Jody Klymak

Like it or not the current style is the current style and it's hard to know whether anyone likes it or not. But our policy is pretty clear that we...

We should discuss on the call, but I'm not convinced having a "pinnable" style, but the main style changing at arbitrary points in the dev cycle is very desirable either....

> This could only be prevented by a "we never change a style ever" policy, In recent times, Matplotlib only changed styles in the 2.0-3.0 transition, with lots of notice...

Please make a reproducible minimal example. We can't test this as-is.

I think we just need a FAQ https://matplotlib.org/stable/users/faq/index.html for these sorts of issues. Basically if you have too many pixels we (or any rendering engine) have to decide which ones...

The contention is you had a rendering issue in both versions, just that it was different.

For instance in the above, at least on my screen, the 3.5.1 version has _more_ detail than the 3.3.4 version. Maybe you got them backwards, but regardless, you ideally will...

Maybe? All I remember us working of for pcolormesh was the shading, which shouldn't affect this...

@mmajewsk try setting snap=False to get the old (still aliased) behaviour.

I can confirm that `snap=False` gives the 3.3.4 behaviour. The fundamental issue still remains that pcolormesh (or `hist2d`) with more bins than pixels is going to be aliased. I strongly...