Re: plotting particle data with halos? [message #33699] |
Thu, 16 January 2003 20:20 |
Craig Markwardt
Messages: 1869 Registered: November 1996
|
Senior Member |
|
|
Matt Wood <wood@astro.nospam.fit.edu> writes:
> I've got particle data generated from 'smoothed particle hydrodynamics'
> code that I'd like to visualize using idl. Are there any routines that
> will plot particles with 'halos' instead of simple colored points? If
> you'd like to see an example of what we're doing, please see
> www.astro.fit.edu/wood/100k34.gif
> It's a simulation of an interacting binary star system that has an
> accretion disk. Color indicates temperature in the disk.
The easiest thing to do might be to plot to the Z buffer, capture it,
then convolve with a gaussian, or even simpler, just SMOOTH it. That
will give a more smooth look, but of course every point will have the
same "halo" size.
However, if you really need variable size smooth particles, then you
need to kick it up a notch. You could compute a 2D gaussian for each
point, add them to an IDL array one at a time, but that might get
pretty computationally expensive. Better might be to make up a
library of 2D gaussians and then add those in turn, picking the
closest one.
A trick combination of the 1st (convolution) and 2nd (addition of
gaussians) approaches would be to sort the points according to their
particle size, partition them into a few groups, make
plots+convolutions of each using the right particle size, then add
them up.
Cool sims! I work on X-ray binaries.
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@cow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
Re: plotting particle data with halos? [message #33703 is a reply to message #33699] |
Thu, 16 January 2003 17:43  |
Rick Towler
Messages: 821 Registered: August 1998
|
Senior Member |
|
|
"Matt Wood" <wood@astro.nospam.fit.edu>
> I've got particle data generated from 'smoothed particle hydrodynamics'
> code that I'd like to visualize using idl. Are there any routines that
> will plot particles with 'halos' instead of simple colored points?
What exactly do you mean by "halo"? Do you mean that literally, as if the
particle were glowing? Or do you mean some sort of trail?
If you mean glowing, there aren't any canned routines for this that I know
of but it could be done to some degree in object graphics. I say some
degree because you would use alpha blending to create the halo effect and
IDL's object graphics renderer is not well suited for this type of rendering
(multiple moving alpha textured objects). A lot will depend on how you view
things.
Another consideration will be the number of particles you wish to create.
At a minimum each particle would be 2d and require 4 vertices in which you
would be restricted to a single static view. Practically 8 or 12 verts
would be required if you were to "billboard" the texture to view from all
angles (I actually don't know how this would render in IDL but am curious to
find out). If you want to draw 10k of these things it could get ugly.
If you are still interested I can get you started.
-Rick
|
|
|