Advantages of >3 vertices in NewtonTreeCollisionAddFace?

A place to discuss everything related to Newton Dynamics.

Moderators: Sascha Willems, walaber

Advantages of >3 vertices in NewtonTreeCollisionAddFace?

Postby Leadwerks » Sat Nov 16, 2019 6:44 am

Does NewtonTreeCollisionAddFace actually produce better collisions along edges if faces are fed in with more than three vertices (a quad, for example), rather than triangulating the entire mesh? I believe this is the case.
User avatar
Leadwerks
 
Posts: 531
Joined: Fri Oct 27, 2006 2:54 pm

Re: Advantages of >3 vertices in NewtonTreeCollisionAddFace?

Postby Julio Jerez » Sat Nov 16, 2019 8:32 am

Only if they are not optimized.
But I strong recommend calling with optimize flag.

You can pass any polygon, even concave faces as long as they are flat and do not self intersect.
The optimizer apply a delauny traingulation and then convert the mesh to convex slices.

My recomention is to pass triangles wich are not anbiguos and conform to the graphic and let the builder figure it out.

adding other kind of polygon like the type generated by splines or procedural algorims, in general will have some curvature, Wich will be a problem for the polygon collider shape.
Julio Jerez
Moderator
Moderator
 
Posts: 11048
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Advantages of >3 vertices in NewtonTreeCollisionAddFace?

Postby Leadwerks » Sun Nov 17, 2019 7:13 am

Okay, thank you for the explanation.
User avatar
Leadwerks
 
Posts: 531
Joined: Fri Oct 27, 2006 2:54 pm

Re: Advantages of >3 vertices in NewtonTreeCollisionAddFace?

Postby JernejL » Wed Nov 27, 2019 1:59 am

I feed trimesh huge grid of rectangle quads and the optimizer does wonders, i'm sure physics simulation is running much more efficient because of this.
Help improving the Newton Game Dynamics WIKI
User avatar
JernejL
 
Posts: 1500
Joined: Mon Dec 06, 2004 2:00 pm
Location: Slovenia


Return to General Discussion

Who is online

Users browsing this forum: No registered users and 7 guests