 |
Author |
Topic: snow (again) (Read 2411 times) |
|
heavysixer
|
snow (again)
« on: Nov 8th, 2004, 8:23pm » |
|
I know that people have been posting snow and sand demos on this board for a while. I tried to take a crack at it too. I wanted some help however i need some advice on the piling code. Actaully, any feedback is welcome.
http://www.flavoredthunder.com/dev/processing/flakes/
Thanks,
Mark
|
|
|
|
|
Fish-face
   
|
Re: snow (again)
« Reply #1 on: Nov 11th, 2004, 5:33pm » |
|
Looks like you need a more complicated CA algorithm for the piling.
try checking further out than just 2 cells - real snow piles up in a curvier way than this simulation; it tends to go at a 45 degree angle, which doesn't look right.
|
--
The C@ S@ on the M@
=Fish-Face=
~#Chris#~
|
|
|
heavysixer
|
Re: snow (again)
« Reply #2 on: Nov 23rd, 2004, 1:26am » |
|
Hi, I totally agree, so you suggestion would be to move one or more pixels out to determine if there is still empty space before i drop?
Thanks,
Mark
|
|
|
|
|
Fish-face
|
Re: snow (again)
« Reply #3 on: Dec 1st, 2004, 10:26pm » |
|
Hmm... It's difficult to say, really.
Basically, if you think about snowflakes falling, if two clumps fall on top of one another, the bottom clump tends to spread out. I'd try to model that: if there's more snow on top of a particle, have it exert force on it's neighbours... This will require modelling of both air and snow, though - i.e. snow with gaps in.
|
--
The C@ S@ on the M@
=Fish-Face=
~#Chris#~
|
|
|
arielm
|
Re: snow (again)
« Reply #4 on: Dec 1st, 2004, 11:07pm » |
|
hi there,
i've been working a lot with sand earlier this year: e.g. the book of sand & mobile sand
the concept that helped me the most was "Self-Organised Criticality" and this page is explaining it very well.
S.O.C is the key to achieve the correct balance, i.e. an equilibrium between sandpiles with ugly 45-deg. angles and sandpiles that are almost "flat"...
hth!
|
Ariel Malka | www.chronotext.org
|
|
|
Fish-face
|
Re: snow (again)
« Reply #5 on: Dec 23rd, 2004, 9:29pm » |
|
The problem for snow is that snow is sticky, and can actually result in a gradient of over 180 degrees... Not sure what to do about that.
One possibility is to measure the gradient at a macro scale, i.e. allow large gradient fluctuations at the micro, but as long as, when averaged out, this is not over the critical angle...
|
--
The C@ S@ on the M@
=Fish-Face=
~#Chris#~
|
|
|
|
