• Create Account

# Force/Torque & Vector Fields

Old topic!

Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

2 replies to this topic

### #1Mark Slater  Members   -  Reputation: 103

Like
0Likes
Like

Posted 13 October 2012 - 04:43 AM

Hey,

For my current game I developed a scripting language to describe vector/scalar fields (similar to the approach on bitsquid). It's working well for simple particle systems, but I'd now like to use it for our main game physics as well. The problem is, how can I generate the torque for a body given a field? A simple, brute force approach would be to sample the field at multiple points of the meshes surface and generate a force/torque. However, this would be:
• Inaccurate
• Too computationally expensive to run on multiple game objects (based on profiling of the particle system and the number of objects in our scenes).
Any suggestions? We already use Helmholtz Decomposition to split our fields into div/curl free parts, which simplifies matters. I'm sure my old electrodynamics books could help, but I can't find them just now.

### #2apatriarca  Crossbones+   -  Reputation: 2275

Like
0Likes
Like

Posted 13 October 2012 - 06:01 AM

They designed that system to animate things like particles, cloths, grass, trees... It does not seem to be very well suited for rigid body simulations.

### #3Mark Slater  Members   -  Reputation: 103

Like
0Likes
Like

Posted 13 October 2012 - 06:15 AM

That's the conclusion I was coming to. It's unfortunate as I have already created some interesting effects using the brute force method and representing a mesh as a sphere tree (with each sphere acting as a constrained particle). If you'd like to see what I mean, I'll be releasing a tutorial on our vector calculus scripting language at the end of next month which includes this demo.

Anyway, given the size of our scenes, I guess the main game objects are stuck with central forces and specific, optimized handlers for non-central forces for now. If anyone has any other ideas I'd welcome them.

Edited by Mark Slater, 13 October 2012 - 08:31 AM.

Old topic!

Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

PARTNERS