[MPlayer-dev-eng] [PATCH] panscanrange option

[The Wanderer]

Reimar Döffinger wrote:

> Hi,
> On Sun, Sep 25, 2005 at 08:48:57PM -0400, The Wanderer wrote:
> 
>> Reimar Döffinger wrote:
>> 
>>> .TP
>>> +.B \-panscanrange <-19.0\-99.0> (experimental)
>>> +Change the range of the pan-and-scan functionality.
>>> +Positive values means you can zoom up to this many time the factor needed
>>> +to remove the borders.
>> 
>> This sentence a) is ungrammatical (at bare minimum, it needs
>> "times" instead of "times" and either "mean" instead of "means" or
>> "A positive value" instead of "Positive values"), and b) is not
>> necessarily clear. Does it in fact mean "there is a factor which is
>> necessary to remove the borders, and this specifies the multiple of
>> that factor to which it is possible to zoom"? If not, then I don't
>> know what is actually meant. (If
> 
> Yes, that is right.

And I presume, from the fact that you didn't address it, that my
assumption that specifying -panscanrange but not -panscan does nothing
is correct.

>> Do I understand correctly that negative values are just a different
>> way of specifying some of the positive values - i.e., the results
>> of '-panscanrange -3' would be identical to the results of
>> '-panscanrange 4'? If not, then I don't understand what is being
>> said here.
> 
> Yes, with one execption: when playing e.g. a 4:3 video on a 4:3
> screen you always have a border size of 0. No matter which positive
> value you specify the 'w' and 'e' keys will do nothing at all (since
> x * 0 is still 0).
> 
> And if -panscanrange -3 corresponds to -panscanrange 4 depends on
> movie and monitor aspect.

...first you say that I did understand correctly, then you say that what
I said is not necessarily true. I'm confused.

You say that the border size (which I infer to mean, the width of the
region cropped off at the sides of the image) will always be 0 if the
video aspect matches the monitor aspect, which makes sense. I'm not
entirely sure how the question came up from what I wrote, though.

Does the fact that you say specifically "No matter which positive value
you specify" mean that the description which follows that applies only
when using positive values, not when using negative values? I don't
quite see how they could behave differently, based on the very limited
understanding I have of the functionality we're actually talking about,
but you obviously know the code much better than I do.

> Best try it out, I actually find only negative values really useful,
> but others might have different opinions.

I've never actually had occasion to use -panscan, so I wouldn't even
know where to begin experimenting. (I don't mean "I don't know the
syntax" - although I don't, that's easily found; I mean that I'd have to
do some experimentation even to figure out what I'd be trying to find out.)


I'd try to go over this in more detail, since it seems to need it, but
my brain is feeling foggy enough that I don't trust myself to get it
right. I'll come back to it after getting some sleep, if possible.

-- 
       The Wanderer

Warning: Simply because I argue an issue does not mean I agree with any
side of it.

Secrecy is the beginning of tyranny.