Andrzej Więckowski, Ph.D.

> Btw. A single test seems to fail with this PR. I'm not quite sure why, but I think it is due to floating point errors, and that the test...

I've also conducted a rough benchmark comparing the performance of `float2` vs `double2` for *classic* triangulation, and it seems that `float2` is slightly faster. Might be helpful for someone. 

Okay, I believe I've completed the review. I'll take another look after you address my issues.

Hi Aron! How's it going? I recently pushed some commits to `main`, and now the `BurstTriangulator` finally has proper documentation, including both a manual and API [**page**](https://andywiecko.github.io/BurstTriangulator/). Any updates regarding...

> Hi Aron! > > How's it going? > > I recently pushed some commits to `main`, and now the `BurstTriangulator` finally has proper documentation, including both a manual and...

@HalfVoxel > I have rebased on your latest commits now, and addressed the review comments. WOW, there are a lot of commits here! In the future, please consider smaller PRs....

Hi @HalfVoxel, How's it going? I've recently made huge progress with the package and made [`v3`](https://github.com/andywiecko/BurstTriangulator/releases/tag/v3.0.0) release! Introducing `int2` support at this stage should be very easy. Please let me...

Closing, PR feature moved to #190.

Hi @anzago, Recently, I've managed to release version [**`v3`**](https://github.com/andywiecko/BurstTriangulator/releases/tag/v3.0.0), which resolves your issue. Triangulator is now based on `double2`. I added commit 233a978 with a test to check this particular...

Hi @katastorm, Thank you very much for your contribution to the project! Could you please provide me with some code that reproduces the bug (at least input data)? I will...