f3kdb(clip c, int "range", int "Y", int "Cb", int "Cr", int "grainY", int "grainC", int "sample_mode", int "seed", bool "blur_first", bool "dynamic_grain", int "opt", bool "mt", int "dither_algo", bool "keep_tv_range", int "input_mode", int "input_depth", int "output_mode", int "output_depth", int "random_algo_ref", int "random_algo_grain", float "random_param_ref", float "random_param_grain") Ported from http://www.geocities.jp/flash3kyuu/auf/banding17.zip . (I'm not the author of the original aviutl plugin, just ported the algorithm to avisynth.) This avisynth plugin debands the video by replacing banded pixels with average value of referenced pixels, and optionally add grain (random dithering) to them. Supported colorspaces: YUY2, YV12, YV16, YV24, YV411 (Progressive only) Parameters: range Banding detection range. Default: 15 Y Cb Cr Banding detection threshold. If difference between current pixel and reference pixel is less than threshold, it will be considered as banded. Default: 1 (dither_algo = 0) / 64 (dither_algo > 0) grainY grainC Valid only when sample_mode is 1 or 2. Specifies amount of grains added in the last debanding stage. Default: 1 (dither_algo = 0) / 64 (dither_algo > 0) sample_mode Valid modes are: 0: Take 1 pixel as refernce pixel. grainY and grainC are ignored. (Deprecated, will be removed in future) 1: Take 2 pixels as reference pixel. Reference pixels are in the same column of current pixel. Add grain after processing. 2: Take 4 pixels as reference pixel. Reference pixels are in the square around current pixel. Add grain after processing. Default: 2 seed Seed for random number generation. blur_first Valid only when sample_mode is 1 or 2. true: Current pixel is compared with average value of all pixels. false: Current pixel is compared with all pixels. The pixel is considered as banded pixel only if all differences are less than threshold. Default: true dynamic_grain Use different grain pattern for each frame. Default: false opt Specifies optimization mode. -1: Use highest optimization mode that is supported by host CPU 0: No optimization (should be supported by almost all CPUs) 1: SSE2 (Pentium 4, AMD K8) 2: SSSE3 (Core 2) 3: SSE4.1 (Core 2 45nm) Default: -1 mt Multi-threaded processing. If set to true, U and V plane will be proccessed in parallel with Y plane to speed up processing. Default: true if host has more than 1 CPU/cores, false otherwise. dither_algo 0: 8-bit processing (Deprecated, will be removed in future) 1: No dithering, LSB is truncated 2: Ordered dithering 3: Floyd-Steinberg dithering * Visual quality of mode 3 is the best, but the debanded pixels may easily be destroyed by x264, you need to carefully tweak the settings to get better result. * Mode 1 and mode 2 don't look the best, but if you are encoding at low bitrate, they may be better choice since the debanded pixels is easier to survive encoding, mode 3 may look worse than 1/2 after encoding in this situation. (Thanks sneaker_ger @ doom9 for pointing this out!) Note: 1. In sample mode 0, only mode 0 is available (it doesn't make sense to use high precision mode). Setting this to other values will cause an error. 2. It is not recommended to use mode 0 anymore (except that you want to use sample mode 0). 3. In high precision mode, threshold and dither parameters are 64 times bigger than equivalent in low precision mode. 4. This parameter is ignored if output_depth = 16. 5. 10bit x264 command-line example: avs2yuv -raw "script.avs" -o - | x264-10bit --demuxer raw --input-depth 16 --input-res 1280x720 --fps 24 --output "out.mp4" - Or compile x264 with the patch on https://gist.github.com/1117711, and specify the script directly: x264-10bit --input-depth 16 --output "out.mp4" script.avs Default: 0 (sample_mode = 0) / 3 (sample_mode > 0) keep_tv_range If set to true, all processed pixels will be clamped to TV range (luma: 16 ~ 235, chroma: 16 ~ 240). It is recommended to set this to true for TV-range videos, since pixel values may overflow/underflow after dithering. DON'T set this to true for full-range videos, as all out-of-range pixels will be clamped to TV range. Default: false input_mode Specify source video type. 0: Regular 8 bit video 1: 9 ~ 16 bit high bit-depth video, stacked format 2: 9 ~ 16 bit high bit-depth video, interleaved format Default: 0 (input_depth = 8 or not specified) / 1 (input_depth > 8) input_depth Specify bit-depth of source video. Range: 8 ~ 16 Default: 8 (input_mode = 0 or not specified) / 16 (input_mode = 1 or 2) output_mode Specify output video type. Meaning of values are the same as input_mode. Only valid when dither_algo = 1 / 2 / 3 . If dither_algo = 0, output_mode is set to 0 and can't be changed. When output_mode = 2, frames will be 2x wider and look garbled on preview, it will return to normal after correctly encoded by high bit-depth x264) Default: 0 (output_depth = 8 or not specified) / 1 (output_depth > 8) output_depth Specify output bit-depth. Only valid when dither_algo = 1 / 2 / 3 . If output_depth = 16, dither algorithm specified by dither_algo won't be applied. Range: 8 ~ 16 Default: 8 (output_mode = 0 or not specified) / 16 (output_mode = 1 or 2) random_algo_ref / random_algo_grain Choose random number algorithm for reference positions / grains. 