| User review:
|
I'm a huge fan of HuffyUV and have been using it for years for capturing and editing. I did some tests in Sony Vegas, encoding the test image provided here:
http://www.adamwilt.com/pix-gens.html#DVtest1
I encoded a five second clip. Loaded the resulting file into Vegas and added a caption 'copy01' - so Vegas was forced to re-encode rather than it doing a 'no recompression' save. I then encoded that, loaded it back in, changed the caption to 'copy02' so I could keep track and so on...
Using the convert to YUY option I found there was slight quality loss on each additional encode with the chroma moving to the left slightly (one pixel?).
In the other modes - predict left, predict median and so on - there was no loss whatsoever. I went up to eight re-encodes using predict left/predict left and stopped, because it was obvious there was no loss. I don't know how anyone could see increased grain or a change of colour in this mode.
Convert to YUY gives a file that is about half the size but it seems you take a small quality hit right from the first capture/encode, compared to the other compression options.
I'm not sure why other people reckon that Convert to YUY is 'lossless'. Maybe they are just cutting the video and resaving which, as I already mentioned, results in a no recompression frame-for-frame save in some software? As soon as you add filters or effects you are reencoding. |
