Alberto

An update on this as i have some updated benchmarks and i can finally show how different changes impact the overall performance benefits of this new implementation. ___ **Note** Currently...

Update on performance. After careful and more general testing i can say that (Best Case: all rects are normalized) the mean improvement across all tests is 41.6%. The smallest performance...

Some newer performance tests: **OLD** ``` CUMULATIVE TIME: COLLIDELIST | 0.0465923 COLLIDELISTALL | 0.1035984 UNIONALL | 0.0792499 UNIONALLIP | 0.0672198 COLLIDEOBJECTS | 0.1883542 COLLIDEOBJECTS_ALL | 0.4826421 ``` ``` MEAN: COLLIDELIST...

Plus having new objects( sneak peek from Starbuck) like circle and line would simplify functions quite a bit both in implementation and in signature. (right now i can only show...

Cool! was thinking about it myself! Should we open an issue on this?

I totally agree! I'll link some PRs to this in a moment. Also I'd like to make a little list for where to use `FASTCALL`, as it might be helpful....

> I have these questions to help us decide if the trade off of not using keyword arguments is worth it for us: We are not saying to stop using...

I can take this one. Problem is a surface.fills() function can be either of: - given a list of surfaces it fills them based on color and rect passed: like...

You are suggesting number 2 but number 1 and 3 are also possible

I'll go with number 3 as it gives you more freedom over how you fill the surfaces.