Hi Renyue,
It turns out that's not such an easy thing to do at the moment. Because of the way the transfer function was constructed, most of the renderings turn out to live in an optically thin regime. This is primarily due to the adaptive nature of grid traversal and the clipping of the transfer function to have a maximum value of 1.0 and the way the integration is approximated. I've added some changes to this in a bitbucket fork that is part of an effort to refactor the volume rendering. If you'd like to try it out, you can clone this fork from:
Note that this also brings with it some OpenMP stuff, and by default it runs with OMP_NUM_THREADS = number of cores the node. You can get rid of this with export OMP_NUM_THREADS=1.
Because of the adaptive nature of our ray-casting, you may need to set alpha to something higher than 1.0 (since it get's multiplied by something like dx during ray tracing). You can find an example script of mine where I add three isocontours, as well as use a camera lighting (yet another feature in the works).
Let me know if you have any luck/difficulty with this. We will be pushing for this in the next stable release.
Best,
Sam
On Tue, Mar 20, 2012 at 10:37 AM, Renyue Cen
<cen@astro.princeton.edu> wrote:
Hi,
For iso-density surface I have a primitive question: how do I make certain contour level totally opaque?
I can do " tf.add_gaussian(-25.5, 0.05, [1.0, 0, 0, 1.0])"
but specifying alpha=1.0 in this example does not seem to make that level (red) totally opaque.
Thanks,
Renyue
_______________________________________________
yt-users mailing list
yt-users@lists.spacepope.org
http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org