RgTools

Latest revision as of 21:05, 25 June 2020

[edit] Description

RgTools is a modern rewrite of RemoveGrain, Repair, BackwardClense, Clense, ForwardClense, VerticalCleaner and TemporalRepair all in a single plugin. RgTools is mostly backward compatible with the original plugins, incompatibilities mainly come from deprecated parameters in the original filters, read filter documentation for full details. Some routines might be slightly less efficient than original, some are faster. Output of some RemoveGrain modes are not exactly identical to the original due to some minor rounding differences; these differences will not affect your output in any significant way (it might make it better)[1]. Other functions should be identical.

[edit] Requirements

• [x86]: AviSynth+ or AviSynth 2.6
• [x64]: AviSynth+
• Supported color formats: Y8, YV12, YV16, YV24, YV411
  • AviSynth+: All planar formats (8/10/12/14/16/32-bit, Y, YUV(A), and RGB(A) with or without alpha) are supported.

• AVX2 support requires AviSynth+ (also OS dependent, AVX2 needs at least Windows7 SP1)

• Microsoft Visual C++ 2019 Redistributable Package (x86 / x64)

*** vcredist_x86.exe is required for RgTools-x86

*** vcredist_x64.exe is required for RgTools-x64

[edit] Filters

[edit] Changelog

Version      Date            Changes

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
                               Though mode 28 has different code from mode 27 so far no clips found producing 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        08/14/2019      - 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        07/02/2018      - 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        06/09/2017      - 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        03/14/2017      - 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        01/26/2017      - 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        11/27/2016      - Same functionality as 0.92.1
                             - New bit depths 10, 12, 14, 16 bits and 32 bit float
                             - New formats: Planar RGB, Planar RGBA and YUVA support
                             - Use aligned loads if possible, internally
                             - 10+ bit formats require SSE4 for fast processing
                             - XP Support
                             - Built with VS2015, requires Visual Studio Redistributable Update 3 or newer
                             - Of course it works with classic Avisynth for 8 bit videos.

v0.92.1      02/10/2014      - Planar parameter added to VerticalCleaner registration function. It worked before but apparently not.

v0.92        02/10/2014      - Planar parameter added to VerticalCleaner for compatibility with QTGMC.
                               Same behavior as in RemoveGrain.

v0.91        11/30/2013      - Planar parameter is added to RemoveGrain and Repair. It allows to skip any colorspace 
                               checks whatsoever, making the filter always assume that planar input is passed.

v0.9         11/26/2013      - Initial release
                             - 24 modes of RemoveGrain, 24 modes of Repair, both modes of VerticalCleaner, and three 
                               variations of Clense implemented.

Back to External Filters ←