|
about the raw size i guess i can see the sense in that, more size=more info to be encoded.
but the problem with that is that it can simply mean a huge amount of action, so if a "calm" series is about 400 and an action based one is 500 its not necessarily a case of higher "detail"..
but all in all i guess you can at least get an estimate using that method..
about your second point, in general i will agree, the most important thing is not losing data..
but using the lines as a measure is only ok as long as the broadcasters aren't using some kind of filters on the video when upscaling it could cause the lines to get thinner (thou i don't really think they do those kind of things, but you can never really know uh? XD)
or they could rescan the originals which were meant for sd and make a higher resolution scan which i think will also cause the lines to be thinner (i think digital upscaling causes lines to grow in a way that is easy to tell, and of course is easier to downscale back without having to deal with disappearing lines, while with scanning the lines will be thinner and thus may have trouble downscaling)
|