[MPlayer-users] Re: -ss switch broken in current 3-pass-mode of mencoder

[D Richard Felker III]

On Fri, Sep 13, 2002 at 10:48:04AM +0200, Dominik Mierzejewski wrote:
> [Automatic answer: RTFM (read DOCS, FAQ), also read DOCS/bugreports.html]
> On Friday, 13 September 2002, Arpi wrote:
> > [Automatic answer: RTFM (read DOCS, FAQ), also read DOCS/bugreports.html]
> > Hi,
> > 
> > > In pass one "-ss 100 -endpos 102" encoding starts at position 100 and
> > > ends at 202, giving a total of 102 seconds to encode.
> > > 
> > > In pass two "-ss 100 -endpos 102" encoding starts at position 100 and
> > > ends at 102, giving a total of 2 seconds to encode.
> > > 
> > > With "-ss 100 -endpos 98" pass one will encode 98 seconds and pass two
> > > will just fail, because it thinks a total of -2 seconds is reqested.
> > > 
> > > So the real problem is that the -endpos parameter is interpreted
> > > differently in different places (and it is not even clear in the DOCS
> > > whether -endpos means "length" or "end position").
> > 
> > as its name says, it's end position.
> > but according your above tests, it seems to be buggy too :(
> 
> Hmmm. I was under the impression that it was, in fact, the duration, only
> misnamed. And I think it's described as such in manpage. Either way, it
> seems it doesn't behave consistently, so it should be fixed and I vote for
> having two options, mutually exclusive: -endpos and -duration.

Sorry, it was originally misnamed but consistent (always a duration)
when I added it a long time ago. Changes since then must have broken
it. I agree both forms should be available, and most importantly, they
should have names that make sense.

Rich