[MEncoder-users] change bpp

The Wanderer inverseparadox at comcast.net
Mon Nov 7 20:35:31 CET 2005


Aleksandar Uzelac wrote:

>>> i don't get it, can you specify which -lavcopts is not proper?
>>> don't worry about second pass.
>> 
>> "Proper" can vary considerably depending on the specific case at
>> hand, but as he points out, using two-pass encoding can potentially
>> produce a vast improvement in quality even with no other change.
> 
> issues above might be a cultural thing, but the thing is that i
> considered all three posts from Rich to be very rude. i tried to
> answer as polite as i could, given the circumstances.

It is always possible to be entirely polite even if the other person is
being rude. Yes, Rich is known as perhaps the hardest member of the
development team to get along with (except for one former member who
shows up to be an ass from time to time), but the reason he's tolerated
despite that is that he does usually know what he's talking about.

> if one bothers to take a look at his posts, one could see that Rich
> stated (on more that one occasion, in several posts) that "bad
> encoding options" are presumably the cause of the problem, but
> refused to state which precisely (since i listed all options in the
> first post).

This could easily enough be seen as a matter of general aggravation,
since "inappropriate options" is just about the most common type of
problem asked about, and I imagine that re-explaining much the same
thing can get rather tiring after a while. There is at least one
alternative explanation which doesn't look bad, but I'm having a hard
time expressing what it would be.

It is also entirely posible, however, that he was just being - "lazy"
isn't quite the right word, since he's certainly under no obligation to
help out, but it's the only one which springs to mind.

> i don't get this second pass issue; i thought that stating vpass=1 is
> sufficient enough for everybody to understand that there is a second
> pass coming after (otherwise vpass=1 would not be stated at all).

No, the mere presence of "vpass=1" does not automatically signal to
people that you are going to do a second pass. It's not unheard of for
some people to put that in there anyway, under the mistaken impression
that it's necessary in order to tell the program to do only one pass, or
to have it in from a two-pass setup and not bother to remove it. The
standard practice, if one is doing a two-pass encode, is to do one of
three things: list both command lines (especially advisable if some
options are different between passes, which is occasionally a good
idea), say something like "vpass=[12]", or explicitly state that you're
doing two passes.

> i took some time and trouble to write a good introductory post
> (short, concise, with keywords and configuration), and i thought that
> would be appriciated. here is an argument to end this issue:
> 
> vpass=<1-3>
>      Activates internal two (or three) pass mode, only specify if you wish to use two (or three) pass encoding.

If everyone were fully competent and guaranteed to have read the man
page in detail and made proper use of it, then this would indeed almost
certainly suffice to end the issue. Since that isn't the case,
however...

>> Fifth: Coming to ask for help, but refusing to follow advice
>> without giving a Very Good Reason for it (and "I don't want to use
>> a pre-release
> 
> there is a very good reason, but it will probably be misunderstood. i
> work on a highly conservative platform where only a couple of
> installations (not that mplayer needs to be installed) are allowed
> during the year. target number is zero changes per year. so, i only
> change something if something must be changed to provide new required
> functionality (which is not the case here).

Okay, yes, that's a good enough reason - in fact, exactly the kind of
reason I was thinking of. Carry on.

(Though you might want to wrap your outgoing messages at no more than 80
columns, and separate paragraphs by a blank line... the fact that you
don't probably derives from the fact that you apparently send out HTML
mail, judging by the other copy of this message I received.)

-- 
       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.




More information about the MEncoder-users mailing list