Volodymyr Agafonkin

@tmeasday Sorry for a late reply, just noticed the issue! I haven't thought about this problem, but open to any improvements — keep me updated on your experiments.

Aren't you able to control this on your side? Run `pixelmatch` without diff, then run it again if there are differences above the threshold (which shouldn't happen often in tests,...

Currently no easy way to compare differently sized images, but it's possible to prepare for it manually if you slice out the relevant pixels from a bigger image before feeding...

No — the CLI that accompanies the library is very basic and can only compare PNGs. We might want to add JPG support though, e.g. with https://github.com/eugeneware/jpeg-js

Pixelmatch measures visual difference, not numeric one, so it blends all transparent/semi-transparent pixels over a white background. If you have transparent pixels mixed with white pixels on a white background,...

Yes, I think a checkerboard-style background would be possible and likely not impacting speed — let's look into that! Maybe we should change the white blending color to gray as...

I have the same problem because of Waves bundle (which contains TONS of plugins). Another way to approach the issue would be to allow specifying a concrete folder (instead of...

This would be very cool indeed, but is unfortunately much more difficult to implement. I don't have any plans to tackle this yet.

@nicoptere the problem with cdt2d is that it's extremely slow. I've added some benchmarks for it (non-constrained version) in the readme.

Thank you! I hope Mikola will some day get to implement his idea of a compiled fast robust arithmetic engine for JS. The ideas he described to me are amazing,...