5 Jun
2013
5 Jun
'13
4:16 p.m.
Am 17.05.2013 um 22:34 schrieb Carsten Kurz:
Interesting. I know there are some shortcuts (in the part of the JPEG2000
encoder which I have looked at) which take advantage of some types of
"simplicity" in the source.
I thought j2k encoding speed was more or less
independent from content complexity. At least I wouldn't expect a 1:4 speed difference
between 'simple' and 'normal' content. But will do more testing while
doing this project.
I just did a test comparison between my frame count test video (white
numbers on black) and a (static) synthetical colour noise pattern (noise
at the single pixel level). Both MP4, 1998*1080 24fps with stereo audio,
all transcoding parameters identical, no additional processing/filters
enabled.
While the very simple frame count images compress at up to 3.6 fps in DVD-o-matic, with
the noise pattern it drops to 0.7fps on the same machine. That is a 1:5 difference between
best and worst case content! Realworld footage from my Sony 1080p cam compresses at
around 1.2fps.
Set J2k bandwith/datarate is irrelevant it seems.
That is again my i3-2310M notebook running at 2.1GHz with WIN7/64 (2core/HT, 4 threads
configured), 100% CPU load.
So, if we want to compare CPUs/machines, we definitely need a common test clip.
That would be good. Perhaps something like the Sintel trailer?
Carl - how is it going with the MacMini - do you have
a Bootcamp windows
installation on it? What is the exact config of that machine?
DVD-o-matic is working on the Mac, I just need to finish the .dmg-based
installer. I am just running OS X at the moment. It's the 2.5Ghz i5
version.
I'll do some benchmarks as soon as I can.
Best regards
Carl