By running ffmpeg -h encoder=apng
, I get this:
APNG encoder AVOptions:
-dpi <int> E..V..... Set image resolution (in dots per inch) (from 0 to 65536) (default 0)
-dpm <int> E..V..... Set image resolution (in dots per meter) (from 0 to 65536) (default 0)
-pred <int> E..V..... Prediction method (from 0 to 5) (default none)
none E..V.....
sub E..V.....
up E..V.....
avg E..V.....
paeth E..V.....
mixed E..V.....
What are the differences among these prediction methods specified with -pred
?
I could not find any documentation on ffmpeg.org or anywhere else.