RgTools

RgTools is a modern rewrite of RemoveGrain, Repair, BackwardClense, Clense, ForwardClense, VerticalCleaner and TemporalRepair in a single plugin. RgTools is mostly backward compatible to the original plugins.

Some routines might be slightly less efficient than original, some are faster. Output of a few RemoveGrain modes is not exactly identical to the original due to some minor rounding differences which you shouldn't care about. Other functions should be identical.

This plugin is written from scratch and licensed under the MIT license. Some modes of RemoveGrain and Repair were taken from the Firesledge's Dither package.

v1.2 (20210526)

• Fix: garbage in 10+ bit Backward/ForwardClense for non SSE4.1 processors (C code part)

v1.1 (20210104)

• Source syntax for GCC
• CMake build environment, builds on Linux, at least on my (pinterf) Ubuntu 19.10 WSL (INTEL_INTRINSICS handling not implemented in the source, so it compiles on Intel at the moment)
• Fix: artifacts in RemoveGrain mode 19 for 10-16 bits

v1.0 (20200427)

• RemoveGrain mode 26-28, (C, up to SSE4.1) by the original author: Mode 26 SmartRGC: 26 = medianblur. Based off mode 17, but preserves corners, but not thin lines. Mode 27 SmartRGCL: 27 = medianblur. Same as mode 26 but preserves thin lines. Mode 28 SmartRGCL2

  I swear that by seeing the code mode 28 is different from mode 27 but I cannot find a clip where 27 and 28 gave different results :)

• Repair mode 26-28 (25 is nonexisting, only exists in RemoveGrain)

• add "opt" debug parameter to Repair. 0:auto, 1:c 2:SSE2 3:SSE4.1 4:AVX2 (latter is not used here atm)

v0.99 (20200426)

• RemoveGrain mode 25 ("nondestructivesharpen") finally reverse engineered and ported from the original RemoveGrain package. (no-no, Repair has still 24 modes) Plus: 10-32 bits, Plus: sse4.1, avx2
• RemoveGrain: Revised, more correct 32 bit float in mode 6,7,8,9,15,16,23,24 (proper clamping, chroma part specially treated)
• General: make C reference more similar to actual SIMD code
• fix: width<32 (or 16) with horizontal subsampling would use AVX2 (SSE2) on chroma
• (noticed: RemoveGrain: mode 11 and 12 SSE is simply avaraging instead of really do 1/4 1/2 1/4 kernel blur like C (todo). This is written in docs but even very old RemoveGrainSSE2 acts like this)
• Avisynth+ V8 header, frame property copy support
• opt debug parameter to RemoveGrain. 0:auto, 1:c 2:SSE2 3:SSE4.1 4:AVX2. Replaces bool "optAvx2"

v0.98 (20190814)

• Include "TemporalRepair" filter from old RemoveGrainT package (rewritten C and SIMD intrinsics from pure inline asm) Add Y8, YV16, YV24 besides YV12, drop YUY2 support. Add 10-32 bit support for Y, YUV and planar RGB formats Add int "opt" parameter (mainly for debug: 0=C 1=SSE2 2=SSE4.1) for testing specific code paths
• Codes for different processor targets (SSSE3 and SSE4.1) are now separated and are compiled using function attributes (clang, gcc).
• Other source changes for errorless gcc and clang build
• LLVM support, see howto in RgTools.txt Note: use at least LLVM 9.0 build 21 June 2019 due to a clang compiler bug (_mm_avg_epu8 related, fixed on April 14 2019) older versions are up-to 1/3 slower than the Microsoft build. See latest snapshot builds at https://llvm.org/builds/
• GCC 8.3 support, CMakeFiles.txt, see howto in RgTools.txt
• RemoveGrain/Repair different code paths for SSE2/SSE4.1/AVX2 instead of SSE2/SSE3/AVX2.
• Add documentation (from old docs, new part: gcc/clang howto)
• todo: Overview 32 bit float parts for some RemoveGrain modes (6, 7, 8, 9, 15, 16, 23, 24)

v0.97 (20180702)

• Remove some inherited clipping to 0..1 range for 32bit float. In general we do not clamp in 32 bit float colorspaces, especially that U/V range is -0.5..+0.5 from Avisynth+ r2728

v0.96 (20170609)

• RemoveGrain: AVX2. Available when Avisynth+ reports AVX2 usability Can be disabled with new parameter: optAvx2=false
• Clense, ForwardClense, BackwardClense: ignore planar colorspace checking when planar=true. Like in RemoveGrain and Repair.
• Fix: Mode 11 and 13 for 32 bit float colorspaces (which worked like mode 10 and 12)

v0.95

• Fix: RemoveGrain Mode 20: overflow at 14 and 16 bit depths in SSE4 (stripes)
• Repair: error on unaligned frames (unaligned crop) instead of access violation error

v0.94

• Clense: new parameter (from v0.9): bool reduceflicker (default false)
• Clense: dummy compatibility parameters: bool planar, int cache
• autoregister filter MT modes as NICE_FILTER for Avisynth+ except for Clense: when reduceflicker is true, MULTI_INSTANCE MT mode is reported
• alignment check in Repair and RemoveGrain (anti-unaligned crop measures)

v0.93: new pixel formats are supported

• 10, 12, 14, 16 bit and float
• Planar RGB, RGBA and YUVA (alpha plane is copied)

v0.92:

• tp7, high quality port of original source full with inline asm. C, SSE2 intrinsics. Modes 1-24

Functions

RemoveGrain(clip, int "mode", int "modeU", int "modeV", bool "planar", int "opt")

Purely spatial denoising function, includes 28 different modes. Additional info can be found in the wiki. Parameters "planar" is dummy, exists for compatibility reasons.

Repair(clip, clip, int "mode", int "modeU", int "modeV", bool "planar", int "opt")

Repairs unwanted artifacts from (but not limited to) RemoveGrain, includes 28 modes, except mode 25. Parameters "planar" is dummy, exists for compatibility reasons.

Clense(clip c, clip "previous", clip "next", bool "grey", bool "reduceflicker", bool "planar", int "cache")

Temporal median of three frames. Identical to MedianBlurTemporal(0,0,0,1) but a lot faster. Can be used as a building block for many fancy medians. If reduceflicker is true, the (n-1)th source frame is reused from the previous "clensed" frame, that the filter stored internally. This works however only if Clense is getting frame requests sequentally. Parameters "planar" and "cache" are dummy, they exist for compatibility reasons.

ForwardClense(clip c, bool "grey", bool "planar", int "cache")

Modified version of Clense that works on current and next frames. Parameters "planar" and "cache" are dummy, they exist for compatibility reasons.

BackwardClense(clip c, bool "grey", bool "planar", int "cache")

Modified version of Clense that works on current and previous frames. Parameters "planar" and "cache" are dummy, they exist for compatibility reasons.

VerticalCleaner(clip c, int "mode", int "modeU", int "modeV", bool "planar")

Very fast vertical median filter. Has only two modes. Parameter "planar" is dummy, exists for compatibility reasons.

TemporalRepair(clip c, clip c, int "mode", int "smooth", bool "grey", bool "planar", int "opt")

In the same way as Repair is derived from RemoveGrain, TemporalRepair is derived from Clense. While Repair is a spatial filter most suitable for removing artifacts of temporal filters like Clense, TemporalRepair is a temporal filter, primarily useful for restoring static (non moving) details of spatial filters like RemoveGrain. Parameter "planar" is dummy, exists for compatibility reasons.