[FFmpeg-user] Is "-q:v" the same as "-qscale:v"?

[Werner Robitza]

On Fri, Oct 25, 2013 at 8:30 PM, A Commuter <acommuter at hotmail.com> wrote:
> Is "-q:v" the same as "-qscale:v"?

Yes.

> I ask because based on the page below, "-qscale:v 10" provides the highest quality video, yet when I use "-q:v 10" in my command line, the resulting file size is smaller than if I use "-q:v 5".

Are you using an older/outdated/buggy version of ffmpeg? I cannot
reproduce this. -qscale and -q both return the same file sizes for me.

Actually, the libtheora encoder source makes it clear that values
outside of 0–10 are clipped and then scaled to the range of 0–63,
which is what the libtheora encoder uses:
http://www.theora.org/doc/libtheora-1.0/structth__info.html#a4cdcf96cb46b256821993e9a830ee02

    if (avc_context->flags & CODEC_FLAG_QSCALE) {
        /* Clip global_quality in QP units to the [0 - 10] range
           to be consistent with the libvorbis implementation.
           Theora accepts a quality parameter which is an int value in
           the [0 - 63] range.
        */
        t_info.quality        = av_clipf(avc_context->global_quality /
(float)FF_QP2LAMBDA, 0, 10) * 6.3;
        t_info.target_bitrate = 0;