[MPlayer-dev-eng] [libass] Ignore PlayResX/Y aspect ratio for font aspect ratio.

[Grigori Goronzy]

Nicolas George wrote:
> Ok, I see the problem. I had never met scripts that overcompensated for the
> scaling of anamorphic content. That is a shame: such scripts will look
> overly thin on settings that scale before rendering the subtitles, such as
> vo=gl.
>

Yes, but libass can compensate for that (the standalone version w/ uau's
MPlayer fork does it by default).
I like to call it "PAR decorrection". :)

> And, as I said, I have seen quite a few scripts that make the opposite
> assumption (so that they look the same on various encodings of the same
> content).
>

Actually I've never seen that.

> I believe this calls for an option. I think I can provide a patch adding:
> 
> -ass-font-scale-x k
> 
> With the following semantic:
> 
> - k > 0: horizontally scale fonts by k; k = 1 means the current (braindead
>   but compatible) behaviour; k = 45/64 compensates for a 64/45 scaling (16/9
>   PAL).
> 
> - k = -2: sets k according to PlayResX/PlayResY and orig_width/orig_height;
>   this gives us the behaviour we had six months ago and that I liked.
>

This seems to be reasonable.

> - k = -1 (default): heuristic guess:
> 
>   - if any style in the script has ScaleX != 1, then behave as if k = 1
>     (assuming that ScaleX is there to compensate for anamorphic content);
> 
>   - otherwise, behave as if k = -2
> 

This doesn't.  Often, ScaleX is used for complex typesetting (i.e.
squishing/stretching on purpose) and if there's a style with ScaleX !=
100 it doesn't really mean anything. So you would possibly misdetect
scripts that need k = -2 as scripts that need k = 1.
(ScaleX/Y is represented in percent, by the way)

Maybe it's a better idea to determine the scaling mode from the content
or container.  If you've got Matroska with embedded ASS/SSA subtitles,
you can be almost 100% sure the "k = 1" behavior is expected. If you
have text (SRT/DVB etc.) subtitles that are converted by MPlayer and
rendered with libass, the "k = -2" behavior is needed, for example.

Grigori