Problems
to Avoid and Ignore
The
simulation has a couple of problems. The first is that the way to simulate
cloth realistically is to use a lot of points in the simulation. This
takes more computation time. High-end animation programs rely on a great
number of particle points for realism. Of course, in other fields, hour-long
render times are perfectly acceptable. In a real-time game, however,
this won’t get you on the cover of any game magazines. You have to sacrifice
realism for speed. This is another good area for scaling game performance.
If the system is running quite fast, subdivide the cloth patches a little
more. Game players with a white-hot system should have smooth-looking
cloth.
Another
problem is that each spring acts independently. This means that each
spring can be stretched to a great extent. In many cases, the amount
of stretch can exceed 100 percent. This is not very realistic. Actual
fabric will not stretch in this manner. The problem I have is that I
am using linear springs when fabric actually displays a nonlinear spring
behavior. As the amount of stretch increases, the strength of the spring
increases also. The fabric will also stretch to some limit and then
if the force continues, it will rip. This is not what I want (at least
for now). This issue, which Xavier Provot (see For Further Info) calls
“the Super-Elastic Effect,” is difficult to handle. Increasing the spring
strength dynamically can lead to instability problems just like any
other stiff spring problem. Provot suggests checking the amount of stretch
in each spring, and if it exceeds a set deformation limit, the springs
are adjusted to achieve this limit. While I agree this solves a definite
problem, a second pass through the springs is costly. For the effects
I have attempted to achieve, I can live with super-elastic cloth.
My
collision system is pretty primitive. To make things easy, I only collide
the vertices of the mesh with the objects. As it stands, if a sphere
is small or the fabric stretches too much, the sphere will pass right
through it. I also don’t handle self-collisions. That is, the fabric
can pass through itself without penalty. This could be corrected by
placing bounding spheres at each vertex. However, applying the sphere
collision test between each vertex gets expensive. So, I just limit
the situation so that either the cloth doesn’t pass through itself,
or so the effect isn’t too noticeable.
Taking
It to the Limit
Once
the system is working, it’s fun to see how it can be extended. I mentioned
the issue of tearing and ripping after the fabric stretches too far.
I can monitor the spring lengths. If they exceed a limit, the spring
can be removed from the system, effectively tearing the fabric. I think
this would be a great way to simulate a cannonball tearing through the
mainsail of a tall ship. This same method of breaking a spring would
work for a simulation of a rope as well. After all, a rope is really
just a one-dimensional version of the cloth patch.
Another
dynamic effect can be achieved by manipulating the flexion springs.
With these springs in place, the fabric will resist folding. However,
if I selectively delete one of these springs, the fabric will be able
to fold nicely where the springs are missing. I don’t know where I can
use that yet, but I’m sure I can find a way.
The
Application
The
application this month was actually pretty easy to build. It’s essentially
the same as last month’s application, but with a few additions. There’s
a function that creates the cloth patch in a sort of macro fashion.
You can set the spring settings for the three types of springs. You
can also drop some collision objects around and watch them interact.
Find the application and the source at http://www.gdmag.com.
Acknowledgements:
Special thanks to Chris Hecker of Definition 6 and Rob Wyatt of Dreamworks
SKG for discussing the issue with me.
For
Further Info:
•
Provot, Xavier. “Deformation Constraints in a Mass-Spring Model to Describe
Rigid Cloth Behavior,” Graphics Interface, 1995, pp. 147-155.
Also available in electronic form at http://www-rocq.inria.fr/syntim/research/provot
•
Baraff, David and Andrew Witkin. “Large Steps in Cloth Simulation,”
Proceedings of SIGGRAPH 1998, ACM SIGGRAPH, pp. 43-54.
There are also fabric simulations available for many professional 3D
animation packages available either as plug-ins or integrated into the
software. I do not know what techniques these products use with the
exception of one. Colin Withers of Topix created a fabric simulation
for Softimage based on the Provot paper. Graciously, Topix released
the source code for this plug-in to the public. See http://www.topix.com
for more info.
Bio:
Jeff Lander prefers to wear comfortable loungewear when hanging out
writing code at Darwin 3D. Drop him a note and let him know what the
fashion conscious are wearing this spring at jeffl@darwin3d.com.
________________________________________________________
[Back
to] Introduction
|
|
|
|
Features
