Starting a Database of *useful* avisynth scripts .avs

[Randall]

I think it would be a great idea if someone (I guess me) would start a database or a website that would have many useful AVIsynth scripts that we could all benefit from. There are script gurus out there that can make just about any source turn into gold, and then there are the script kiddies (me) that hack away at the most basic of scripts and wonder how they managed to fux0r up a perfectly good .vob source. I know there are some tutorial scripts out there that show some useful features of avisynth, but if you just can't be bothered with countless hours of experimentation, then it would be helpful to have a bunch of proven scripts to use as a general guidline to success. A major roadblock to this approach is that no two video sources are identical, so what worked for you might not work for me, and what works for NTSC probably won't work for PAL sources etc. But my idea is we can break it down into categories, for example we could have a category of scripts for Film, TV Video, Teleclined Video, and then apporpriate subcategories like Interlaced or Progessive Scan, then FPS, and finally Aspect Ratio. Then people would find any pre-built scripts that are available far more useful to his/her particular senerio. A heirchy like this:

Video
|-- Interlaced
| `-- 29.97 FPS
| `-- available scripts found here (with Deinterlacing advice)
|-- Progressive
| `-- 29.97 FPS
| `-- available scripts found here
|-- Teleclined
| `-- 29.97 FPS
| `-- available scripts found here (with Reverse Telecline advice)


Here are my home made and commonly used scripts for NTSC 29.97 FPS Video sources that are FULL SCREEN 4:3 Aspect Ratios (Sitcoms, Cartoons, made for TV stuff in Region 1, etc.):
Here is my example for a NTSC 29.97 FPS Progressive Scan Video this format is ideal, we don't need to do anything except resize and clean:
# LOAD THE PLUGINS
LoadPlugin("C:\Program Files\DGMPGDec\DGDecode.dll")
LoadPlugin("C:\Program Files\AviSynthPlugins\UnDot.dll")

# VIDEO SOURCE
mpeg2source("C:\VIDEO_TS\test.d2v")

# CROP THE IMAGE
crop(42,0,636,480)

# RESIZE (SHARP) FULLSCREEN SOURCE
LanczosResize(512,384)

# CLEAN THE PICTURE
Undot() Here is my example for a NTSC 29.97 FPS Teleclined Video (we use Telecide filter to pull down the frame rate to 23.976 FPS, so be aware of the frame rate change) This will remove the Teleclined mice teeth artifacts that were left behind (common on The Simpson’s as well as the Seinfeld DVDs, and many many other 'made for TV' programs):# LOAD THE PLUGINS
LoadPlugin("C:\Program Files\DGMPGDec\DGDecode.dll")
LoadPlugin("C:\Program Files\AviSynthPlugins\decomb.dll")
LoadPlugin("C:\Program Files\AviSynthPlugins\UnDot.dll")

# VIDEO SOURCE
mpeg2source("C:\VIDEO_TS\test.d2v")

# INVERSE TELECLINE OPTIONS
Telecide(order=1,guide=1).Decimate()

# CROP THE IMAGE
crop(42,0,636,480)

# RESIZE (SHARP) FULLSCREEN SOURCE
LanczosResize(512,384)

# CLEAN THE PICTURE
Undot() Here is my example for a NTSC 29.97 FPS Interlaced Video Some DVDs never even bother to deinterlace the source, relying on your Progressive Scan DVD players to do the dirty work for you. Unfortunately we can't use a hardware deinterlacer, but TDeint and BlendBob are some fine filters that will do it for us.:# LOAD THE PLUGINS
LoadPlugin("C:\Program Files\DGMPGDec\DGDecode.dll")
LoadPlugin("C:\Program Files\AviSynthPlugins\TDeint.dll")
LoadPlugin("C:\Program Files\AviSynthPlugins\BlendBob.dll")
LoadPlugin("C:\Program Files\AviSynthPlugins\UnDot.dll")

# VIDEO SOURCE
mpeg2source("C:\VIDEO_TS\test.d2v")

# DEINTERLACE OPTIONS
TDeint(mode=1,order=1,mtnmode=1)
BlendBob()

# CROP THE IMAGE
crop(42,0,636,480)

# RESIZE TO FULLSCREEN 4:3 ASPECT RATIO
LanczosResize(512,384)

# CLEAN THE PICTURE
Undot()All of these avisynth filters can be downloaded at http://www.avisynth.org/warpenterprises/ if you don't have them. Make sure you give them the proper path when you LoadPlugin("Path to your plug-in here").

We should all try to comment our sripts in a readable mannor, so that others can make enhancements and changes where necessary. We could license these scripts under LGPL or Creative Commons, so everyone can benefit. If anybody thinks that this is a good idea, then please post some of the scripts that you feel would be quite useful to other people on this thread. If we get enough avisynth scripts, then perhaps we could look for a general hosting solution or central repository for all to benefit from. :script: :)

[Fizick]

http://www.avisynth.org/ShareFunctions

[Randall]

So then the general consensus is that this idea is no good? Thanks for pointing me toward the shared functions, but I was hoping that we could get everybody's scripts together in some kind of archive. Functions are useful, but they aren't scripts.

[raquete]

for my taste is very good Randall. :goodpost:

[scharfis_brain]

there are no general scripts.

one only can to some script-snipplets, cause the situations aways differ.

that's why I am doing a lot AVS functions ;)

[Randall]

there are no general scripts.

one only can to some script-snipplets, cause the situations aways differ.

that's why I am doing a lot AVS functions ;)
I know exactly what you mean. That's what I mentioned in my first post. So that is why I propose to break the script listing up into categorys like I mentioned until you finally get to a specific script that will suit you. It seems easier then writing a script for say "The Simpsons Season 1" and another for "Terminator 2" etc. Plus, we could get around any legal issues of writing scripts specifially for a certain copyrighted source, and instead merely suggest that this script X will work beautifully for movie Y, etc etc.

@scharfis_brain: can you post those dolby downmix functions you wrote directly into an .avs script or does it have to be an *.avsi file?

[scharfis_brain]

@scharfis_brain: can you post those dolby downmix functions you wrote directly into an .avs script or does it have to be an *.avsi file?

Uhm no. Why? The function code listing is fully sufficient.

[Randall]

No you misunderstood me. I was asking if anybody could just use those functions as is in a regular .avs file as opposed to an .avsi file?

[Fizick]

I am still think, that additions to "shared functions will be very useful.
It is almost a script, and you always can add function example using as full script.
I suggest to everbody to contibute to http://www.avisynth.org/ShareFunctions
(how many "to" in my "English" ! :))

[Randall]

I am still think, that additions to "shared functions will be very useful.
It is almost a script, and you always can add function example using as full script.
I suggest to everbody to contibute to http://www.avisynth.org/ShareFunctions
(how many "to" in my "English" ! :))

It's a good start, but half of those functions are 404'd anyway, and outdated. I think that full scripts would be more useful if they were mapped to a source that can be used for it. For example, If I wanted to encode my Seinfeld Season 3 DVD, I would use my NTSC 29.97 FPS Teleclined Video script that I posted at the top of this thread, since it is progressive and teleclined fullscreen. I can think of maybe 15-20 simple scripts like this, that would cover just about every DVD out there in the NTSC format. The hard part is getting the aspect ratio right on widescreen sources. And even then there are only a dozen commonly used ratios.

[Fizick]

May be, some new steaky thead will be useful, with links only to others topics and posts with such scripts. Without any discussions.
Probably your classification may be some basis. But there are many others possible classifications too, for example with dependence on source (VHS, DV,DVD,TV),
noise level, blendings, etc, etc.

[mg262]

Some DVDs never even bother to deinterlace the source, relying on your Progressive Scan DVD players to do the dirty work for you:confused: I don't see that it's less legitimate to provide interlaced content on a DVD -- many people will be watching it on a standard TV, after all.

All your functions presuppose cropping by particular amounts and resizing to a fixed target resolution (512 x 384). But there's no reason to believe that everyone will always want this target resolution... and also whether you crop and/or letterbox, and by how much, will depend not just on the source and target aspect ratios but also on the source. Leaving that last point aside, there are too many possible target resolutions to capture all the relevant scripts in some sort of tree diagram.

But IMO the biggest point is that many of the steps are largely independent. So for example there are many many possibilities for denoising (Undot() is extremely minimal), and in many cases you will be able to use the same denoising method regardless of whether your source was originally telecined or not... but the tree-diagram method you suggest would require one script either way, i.e. 2x the number of scripts.

I'm not trying to shred your suggestion... I think that what you suggest could still be useful, if it were limited in scope. In particular, what you seem to be concerned with is the aspect ratio and I think that having example aspect ratio scripts is a decent idea... although of course people might still want to vary the target resolution.

[Randall]

:confused: I don't see that it's less legitimate to provide interlaced content on a DVD -- many people will be watching it on a standard TV, after all.

All your functions presuppose cropping by particular amounts and resizing to a fixed target resolution (512 x 384). But there's no reason to believe that everyone will always want this target resolution... and also whether you crop and/or letterbox, and by how much, will depend not just on the source and target aspect ratios but also on the source. Leaving that last point aside, there are too many possible target resolutions to capture all the relevant scripts in some sort of tree diagram.

But IMO the biggest point is that many of the steps are largely independent. So for example there are many many possibilities for denoising (Undot() is extremely minimal), and in many cases you will be able to use the same denoising method regardless of whether your source was originally telecined or not... but the tree-diagram method you suggest would require one script either way, i.e. 2x the number of scripts.

I'm not trying to shred your suggestion... I think that what you suggest could still be useful, if it were limited in scope. In particular, what you seem to be concerned with is the aspect ratio and I think that having example aspect ratio scripts is a decent idea... although of course people might still want to vary the target resolution.

Those are very good points. I guess my idea is a little ambitious, but I felt obligated to get a standard guideline or something going for all the people out there that would like to recode their content. I know there are many aspect ratios out there, and countless formats and situations. This would make any kind of scripting archive very difficult to implement, as many of you pointed out. How about instead of a complete script archive, we could start a "snippet" archive that points out useful techniques for certain situations. This will broaden the scope considerably, while still keeping a good goal in mind. Let's try it out here, and then if we get enough snippets, I can start to try and find a good method of archiving them, with the help of anybody interrested. I suggest commenting your snippet with what it does, some of the options you use, and which avisynth plugin(s) are required for it to function. For example...

NTSC 29.97 FPS Teleclined Video# INVERSE TELECLINE requires decomb.dll plugin
# Uses Telecide filter and Decimate() to pull down teleclined footage
# Restores to original framerate of 23.976 FPS
Telecide(order=1,guide=1).Decimate() NTSC 29.97 FPS Interlaced Video# DEINTERLACING requires TDeint.dll and BlendBob.dll plugins
# TDeint() is a bi-directionally, motion adaptive (sharp) deinterlacer
# BlendBob() tries to match fields from the same original frame and
# blends them to attempt to restore the original quality.
TDeint(mode=1,order=1,mtnmode=1)
BlendBob()

[mg262]

something going for all the people out there that would like to recode their content.I think a big problem in this regard is that there aren't actually thorough introductions to AVISynth out there -- there's bits in the analog capture guide, and a little on the website, and a reference-y bit in the manual, but I'm not sure any of those are very accessible. I think we need a guide with lots of pictures, both stills of videos and pictures to demonstrate how the script behaves like a pipeline. I'd love to write something like this, but I have never found the time...

[Randall]

I think a big problem in this regard is that there aren't actually thorough introductions to AVISynth out there -- there's bits in the analog capture guide, and a little on the website, and a reference-y bit in the manual, but I'm not sure any of those are very accessible. I think we need a guide with lots of pictures, both stills of videos and pictures to demonstrate how the script behaves like a pipeline. I'd love to write something like this, but I have never found the time...
Yeah I think you hit the nail on the head there. Not enough documentation out there to work with. Time is an issue with me too. Maybe if we got something started like wikimedia for avisynth, where everyone that is knowledgable could contribute a useful demo or workable script snippet...

[laserfan]

Randall I am a Rank Amateur in the use of avisynth and filters/plugins, having done only a handful of attempts thus far.

What I have found most impossible about the various sites/links/information, including those mentioned here already (warp, sharefunctions) is that they are organized by Filter and the Filter description, not by the Function they perform. What on Earth are TomsMoComp, Deen, and DeKafka for example!!?? If only someone would organize these by their FUNCTION--de-noisers, de-interlacers, re-sizers, etc.

We newbies are expected to learn by examples, and there are hundreds and hundreds posted here, many are incomplete, and even more than that are uncommented and give a reader only the vaguest idea of what the various functions are actually intended to do.

The couple of examples you posted intrigue me already, as first they tell what they intend to DO, then individual filters are commented as to their function (though these could be elaborated upon some more I think). I saw another script somewhere about a color correction filter--it had various choices (comment-out the ones you don't want) labeled according to source (one line was for a VHS tape, another for a Laserdisc). Great, very clearly done.

Sorry I'm rambling a bit now, but to me the ultimate website would be one where I could go to look for ideas for Sharpening, or Removing Grain, or Color Correction, and it would have options, samples, examples, and even before-and-after images showing the effects. Perhaps if such website got traction, users would post their successes "here's what I used to fix the grainy, 4x3 letterboxed DVD of 'Quiz Show' to full anamorphic glory". I can dream...

[Randall]

Randall I am a Rank Amateur in the use of avisynth and filters/plugins, having done only a handful of attempts thus far.

What I have found most impossible about the various sites/links/information, including those mentioned here already (warp, sharefunctions) is that they are organized by Filter and the Filter description, not by the Function they perform. What on Earth are TomsMoComp, Deen, and DeKafka for example!!?? If only someone would organize these by their FUNCTION--de-noisers, de-interlacers, re-sizers, etc.

We newbies are expected to learn by examples, and there are hundreds and hundreds posted here, many are incomplete, and even more than that are uncommented and give a reader only the vaguest idea of what the various functions are actually intended to do.

The couple of examples you posted intrigue me already, as first they tell what they intend to DO, then individual filters are commented as to their function (though these could be elaborated upon some more I think). I saw another script somewhere about a color correction filter--it had various choices (comment-out the ones you don't want) labeled according to source (one line was for a VHS tape, another for a Laserdisc). Great, very clearly done.

Sorry I'm rambling a bit now, but to me the ultimate website would be one where I could go to look for ideas for Sharpening, or Removing Grain, or Color Correction, and it would have options, samples, examples, and even before-and-after images showing the effects. Perhaps if such website got traction, users would post their successes "here's what I used to fix the grainy, 4x3 letterboxed DVD of 'Quiz Show' to full anamorphic glory". I can dream...
This is exactly the kind of thing we should be after. Well commented, documented, and sorted by functionallity. Thank you for your input laserfan. :thanks:

OMG I just saw your letterboxed to anamorphic source, and this is something that we could make a nice guide to. I have Quiz Show, but sadly not on DVD, and no way of editing the VHS at the moment. Essentually you want to crop out the black boxes, and resize accordingly. I had the same issue with Top Gun, and I wish I still had the script I wrote for it. D'oh! I guess you could try
RemoveGrain(mode=4,modeU=2)
Crop(0,60,-0,-60)
LanczosResize(720,480)
LimitedSharpen(ss_x=1.0,ss_y=1.0,Smode=3,strength=40)RemoveGrain() works well with Limitedsharpen() for removing fine grain. This will remove the grain, crop out what I think will be the height of the two black bars, and then resize to a 16:9 aspect ratio. You get the idea. The settings I've found to be optimal for average grain in DVDs. You can make any adjustments from the plugin documentation that you see fit for the best results on Quiz Show. Cheers.

P.S. more about Limitedsharpen() here (http://forum.doom9.org/showthread.php?s=&threadid=84196).

[mg262]

laserfan,

Have you seen this? Plug-ins organised by function.

http://www.avisynth.org/Section+3%3A+Filters%2C+plugins+and+colorspaces#q3.5

Having said that, I think that it's much preferable to have such material collected on this forum rather than the AVISynth website -- the website is updated piecemeal, as someone public-spirited takes it upon themselves to do it, but that generally means it's not that up-to-date.

I think a decent start would be to collate a list of types of filter, where the types are defined to solve particular tasks. For example,deinterlacers, IVTC filters, resizers, frame rate changers, denoisers, sharpeners, second-order filters (filters supporting other filters, e.g. motion compensation), scratch removal, grain removal, chroma correction, field blending removal, rainbow removal, dotcrawl removal, stabilisers.

In fact, having thrown those out, it occurred to me that most of them are intended to remove some particular effect -- e.g.dotcrawl, jitter, telecine. So perhaps one could organise the main body of filters by the effect they are meant to remove.

Edit: avsi was explained in gory detail here: http://forum.doom9.org/showthread.php?t=103594

[Wilbert]

Why not update http://www.avisynth.org/Section+3%3A+Filters%2C+plugins+and+colorspaces#q3.5 (note more recent descriptions can be found in the install docs ...\AviSynth25\Docs\english\externalfilters), and use more categories? It's at least a start.

Of course if nobody updates http://www.avisynth.org/ShareFunctions, those functions will be 404 and outdated.

These are two things which can be done without much work (relatively speaking :)). We just need some volunteers! If people have other *concrete* ideas i would be glad to help.

[smok3]

cough, has anyone done a nice countdown clock (5s is enough) that would nicely scale to random video source, so i get '5s clock + Ns somevideo + 2s of black' for example?

(clock should have some sort of text place for title and duration as well - where title is generated from filename and duration provided automagically by avisynth)

[Randall]

cough, has anyone done a nice countdown clock (5s is enough) that would nicely scale to random video source, so i get '5s clock + Ns somevideo + 2s of black' for example?

(clock should have some sort of text place for title and duration as well - where title is generated from filename and duration provided automagically by avisynth)
You mean a Counting Leader? Try this, Counting Leader MPEG1 (2.8 MB) (http://www.archive.org/download/Universal_Counting_Leader/universal_counting_leader_vbr.mpg) right mouse click and choose save as... ect.

[Mug Funky]

hehe. i like the old fashioned countdowns. then there's the British kind where it's a picture of a clock and the hand moves toward 0 every field (quite useful actually).

then there's the old film kinds with sideways letters, upside-down numbers and 1kHz bursts.

btw, as far as a heirarchical database of script options, i think it's a good idea. even though there are no general scripts, there is a general method that can be followed to find the right script - it could be like windows help troubleshooters, but actually useful...

something like:

1. what is your destination format (choose DVD backup, 1 CD backup, streamable, etc).

2. what is your source (choose analog video, film, DVD, DV, etc)

depending on that you could then ask questions about the source and construct a script based on the "questionnaire".

[Randall]

Getting back on track... here are some more great script snippets that I have used before.

Smoothing filter - A smoothing filter can hide compression artefacts or compensate for other sources of noise and/or color errors.

Deen - 6 filters in one. Can operate in 2d or 3d space with a variety of filtering types. My personal favorite smoother.

# SMOOTHER requires deen.dll plugin
# Options for are deen(filter type, area, luma threshhold, chroma threshold)
# "c2d" - basic 2d smoothing.
# "c3d" - convolution 3d... smooths both within the frame and from frame to frame.
# "a2d" - smoothing optimised for areas of colour like in anime.
# "a3d" - same as a2d but also smooths from frame to frame.
# "w2d" - smoothing weighted by distance in order to retain gradients.
# "w3d" - like w2d but also smooths from frame to frame.
# area - can go from 1-4, with 4 being a 9x9 pixel area.
# luma threshold - default 7. Be careful here, too high will damage details.
# chroma threshold - default 9. Generally you can go higher then luma here.
#deen("c3d",2,7,9) #setting for Film, Video, and DVDs that need it.
deen("a3d",4,8,10) #setting for Anime and Cartoons that need smoothing. The number that comes after the filter type is the area that is to be smoothed. With the "2d" filters this are can go from 1 to 7 (with 7 being a 15x15 area) and when using the 3d filters this value can go as high as 4 (9x9).

The third variable, 8, represents the luma threshold, increasing it will increase the amount of smoothing on the luma part of the image. The last argument, 10, represents the chroma threshold. The defaults are 7 for luma and 9 for chroma. Generally you can smooth chroma more than you can luma without damaging details.

[Randall]

hehe. i like the old fashioned countdowns. then there's the British kind where it's a picture of a clock and the hand moves toward 0 every field (quite useful actually).

then there's the old film kinds with sideways letters, upside-down numbers and 1kHz bursts.

btw, as far as a heirarchical database of script options, i think it's a good idea. even though there are no general scripts, there is a general method that can be followed to find the right script - it could be like windows help troubleshooters, but actually useful...

something like:

1. what is your destination format (choose DVD backup, 1 CD backup, streamable, etc).

2. what is your source (choose analog video, film, DVD, DV, etc)

depending on that you could then ask questions about the source and construct a script based on the "questionnaire".

That is a brilliant idea. I could make it platform independant too with C++ and make it a console based application that asks the questions. Although an html based version with example pictures would probably be far more useful.

[Randall]

Does anybody have any great filter snippets that they would like to contribute to this thread? So far I'm a one man band over here. :) :script:

[laserfan]

laserfan,

Have you seen this? Plug-ins organised by function.

http://www.avisynth.org/Section+3%3A+Filters%2C+plugins+and+colorspaces#q3.5

...it occurred to me that most of them are intended to remove some particular effect -- e.g.dotcrawl, jitter, telecine. So perhaps one could organise the main body of filters by the effect they are meant to remove.In fact I never have seen that page. Thanks mg262, I immediately bookmarked it.

Love your idea too of organize by "effect"--or maybe that is something for a true Relational Database, Randall.

Yah in fact I *have* converted a few 4:3lb to anamorphic already, with filters as you suggested, though most recently my "Quiz Show" conversion was done automagically by DVD-RB Pro. This proggie is kind of a swiss army knife for converting DVDs to DVD-5, with some flexibility to incorporate avisynth plugins. But I digress...

[stickboy]

No you misunderstood me. I was asking if anybody could just use those functions as is in a regular .avs file as opposed to an .avsi file?There is no difference between an .avs file and an .avsi file.

[smok3]

randall, no i would need 'resolution independant/automagic' solution, i can easily make counting leader using premiere everytime thought.

[krieger2005]

There is no difference between an .avs file and an .avsi file.
There is a difference: in the avisynth-auto-load-plug-folder. The avsi-file will be loaded whlie the avs-file not. But i don't know any more differences.

[DryFire]

There is a difference: in the avisynth-auto-load-plug-folder. The avsi-file will be loaded whlie the avs-file not. But i don't know any more differences.

I believe randall's saying he wants actual scripts not funtions which are usually put in .avsi instead of .avs (unless you like typing import("...") multiple times.


#How I normally deal with hybrid anime content on a DVD
#that is not field blended
mpeg2source("...")
Dedot()
tfm(mode=3,pp=7, chroma=true) #pp=7 produces softer lines like Tdeint mode=3 IIRC
Tdecimate(mode=1)

#denoising (usually varies source to source)
removegrain()
fluxsmoothST()
deen()
unfilter(-5,-5)
crop() #varies too much
billinearresize() #varies too much
dull = last
sharp= dull.LimitedSharpen(ss_x=2, ss_y=2, smode=2, strength=75)
soothe(sharp, dull, 35)


I'm sure I messed something up there but I literally just woke up.

edit: just some cleaning up.

[Randall]

Thank you for contributing DryFire :)

[Randall]

A sharpening filter. Sharpening typically requires a high bitrate to work with, meaning that the original source should be a DVD vob or something with plenty of bitrate to work with. The example I'm giving you is all default values except strength=40. I have found these settings to be very versitile and good for just about any source with enough bitrate. LimitedSharpen() is not like other sharpeners that usually produce a lot of artifacts. This plugin is not a typical .dll plugin file, but it does require Masktools.dll and WarpSharp.dll to be loaded. You can download LimitedSharpen.avsi here (http://www.avisynth.org/LimitSharpen). Make sure that you right click, and save the text as LimitedSharpen.avsi and put in your plug-ins directory. When you are writing your avisynth script, you can load LimitedSharpen() function using the line: Import("path to the plug-ins\LimitedSharpen.avs"). Good luck.

Limited Sharpen
# SHARPENER requires Masktools.dll and WarpSharp.dll
# You also must download the LimitedSharpen.avsi plug-in, and
# Import("path to the plug-ins\LimitedSharpen.avsi")
# LimitedSharpen() has many options available in the documentation.
# ss_x - float factor for supersampled operation in the x axis.
# ss_y - float factor for supersampled operation in the y axis.
# Smode - Sharpen mode, can be 1, 2, or the default value 3.
# 1 = UnsharpMask() [from WarpSharp.dll package]
# 2 = Sharpen()
# 3 = "MinMaxSharpen()" [private routine of LimitedSharpen]
# strength - obviously, the strength of sharpening. Typically (0-100).
LimitedSharpen(ss_x=1.5,ss_y=1.5,Smode=3,strength=40)

[Randall]

A noise filter. Undot() is a simple median filter for removing dots, that is stray orphan pixels and mosquito noise. It basicly just clips each pixel value to stay within min and max of its eight surrounding neighbors.

Noise Filter
# NOISE FILTER requires UnDot.dll plugin
# UnDot() is a simple median filter for removing mosquito noise.
UnDot()

[Chainmax]

Here's my script for animated content (only telecined or interlaced):

DeDot() (http://nullinfo.s21.xrea.com/#DeDot_YV12)
BiFrost() (http://yatta.mellbin.org/misc/)
TFM(d2v="X:\wherever\yourd2v.d2v",mode=6,PP=7,slow=2,mChroma=true,chroma=true)
TDecimate(mode=1) <== TFM and TDecimate require TIVTC (http://forum.doom9.org/showthread.php?t=82264&page=22&pp=20)
TDeint(type=1) or (type=2) (http://forum.doom9.org/showthread.php?t=82264&page=22&pp=20)
BiFrost(interlaced=false) <== (read its readme to know wether to apply it before or after Deinterlacing/IVTCing)
Crop(your_settings,align=true)
SmartSSIQ() (http://forum.doom9.org/showpost.php?p=744264&postcount=14) (also needs SSIQ (http://yatta.mellbin.org/misc/))
Cnr2("xxx",4,5,255) (use v2.61, tritical's website seems to be down)
LRemoveDust_YV12(17,2) (http://forum.doom9.org/showpost.php?p=662180&postcount=218) or FFT3DFilter(sigma=3 or 1,bw=32,bh=32,bt=3,ow=16,oh=16) (http://bag.hotmail.ru/)
BicubicResize(H_Res,V_Res,0,0.5) or Lanczos4Resize(H_Res,V_Res)
DeHalo_Alpha() (http://forum.doom9.org/showpost.php?p=738264&postcount=43)
aWarpSharp(depth=16,cm=1) (http://ziquash.chez-alice.fr/)
LimitedSharpen() (http://forum.doom9.org/showpost.php?p=559994&postcount=3)

I tried to make this as general as possible, and there are a few points I don't cover, like chroma smoothing and deblocking, but I hope this is useful.

[DryFire]

Why are you using Tdeint after TFM and Tdecimate?

Also have either of you used teh faster version of limited sharpen?
http://forum.doom9.org/showpost.php?p=745493&postcount=318

I haven't gotten a chance to, but I'm waiting on a dvd.

[Anonymouses]

A sharpening filter. Sharpening typically requires a high bitrate to work with, meaning that the original source should be a DVD vob or something with plenty of bitrate to work with.

This doesn't make sense. A sharpening filter in no way is dependant on the bitrate of the source. As long as there is some detail to sharpen it will work regardless of the bitrate. Now if you were trying to say that a sharpened source will need more bitrate to encode then yes, you're right.

[Randall]

This doesn't make sense. A sharpening filter in no way is dependant on the bitrate of the source. As long as there is some detail to sharpen it will work regardless of the bitrate. Now if you were trying to say that a sharpened source will need more bitrate to encode then yes, you're right.
What you said makes no sense. Let me break it down for you, Detail = Bitrate. so more detail to sharpen will require more bitrate.

[Chainmax]

Why are you using Tdeint after TFM and Tdecimate?

Also have either of you used teh faster version of limited sharpen?

It's meant to be a general script, trying to cover as many sides as possible. Not all sources will require dotcrawl removal (DeDot) or DeRainbowing (SmartSSIQ) or chroma smoothing (CNR2). The user has to decide which filter to use.
About LimitedSharpen, I'm waiting for an official release before recommending it.

[berrinam]

This doesn't make sense. A sharpening filter in no way is dependant on the bitrate of the source. As long as there is some detail to sharpen it will work regardless of the bitrate. Now if you were trying to say that a sharpened source will need more bitrate to encode then yes, you're right.
Ummm.... doesn't sharpening a low-quality source result in the artifacts becoming sharpened and being more visible -> making it look worse instead of better?