MJPEG Tools Svn

assumptions about chroma subsampling should not be coded in as explicit
"divide by 4" (which assumes 4:2:0). Use the Y4M API to dynamically compute
the chroma plane size. I.E. do not do 'w * h / 4' but instead use
y4m_chroma_ss_x_ratio (and 'y').

Fix RPM build

accidentally left out an arg in the usage summary for the chroma option. -x
takes an argument

Overlooked the 'k' (flip bytes) option in the getopt() call.

A nev version for the comming release

Oops - single quotes instead of double quotes in a few places (PPC).

Fix gcc4 support for PPC arch

Changes from Burkhard Plaum (libquicktime) for gcc4 support.

added a motion-compensation to the denoiser, again (as the temporal
filter seems to be very robust against motion-errors, now)
...but a lot(!) left to do...

mpeg2enc rather consistently (even on single cpu systems) crashes on the
linux systems (OS/X doesn't appear to have the problem):
INFO: [mpeg2enc] Signaling last frame = 499
mpeg2enc: seqencoder.cc:433: void SeqEncoder::EncodeStream(): Assertion `pass1coded.size() == 0' failed.
Abort

So, make the default -M value 0 to disable threading. Actually for buffering
the 'bfr' program can be placed in front of mpeg2enc and obviate the need for
threading in the encoder completely (it doesn't give much of a speed increase
as it turns out).