Pixel Processing Optimisations
This is a small maintenance release focussing on performance optimisations:
-
SDL_SWIZZLEBGR
support has been enabled for builds using the official (i.e. old) GCW0 toolchain. This provides a more efficient mechanism for packaging pixels for display purposes. Since I only use the official toolchain, this improves the performance of the binary posted here :) -
Interframe blending and the x2 hardware scanline/grid scalers have been optimised using blargg's packed pixel mixing method: http://blargg.8bitalley.com/info/rgb_mixing.html. This reduces the performance overheads of these post-processing techniques by ~20-30%.