[Warner306]

In-progress (not yet added to the guide):

Low Processing

enhance detail:

Focuses on making faint image detail in flat areas more visible. It does not discriminate, so noise and grain may be sharpened as well. Not a form edge enhancement but can work well with line sharpening algorithms to provide "complete" sharpening of the image.

Medium Processing

crispen edges:

Focuses on making high-frequency edges crisper by adding light edge enhancement. The thin edge look should lead to an image that is more high-definition.

High Processing

sharpen edges:

A line/edge sharpener similar to LumaSharpen and AdaptiveSharpen. Unlike these sharpeners, sharpen edges introduces less bloat and fat edges. More aggressive than crispen edges.

thin edges:

Attempts to make edges, lines and even full image features thinner/smaller. This can be useful after large upscales, as these features tend to become fattened after upscaling. May be most useful for animated content and/or used in conjunction with SuperRes at low values.

General Usage:

Each sharpener serves a different purpose. It may be desirable to match an edge sharpener with a detail enhancer such as enhance detail. The two algorithms will sharpen the image from different perspectives, filling in the flat areas of an image as well as its angles. A good combination for 1080p content shown at 1080p might be:

sharpen edges (0.2)
enhance detail (0.5)

sharpen edges provides subtle line sharpening for an improved 3D look, while enhance detail brings out texture in the remaining image.

I have gotten good results with this combination. However, the rendering queue is too high for this combination to be used on my system at this time. Updates should improve its performance, hopefully.

Note to DSPlayer Users:

sharpen edges and thin edges are not yet performance-maximized and remain quite costly to use. None of the new algorithms have made their way into DSPlayer but can be used with the Kodi madVR GUI turned off. enhance detail can be used with integrated graphics cards without much penalty. The other algorithms are bigger resource hogs.