2 ms·
The problem with these sorts of analytical approaches is how to handle backgrounds, depth and intersections. There are good reasons why GPUs rely on variations
by david-gpu 2y ago
The problem with these sorts of analytical approaches is how to handle backgrounds, depth and intersections. There are good reasons why GPUs rely on variations of multisampling. Even CPU-based 3D render engines use similar methods rather than analytic filters, as far as I know.
A more interesting approach to antialiasing, in my opinion, is the use of neural nets to generate aesthetically pleasing outputs from limited sample data, as seen for example in NVidia's DLAA [0]. These methods go beyond trying to optimize over-simplistic signal processing reconstruction metrics.
[0] https://en.wikipedia.org/wiki/Deep_learning_anti-aliasing https://en.wikipedia.org/wiki/Deep_learning_anti-aliasing
- adastra22 2y agoYou order your operations by depth.
- david-gpu 2y agoIf only it was so simple. You typically don't have the memory capacity and computational budget to sort and render the whole scene back to front. You can use bucketing and other tricks to try and do a better job, but at the end of they day it is just impractical. This method has been studied for decades and it is still not in common use.