Video Resizing Methods

[5k0rp10n]

What's the difference between all these resizing methods that are found in programs such as FlasKMPEG (i.e. bilinear filtering, bicubic filtering,...) and Gordian Knot (i.e. soft bicubic, neutral bicubic, sharp bicubic...)?
Are some better used in certain situations than others (encoding a movie at a high bitrate/on 2 CDs, encoding at a low bitrate, amount of action within a movie, etc.)

Any info would be appreciated.

[Kevin AMD]

the bi**** options are used to make the resize with bigger or smaller movie screen but i don't know wich one is for smaller en wich one is for bigger screen...

[Enchanter]

Re: Video Resizing Methods

Originally posted by 5k0rp10n
What's the difference between all these resizing methods that are found in programs such as FlasKMPEG (i.e. bilinear filtering, bicubic filtering,...) and Gordian Knot (i.e. soft bicubic, neutral bicubic, sharp bicubic...)?
Are some better used in certain situations than others (encoding a movie at a high bitrate/on 2 CDs, encoding at a low bitrate, amount of action within a movie, etc.)

Any info would be appreciated.

Bicubic is more accurate than bilinear and hence produces sharper images. This is a good thing of course, but you will need to ensure that you are using sufficiently high bitrate as well to accomodate for the added details. Precise is as what it says, even more accurate resizing and presumably sharper image.

Bilinear is suitable in lower bitrate conditions where a softer (and hence more easily compressible) image is preferred.

p.s. Forget Flask. It's resizing algorithm is terrible. Even its bilinear mode makes the image hard to compress.