Bill Pragnell wrote:
>>Kirk Andrews wrote:
>>
>>>Thanks to Sam for introducing me to the concept.  I had wanted to do something
>>>like that before, but had never thought of using averaging an object pattern.
>>>Some initial results:
> 
> 
> Ooh yes, this is a lovely technique. Thanks for the hint guys, I would never
> have thought of using object patterns in this way. I must play with this in
> future projects. Attached is my best experiment of the day, nowhere near as
> believable as that column but it's just proof-of-concept for me!
> 
> stbenge <THI### [at] hotmailcom> wrote:
> 
>>I have an array of 251 points set in a regular spherical/cubic fashion
>>that I use for this very purpose. It effectively distributes the object
>>pattern in 3D space, thus reducing any inefficiencies arising from
>>complete randomization. A little randomization, though, helps avoid
>>other artifacts associated with such a regular pattern. Attached is a
>>pic showing how the elements are placed.
> 
> 
> I'm using a pure random distribution. I guess you could just jitter your points
> in that illustration to keep a random element if it was ending up too regular.
> 
> This is interesting stuff. :)
> 
> Bill
> 
> 
> ------------------------------------------------------------------------
> 
Though I already replied so second reply:
Yes Bill, with your mesh macros you can do the same averaging in certain 
cases before applying the pattern to the mesh to get some similar 
weathering effects.  Here I jitterd and averaged the radial pattern that 
was used to get the fluting in the column to get a weathered look, then 
applied it to the mesh.

http://tinyurl.com/5oqr7u

Post a reply to this message