Northpack
3rd April 2002, 13:59
Please tell me whether my assumption is wrong or not:
I usually encode 2 CD rips at full DVD Resolution. I crop the source to a width of 704, thus resizing is only needed to squash the picture vertically. My assumption is that in this case there is no advantage (perhaps rather a disadvantage?) of using a bicubic resize filter? Sure, if the picture is resized two dimensionally, a bicubic algorithm makes sense IMHO, but if the width remains untouched, shouldn't a plain bilinar filter achieve the optimum quality?
Northpack
I usually encode 2 CD rips at full DVD Resolution. I crop the source to a width of 704, thus resizing is only needed to squash the picture vertically. My assumption is that in this case there is no advantage (perhaps rather a disadvantage?) of using a bicubic resize filter? Sure, if the picture is resized two dimensionally, a bicubic algorithm makes sense IMHO, but if the width remains untouched, shouldn't a plain bilinar filter achieve the optimum quality?
Northpack