mirror of https://github.com/mpv-player/mpv
Few fixes and suggestions by Jeff and Diego
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@15925 b3059339-0415-0410-9bf9-f77b7e298cf2
This commit is contained in:
parent
2aa6a03251
commit
3249e16758
|
@ -2726,25 +2726,27 @@ codec</title>
|
||||||
The quality of the encode yielded by this option highly depends
|
The quality of the encode yielded by this option highly depends
|
||||||
on personal preferences and on the type and monitor settings
|
on personal preferences and on the type and monitor settings
|
||||||
used to watch it (typically, it will not look as good if it is
|
used to watch it (typically, it will not look as good if it is
|
||||||
bright or if it is a TFT monitor.
|
bright or if it is a TFT monitor).
|
||||||
</para></listitem>
|
</para></listitem>
|
||||||
|
|
||||||
<listitem><para>
|
<listitem><para>
|
||||||
<emphasis role="bold">qpel</emphasis>
|
<emphasis role="bold">qpel</emphasis>
|
||||||
Increases the precision of the motion estimation from halfpel to
|
Raise the number of candidate motion vectors's by increasing
|
||||||
|
the precision of the motion estimation from halfpel to
|
||||||
quarterpel.
|
quarterpel.
|
||||||
The idea is to find better motion vectors which will in return
|
The idea is to find better motion vectors which will in return
|
||||||
reduce bitrate (hence increasing quality).
|
reduce bitrate (hence increasing quality).
|
||||||
However, quarterpel precision motion vectors are coded in the
|
However, motion vectors with quarterpel precision require a
|
||||||
bitstream with a few bits, so if the content do not feature
|
few extra bits to code, but the candidate vectors do not always
|
||||||
enough possibilities to find better motion vectors through
|
give (much) better results.
|
||||||
<option>qpel</option>, it will in fact hurt compressibility.
|
Quite often, the codec still spends bits on the extra precision,
|
||||||
|
but little or no extra quality is gained in return.
|
||||||
Unfortunately, there is no way to foresee the possible gains of
|
Unfortunately, there is no way to foresee the possible gains of
|
||||||
<option>qpel</option>, so you need to actually encode with and
|
<option>qpel</option>, so you need to actually encode with and
|
||||||
without it to know for sure.
|
without it to know for sure.
|
||||||
</para><para>
|
</para><para>
|
||||||
<option>qpel</option> can be almost double encoding time, and
|
<option>qpel</option> can be almost double encoding time, and
|
||||||
requires as much as 30-60% more processing power to decode.
|
requires as much as 25% more processing power to decode.
|
||||||
It is not supported by all standalone players.
|
It is not supported by all standalone players.
|
||||||
</para></listitem>
|
</para></listitem>
|
||||||
|
|
||||||
|
@ -2752,12 +2754,12 @@ codec</title>
|
||||||
<emphasis role="bold">gmc</emphasis>
|
<emphasis role="bold">gmc</emphasis>
|
||||||
Tries to save bits on panning scenes by using a single motion
|
Tries to save bits on panning scenes by using a single motion
|
||||||
vector for the whole frame.
|
vector for the whole frame.
|
||||||
This almost always raise PSNR, but significant slows down
|
This almost always raise PSNR, but significantly slows down
|
||||||
encoding (and also needs more processing power to decode).
|
encoding (significantly slows down encoding).
|
||||||
Therefore, you should only use it when you have turned
|
Therefore, you should only use it when you have turned
|
||||||
<option>vhq</option> to the maximum.
|
<option>vhq</option> to the maximum.
|
||||||
<systemitem class="library">XviD</systemitem>'s GMC (which are
|
<systemitem class="library">XviD</systemitem>'s GMC is more
|
||||||
more sophisticated than DivX's are not supported by all
|
sophisticated than DivX's, but is only supported by few
|
||||||
standalone players.
|
standalone players.
|
||||||
</para></listitem>
|
</para></listitem>
|
||||||
|
|
||||||
|
|
Loading…
Reference in New Issue