Do you have some screen grabs to compare?
Keep in mind that taking square pixel 1920×1080 down to non-square 720×480 isn’t a process which will handle light weight or small point-size fonts very well. You are taking each frame, represented by over 2 million pixels and forcing it to map the same image into 345,000 pixels…17% of the original pixel count.
Years ago there used to be a lot of discussion about various encoders and what one does better than another, or whatever ‘secret sauce’ conjecture was floating around (I remain a HUGE fan of GV ProCoder for transcoding)…over the last 5-6 years, Adobe’s encoder has historically worked well for me.
However when I go to DVD, I’m typically going from 1920x1080p23.976 square pixel to 720x480i29.97 non-square (1.2) pixel…but if I know I’ll end up on a standard def distribution copy, I don’t use exceptionally fine onscreen text…or at least I don’t expect that it will reproduce well at less than a quarter of the original resolution and a change in pixel shape on top of it (we haven’t even addressed whether you might be going from progressive to interlace.)
So, I haven’t seen your results of course, but I doubt anyone is truly ‘satisfied’ with onscreen text that goes through that sort of resolution reduction. I don’t do a lot of SD delivery anymore, so I haven’t noted any specific changes in Adobe’s encoder in the last handful of releases myself…
TimK,
Director, Consultant
Kolb Productions,
Adobe Certified Instructor