proposed interaction physics <=> graphics

A place to discuss everything related to Newton Dynamics.

Moderators: Sascha Willems, walaber

proposed interaction physics <=> graphics

Postby Carli » Thu Oct 15, 2009 3:28 am

Hi,
we (the Toast Engine developement team) want to use newton in our Engine but we want to get an issue, how we should handle graphics/physics.
Should we
a) Load a graphic and convert it to a newton collision
b) generate a newton Collision and generate a newton mesh
c) should we generate newton collision and mesh separatly

thanks for helping.
Carli
 
Posts: 245
Joined: Fri Oct 02, 2009 5:28 am

Re: proposed interaction physics <=> graphics

Postby JernejL » Thu Oct 15, 2009 6:09 am

Carli wrote:Hi,
we (the Toast Engine developement team) want to use newton in our Engine but we want to get an issue, how we should handle graphics/physics.
Should we
a) Load a graphic and convert it to a newton collision
b) generate a newton Collision and generate a newton mesh
c) should we generate newton collision and mesh separatly

thanks for helping.


It really depends on what kind of environment it is, for example if you have big worlds, it would be benefical, to create low-poly meshes of the world to use for newton trimeshes, usually a collision mesh has less faces, because it doesn't need detailed precision (like all windows on a building, etc..), GTA games all use this approach (gta3, vc, sa, IV..).

That's one example for trimeshes, you should show how your game looks like so we can give better suggestions.
Help improving the Newton Game Dynamics WIKI
User avatar
JernejL
 
Posts: 1587
Joined: Mon Dec 06, 2004 2:00 pm
Location: Slovenia

Re: proposed interaction physics <=> graphics

Postby Carli » Thu Oct 15, 2009 6:47 am

It would be a greater world, that means i should use simple newton bodies like boxes, spheres, cylindres and cones and the "real" model for graphics?


or is it better to generate a low-poly model? (when yes: what is the easiest format to generate a newton collision from a mesh?)
Carli
 
Posts: 245
Joined: Fri Oct 02, 2009 5:28 am

Re: proposed interaction physics <=> graphics

Postby JernejL » Thu Oct 15, 2009 7:39 am

Carli wrote:It would be a greater world, that means i should use simple newton bodies like boxes, spheres, cylindres and cones and the "real" model for graphics?


You can still use compound collisions, convex hulls, trimeshes, etc.. if you meant a greater free-roaming world (like gta).

Carli wrote:or is it better to generate a low-poly model? (when yes: what is the easiest format to generate a newton collision from a mesh?)


It is.. if you have enoremous worlds, otherwise you'll use up a lot of ram.
Help improving the Newton Game Dynamics WIKI
User avatar
JernejL
 
Posts: 1587
Joined: Mon Dec 06, 2004 2:00 pm
Location: Slovenia

Re: proposed interaction physics <=> graphics

Postby Stucuk » Thu Oct 15, 2009 7:23 pm

If you have a football in your world, you could just make it a NewtonSphere. If however you had a wrist watch (Which you wanted to have a precise physics body) then making a low poly model for the physics would be a good idea. It all depends on the objects in your world and the level of precision you want from there bodies. In general most objects should be able to be made out of compound collisions, but there are always things that can't without sacrificing precision.
User avatar
Stucuk
 
Posts: 801
Joined: Sat Mar 12, 2005 3:54 pm
Location: Scotland

Re: proposed interaction physics <=> graphics

Postby Carli » Fri Oct 16, 2009 7:14 am

ok, i will se how
Carli
 
Posts: 245
Joined: Fri Oct 02, 2009 5:28 am


Return to General Discussion

Who is online

Users browsing this forum: No registered users and 0 guests

cron