Been doing some simple tests on QT7/H.264 in comparison to
QT7/Sorenson 3. No I haven't spent hours tweaking anything.
In one test - the QT7/H.264 comes in at 1.82 MB/118kbits/sec - while
the QT/Sorenson 3 is 8.68 MB/564 kbits/sec. The QT7/H.264 took about 4
times longer than the QT/Sorenson 3 to encode [on my old, old, old
Apple computer ;) ]
Harry, how can you do comparisons without targeting the same bitrate?
This makes a direct comparison impossible.
That said, H.264 is clearly the winner in quality at the same bitrate,
at least in my own testing.
Quality wise - the Sorenson video looks as if I am watching with my
glasses off - in comparison to the QT7/H.264 video.
Well, I guess if H.264 wins hands down even with 1/4 the bandwidth then
I guess a direct comparison isn't really necessary... though I must
admit, I haven't seen it as THAT radical. And there's the rendering
issue with Apple's H.264 decoder I've been hoping to get some feedback
on.
I used FinalCutPro to do the encoding - for some reason iMovie keeps
crashing [being a good boy I did send a "telegram" to Apple;)]
Can FCP do two-pass mode for H.264? I know it can't for S3...
Question: I am shooting time lapse (every 30 sec for 2 sec) - in
progressive scan. Does progressive scan make any sense in this time
lapse (interval) case - if the video is for H.264/internet playback?
Absolutely - if I was doing time lapse (as with the sample I showed
this week) I would only use a progressive capture mode... though I do
my time lapse with digital SLR's, so progressive is a given!
