warpsharp (floating point ver)

[feisty2]

source: https://github.com/IFeelBloated/warpsharp (requires c++2a compiler)
windows binary: https://github.com/IFeelBloated/warpsharp/releases

namespace: warpsf

warpsharp plugin that works on fp32 video clips, the functionality is largely the same as the one coded by jackoneill (https://forum.doom9.org/showthread.php?t=172721), except the following crap are removed.
- support for clips with subsampled chroma
- "cplace" parameter, aint no such thing as chroma placement for 4:4:4 or gray
- "opt" parameter, no I'm not interested in manual labor for SIMD optimization
- "AWarpSharp2" function, this no good shit is useless, just call the ASobel+ABlur+AWarp combo in ur script

"thresh" now accepts floating point values, the scale is still 0.0 - 256.0

[feisty2]

just tried, well I guess it also compiles with -std=c++17 but that's about the oldest possible c++ standard, -std=c++14 fails cuz I seemed to have used tons of core language features introduced in c++17 (structural bindings, constexpr if, if statement with initializer, constexpr lambdas, etc).

[ChaosKing]

Thx for fp32 support.
Could you zip/7z your release on github so it can be auto updated via vsrepo? See also https://github.com/vapoursynth/vsrepo/issues/112#issuecomment-524619652

[feisty2]

Thx for fp32 support.
Could you zip/7z your release on github so it can be auto updated via vsrepo? See also https://github.com/vapoursynth/vsrepo/issues/112#issuecomment-524619652

done.