Volumes Of Fun http://www.volumesoffun.com/phpBB3/ |
|
Coplanar face merging - OPTIMIZED! http://www.volumesoffun.com/phpBB3/viewtopic.php?f=2&t=98 |
Page 2 of 2 |
Author: | onimatrix [ Fri Dec 10, 2010 7:05 am ] | |||
Post subject: | Re: Coplanar face merging - OPTIMIZED! | |||
I got texturing to work For the moment, I scale the base UV of each quad by a merge count in the two posible axes. Setting GL_REPEAT to both texture wraps, this gives the illusion that all the voxels are textured individually. I expect to run into some problems with this solution and texture atlas (They are pretty much incompatible). Maybe instead of texture atlas I can batch all quads with the same texture. Here are some nifty pics:
|
Author: | David Williams [ Fri Dec 10, 2010 7:03 pm ] |
Post subject: | Re: Coplanar face merging - OPTIMIZED! |
Wow, excellent work again... I'm actually suprised your getting such good results and performance using the fixed function pipeline and immediate mode. But well done. onimatrix wrote: Maybe instead of texture atlas I can batch all quads with the same texture. You might find 'SurfaceMesh<VertexType>::extractSubset' to be useful, though I don't know if it works with the cubic mesh. It might be easier to write your own version which works directly with your quads. |
Author: | onimatrix [ Fri Dec 10, 2010 7:25 pm ] | ||
Post subject: | Re: Coplanar face merging - OPTIMIZED! | ||
I'm using display lists, not immediate mode. Performance wise, they are at the same level that VBOs (Sometimes even faster!). Only two drawbacks that I'm aware of: First, they can't be modified. Second, to create display lists in another thread you have to do some funky rendering context switches, as OpenGL has zero (0) threadsafeness. I got threading to load/generate/save chunks on demand, but I build the display lists in the main thread because I kind of like my current head and wouldn't like it to explode.
|
Page 2 of 2 | All times are UTC |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |