Well I'll do part 2 anyway since it could help other people
After you have the video in avisynth, you can start filtering it.
The amvapp comes with some assortment of filters, but there might come a time where you need to download a filter. First stop for filters, the
avisynth mediawiki. That link brings you to a page with a pretty good list of filters categorized by what they do (denoising, deinterlacing, sharpening, etc...) really useful when you know what you want to do but not what filters do it. After that, google is your friend. You can usually get somewhere by putting "avisynth" after whatever you want (ex. dot crawl avisynth brings up hits on filters that remove dot crawl, TFM() avisynth gives you a hit to where you can download the TFM() filter)
most times you'll find .dll or .avsi files. After downloading, you can put them into avisynth's plugins directory (C:\program files\avisynth 2.5\plugins\) to autoload them. Doing so makes it easier for you later on, because if you don't put them in the plugins directory you have to manually load each one that you use by putting either LoadPlugin() or Import() into your script for .dll and .avsi files respectively.
Filtering is the toughest part of using avisynth because it requires:
1) That you know what footage issue you're trying to correct
2) That you know what filter(s) deal with the issue(s) you have
3) Patience to dial in the settings for each filter
In addition, there are 2 basic laws of filtering:
1)You should only use a filter if it can solve an issue that you see.
2)Your output quality is limited by your input quality
Knowing what footage issues you have takes time to learn, and probably the best way to learn is to post screenshots so other members can tell you what needs fixing. After a while, you'll be able to spot the issues on your own. This will also give you an idea of what filters do what. The biggest pitfall of people starting to work with avisynth is that they'll take a script for one series and just use it for everything. Generally speaking though (very generally speaking), each series needs a different avisynth script.
Each filter has different settings that control how the filter does what it does. Most filters are well documented, explaining what each parameter controls and how changing the value will affect your video. It takes a bit of time and a bit of determination to work through the documentation, and then some trial and error testing to get the "best" settings.
A relatively simple filter is fluxsmoothst(). It is a type of smoothing filter and has 2 parameters, spatial_threshold and temporal_threshold, the parameters are both int type (whole numbers like 1, 2 , 3, etc...) and increasing the parameters will increase the amount of smoothing that takes place. It would look something like
Code: Select all
fluxsmoothst(spatial_threshold = 8, temporal_threshold = 9)
in your script, and you keep changing the values of the parameters until you are satisfied.
The order that you use the filters in does make a difference, so
and
will give you different looking results.
There are a few rules for where to place different filters:
1) It is a very good idea to Deinterlace/IVTC before resizing
*related to resizing, keep your width and height as even numbers, some filters or codecs may require that your width and height be mod 16 (ie, width/16 = a whole number)
2) Sharpening filters should generally go after your smoothing filters
3) Some filters need (or prefer) interlaced input, so those would go before you deinterlace/IVTC
...And that's the end of my primer. I think everything else is best learned by doing, the only other thing is that Avisynth only works in windows. So if you have yourself a windows machine and still want to give it a shot you can post some screenshots so we can get started
