Better mipmap halfband filter about dplug HOT 10 CLOSED

https://observablehq.com/@jobleonard/gaussian-kernel-calculater

from dplug.

it makes little difference it seems.

^highlights exagerated 8x.

Left: baseline 1331 4 taps
Right: handmade Magic Kernel 4 taps.
Still not anisotropic.

It seems the biggest visual issue is the bilinear sampling itself.

from dplug.

All cubic sampling (right) vs all linear sampling (left) on the 5 mipmap levels. Doing it on later levels would be a bit more expensive but also solves part of the issue. Agair both our mipmapping filter don't make a difference, I think those filters are wrong.

from dplug.

• Instead of separate convolution with 1D near-gaussian, try a "round" kernel in the first place => fitting this kernel with exp(-xxvariance) exponential leads nowhere because our kernel is anisotropic.
  The function we fit is a 2D gaussian with a variance of 0.5493

from dplug.

• Softer kernel (larger variance)? => bingo

• Sharper kernel (lower variance)? => nope
  
  With cubic-sampling and no noise.
  Left: exp(-x*x*05493 * 1.5)
  Right: exp(-x*x*05493 * 0.5)
  Now we got something.

• Dichotomize to find best value between 1 and 0.1 for this factor. => 0.35

from dplug.

Basically making the mipmap blurrier works to mask anisotropy. That would be just for diffuse (emisive effect) since depth would probably suffer from changes like this.

from dplug.

Baseline | New mipmap kernel for diffuse + one cubic sample at highest level | New mipmap kernel for diffuse + two cubic samples at highest level

aaaaand the new kernel isn't especially nicer on Graillon. It's mostly just a bit blurrier.
Let's see how much cubicSample we need first.

from dplug.

2 samples of cubic > all samples of cubic (!!!) >> 1 samples of cubic >> all linear samples >> no futurePBREmissive
futurePBREmissive gets 2 samples of cubic for the higher mipmaps then.

from dplug.

mmmm, perhaps 0.5493 * 0.35 as variance was too small, try higher. =>no, not worth it

from dplug.

futurePBREmissive has changed in a small way

from dplug.