For few years I’ve been reading (and using) the term Uprez whenever the people try to make a picture bigger, what ever the original size and whatever the target size. However if you show me a serious document stating what you say i will happily accept it.
I have no serious documents to offer, just logic.
Insignificant to make 1920×1080 pixels from 1280×720 pixels?
Insignificant to make 1280×720 Square pixels from 960×720 HD pixels?
Apparently your notions of size is constrained to the individual frame, which would make sense if we were talking still photography. But we are not, we are talking about motion pictures, and here temporal considerations also apply.
So yes, the differences are insignificant compared to going from 480X720 interlaced to 1280×720 progressive. The reason 720 is comparable to 1080 is the progressive frame rate. If you actually count pixels per second, 1080p24 has 1920x1080x24 (49,766,400 pps) 720P60 has 1280*720*60 (55,296,000 pps) which comes to around a 10% difference, with what you consider the smaller size having the larger amount of information. If you were using 1080i30 it would be approximately 62,000,000 pps still about a 10% difference, albeit the other way. However standard def has 480*720*30 (10,368,000 pps) which is aprox. an 80% difference from 720P. So I stand by my statement that the difference between the various HD formats is “insignificant COMPARED to the difference between SD and HD.”
The logic in using one term “up rez” to indicate the always huge and one way change in data and the term “cross convert” to indicate a much smaller and often more complicated change in data size makes a lot of practical sense.
Herb Sevush
Zebra Productions