Original Post
So I was thinking about ways to minimize the amount of voxel data required, in other words ways of compressing voxel data. What I thought of so far, is using a kind of quad tree for blank areas in the 3d world. Whereby if 8 small voxels are all blank, one larger blank voxel with dimensions of 2x2x2 is formed in memory and then 8 of these larger blank voxels, and so on. So any rays or whatever, would first traverse the large voxels, and if it is blank can skip right over it, if it is not blank, check the next level of smaller voxels it contains and so on. In this way not only do you have much much less information to store, but also alot less voxel space to traverse, which means less calculations. This was my ideas so for, any body care to expand or provide some other good ideas?
