Efficiently managing the visual results of texture compression
Posted by Sachin Garg on 24th November 2005 | Permanent Link
Although as I am finding more and more about where all compression is applied, the amount of surprise I feel upon finding an unusual application is relatively reducing. But still every such discovery brings its fair share of excitment.
I had recently talked about how Graphics Cards Use Compression for Reducing Memory Bandwidth.
Anyway, this article is nowhere near as surprising as that one. In this Gamasutra article, Riccard Linde discusses DXT compression method which is geared towards compressing textures used in games. He details what all can be taken care of to limit artifacts and ensure minimum loss during compression.
Throughout this article we will mainly concentrate on DXT1, as it the most useful format due to its effective memory size - a 1024*1024 texture is only 603KB in size. With a compression ratio of 1:8, DXT1 keeps the memory size low by using an algorithm that divides up the texture into 4×4 pixel blocks, where each block only stores 2 unique colors in the combined 3 channels from a 65k palette. If you created a 16bit TGA texture with the same resolution, it would be 2.4MB in size.