Thomas Daede wrote:
Another way to specify the ranges is by metric score, rather than rate, but determining the ranges is a bit problematic. How I would probably set these ranges would be to take these rates, and use them to set the bounds based on a well regarded encoder (x264, for example).
This is also highly content-dependent (i.e., it can vary by an order of magnitude). So if you want to go this way, you will likely have a different range for each set of test sequences, which may not be a bad thing.
_______________________________________________ video-codec mailing list [email protected] https://www.ietf.org/mailman/listinfo/video-codec
