Original Post
hello, I have this question about Percentage closer filtering. I have implemented the algorithm but I do not know how it can actualy work. all it does is that it averages values of pixels in depth map and uses this averaged value as a shadow factor? I have implemented the algorithm and that's all it does. It takes 4 pixels from depthmap, adds them together and multiply by 1/4 and thus we have a shadow factor? I really do not get it. But it works... but how? I need to understand becouse the artefact of PCF for me is that the areas that are further in depth map, thus having higher depth value, are lighter than areas closer in depth map, and also it results in shadow variance as my shadow map changes. Also areas close to occluder are lighter then those far from occluder. What could be the work around?