0: Algorithm in old versions 1: Uniform distribution 2: Gaussian distribution (StdDev (sigma) is settable through random_param_ref / random_param_grain, Only values in [-1.0, 1.0] is used for multiplication, numbers outside this range are simply ignored) Default: 1 / 1 random_param_ref / random_param_grain Parameter for the respective random number generator. Currently only have effect for the Gaussian generator. Default: 1.0 -------------------------------------------------------------------------------- f3kdb_dither(clip c, int "mode", bool "stacked", int "input_depth", bool "keep_tv_range") Downsamples high bit-depth video to regular 8-bit video using dither routines of flash3kyuu_deband. Supported colorspaces: The same as flash3kyuu_deband. Requires SSE2-capable CPUs (Pentium 4 / AMD K8 or later) Parameters: mode 0: Ordered dithering 1: Floyd-Steinberg dithering Default: 1 stacked true: Input data is stacked (like output of flash3kyuu_deband output_mode = 1) false: Input data is interleaved (like output of flash3kyuu_deband output_mode = 2) Default: true input_depth Specifies input bit depth Default: 16 Valid range: 9 ~ 16 keep_tv_range See description of keep_tv_range in flash3kyuu_deband Default: false -------------------------------------------------------------------------------- Change log 1.5.1 (2012-04-07) * Supports setting StdDev (sigma) for the Gaussian random number generator 1.5.0 (2012-03-12) * (There isn't any new feature in this version, only some parameters are modified to reduce user confusion) * ditherY/ditherC are renamed to grainY/grainC * dynamic_dither_noise is renamed to dynamic_grain * precision_mode is renamed to dither_algo, mode 4 and 5 are removed * random_algo_dither is renamed to random_algo_grain * enable_fast_skip_plane is removed, this optimization will be enabled implicitly whenever possible (Filter result won't be changed by this optimization) 1.4.2 (2011-11-10) * Fixed crash on some non-mod16 videos 1.4.1 (2011-11-05) * Fixed broken YUY2 support (still slow) * Improved default value handling of bitdepth-related parameters * precision_mode 4 / 5 are now deprecated and may be removed in future versions, you can use output_mode 1 / 2 to achieve the same result 1.4.0 (2011-10-30) * 9 ~ 16 bit-depth input/output * Related parameters: input_mode/input_depth/output_mode/output_depth * New random number generator, reference position and dither noise can be generated in uniform or gaussian distribution * Related parameters: random_algo_ref / random_algo_dither * diff_seed is replaced with dynamic_dither_noise, when enabled, noise pattern will be different for each frame * Another new parameter: enable_fast_skip_plane * Short filter alias: f3kdb * Now the ICC-compiled DLL should be runnable on pre-SSE2 systems (untested) * Several bug fixes 1.3.0 (2011-09-07) * Added x64 version * Added a downsample filter: f3kdb_dither * Internal precision is increased to 16bit * New parameter: keep_tv_range, please see readme.txt for details * Default sample_mode is changed to 2 as it is better in most cases * Fixed: Floyd-Steinberg dithering may produce incorrect result for full-range videos * Fixed: Broken YUY2 debanding * Minor optimizations 1.2.0 (2011-08-01) * Added support for YUY2 (not optimized yet) * Added support for all planar YUV format in AviSynth 2.6 * The filter is now compatible with both AviSynth 2.5 and 2.6 * 16bit output (precision_mode = 4/5) * Note: The internal processing precision is still 14bit, this will be improved in future versions 1.1.0 (2011-06-18) * Fixed a bug that high threshold values would produce incorrect result in high precision modes. * Threshold values was scaled based on other parameter in previous versions, it is unscaled now to increase flexibility. Using same value has weaker effect in new version. Effect of default parameter set is also changed. * SSE optimization for high precision mode. * Rejects some invalid parameter combination instead of silently skips them 1.0.2 (2011-06-06) * High precision mode * * (currently non-optimized, SSE routine will be added later) * Frame edges are properly processed now * Fix crash in some cases (unaligned frames are handled correctly) * Other bug fixes 1.0.1 (2011-05-27) * Multi-threaded processing * Skip planes which threshold is 0 1.0.0 (2011-05-21) * Lots of bug fix * SSE optimization, massive speed up 0.9.1 (2011-05-05) * Fix: Incorrect results when blur_first=true 0.9 (2011-05-04) * Initial release -------------------------------------------------------------------------------- Web: http://forum.doom9.org/showthread.php?t=161411 http://www.nmm-hd.org/newbbs/viewtopic.php?f=7&t=239 Acknowledgment: flash3kyuu (http://www.geocities.jp/flash3kyuu/) Source code: https://github.com/SAPikachu/flash3kyuu_deband Copyleft 2011 SAPikachu (sapikachu@gmail.com, http://www.sapikachu.net)