mightymarcus Posted November 15, 2016 Share Posted November 15, 2016 I have a simple question about culling of meshes to find out whats the best strategy to keep the performance high. What is the better approach: 1. merge static meshes to a single terrain-mesh. I guess the culling doesn't cull triangles but the whole mesh. So a part of the big terrain-mesh is always on screen, this means that the complete terrain-mesh will always be drawn, even if we see just a small part of the mesh, right? There isn't something like faces culling? 2. merge static meshes to multiple terrain-meshes. This way the amount of drawcalls is reduced too, but only the parts of the terrain are drawn that are on screen. E.g. I split up the terrain into 4 - 8 single meshes. I guess it's 2. Can someone enlighten me? I make a game with a top down view, where the player can see just only a small part of the terrain at once, so I don't want the whole geometry to be drawn in every frame. I am attaching terrain and decorations like trees and some other stuff to a single mesh and wonder if that's a overkill, if the mesh gets a high vertices count and will always be drawn completely. Thanks. Quote Link to comment Share on other sites More sharing options...
GameMonetize Posted November 15, 2016 Share Posted November 15, 2016 Hello 1. Correct: one big mesh will not be culled 2. You can still merge the big mesh but then subdivide it using submeshes: This way you reduce the dawcalls, use cache and keep the culling on submeshes are used in this article for instance: https://www.eternalcoding.com/?p=283 Quote Link to comment Share on other sites More sharing options...
mightymarcus Posted November 15, 2016 Author Share Posted November 15, 2016 Great, thanks. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.