Game Development Community

Why CSG?

by Keith Johnston · in Technical Issues · 04/15/2003 (11:22 pm) · 3 replies

Why is it that so many games use Constructive Solid Geometry (convex brushes) as the basic element in making levels? Is this a relic of the past when polycounts had to be very low? Or are there other reasons for using CSG? I imagine that it would make various lighting / visibility calculations easier - but as processor and graphic card speeds increase, will it still be used?
NextNext Thread
PrevPrevious Thread
More from Technical Issues


Diogo de Andrade
#1
04/16/2003 (2:33 am)
I'm probably not the best person to answear this, but I figure that they are used because they simplify the process of building the BSP tree of a level (in engines that still use it, don't know if Torque does), although I think the BSP is dying... On the other hand, I think that the style of levels in older games was such that the "logic" way to build them was using convex brushes, etc...

But I've been know to be wrong... :)

Torque Owner Sander Van Rossen
#2
04/16/2003 (3:30 am)
Well keep in mind that there are more things to 'levels' then just rendering, there are also things like collision detection.

Torque Owner Jarrod Roberson
#3
04/16/2003 (9:03 am)
with CONVEX CSG alot of assumptions can be made and optomizations about what you can expect with the data set. And Sander is right, collision detection is greatly simplified if you know for a fact that ALL your geometery is convex.