[sldev] Hacking the wind

[Aimee Trescothick]

On 27 Jul 2009, at 01:59, Teravus Ovares wrote:

> This '2D'
> representation of wind makes it almost impossible to simulate a
> tornado.   Also, the resolution of wind is not great for simulating a
> tornado because the sim is broken up into 16 infinitely tall columns.
> You'd have an easier time simulating a hurricane then a tornado :).
> A tornado would need a higher resolution representation as well as a
> representation in 3D space with 3D wind.

3D wind would be nice for flight, lift from thermals etc. The 2D wind  
can still do some pretty cool things at times though :)

http://www.flickr.com/photos/aimeethyst/2878821401/sizes/o/in/photostream/
http://www.flickr.com/photos/aimeethyst/2878884819/sizes/o/in/photostream/

1400 windsocks ... the black dot near the bottom centre is me :)


For reference there's a brief comment on the workings of SL wind from  
Andrew Linden here:
http://forums.secondlife.com/showthread.php?t=146384&lang=en

... might be an interesting conversation topic to bring up at his  
office hour some time.

The paper by Jos Stam he mentions it's based on is here:
http://www.dgp.toronto.edu/people/stam/reality/Research/pdf/ns.pdf

... and some explanatory notes on it here:
http://techhouse.brown.edu/~dmorris/projects/summaries/dmorris.stable_fluids.notes.pdf

Aimee.