are "inside" the icicle  
object.
Right now I construct a sphere based approximation of the icicle and test  
against that to determine if the air bubbles are "inside" the real icicle  
object. I suppose if I executed a recursive orbiting trace() of that  
object, placing air bubbles as well as perhaps *slightly* varying the IOR  
of each layer, I might be able to emulate something like what you see at  
the following link:  
http://www.physics.utoronto.ca/~smorris/edl/icicleripples/icicleripples.html

Any thoughts?

Branching as seen in the following link is next:  
http://farm4.static.flickr.com/3475/3238212338_9245029bf5.jpg?v=0

-- 
Ian McDonald
Lean Agile .NET 4.0/MVC
Senior Application Architect,
Developer and Security Analyst

Post a reply to this message

Attachments:
Download 'blobtest2011_7.png' (45 KB)

Preview of image 'blobtest2011_7.png'