This'll be the last one until I get the object-interaction done.
I've just implemented the possibility to have the simulation
run in a subpart of the animation. Which means that it
isn't based upon the standard-variable "clock", and also isn't
calculated all the time.
I can now set the simulation run only from clock=.25 to clock=.5,
still retaining the mesh before and after those values. This
is useful if for example you know that the water only needs to be
simulated after the clock has reached a certain value, and, if
it isn't in the image anymore or has calmed down, it won't be
simulated later on. This can save immense amounts of parsing
time on multistage animations.

In this test, it rains for the first 6 frames or so (which means that
wave emitters are caused in those frames, waves won't disappear
after that), then we get a stomp (must be some large T-Rex nearby)
and finally the sphere starts moving, causing a trail-wave. I actually
like how the sphere causes a nice dip when entering and
exiting the liquid. The trail waves' height are affected by how close
the sphere's center is, which is just a very simply and crude method
of simulating less touch vs heavy touch of the trail-wave.

Well, enjoy. In order not to cramp this group with my animations,
I'll post the next one once I've got object-interaction implemented
(which I plan on beginning tomorrow).

If there's interest in the simulation so far, send me an email. I'll try and
slap a quick example scene-file together with the simulation-include, and
you may fumble with it a little. Notice though that lots of the code will
change again once I've implemented object interaction.
As a side-note:
It requires my I/O-Macros for data-handling and my MMM for visualization.

Regards,
Tim

--
Tim Nikias v2.0
Homepage: http://www.digitaltwilight.de/no_lights
Email: Tim### [at] gmxde

Post a reply to this message

Attachments:
Download 'water_allmix.mpg' (349 KB)