[MEncoder-users] Different parameters in two-pass encoding

[Stjepan Brbot]

On Sun, 2009-02-01 at 23:26 +0100, Guillaume POIRIER wrote:
> Hi,

Thanks Guillaume for your answer.

> On Sun, Feb 1, 2009 at 11:12 PM, Stjepan Brbot
> <stjepan.brbot at zg.t-com.hr> wrote:
> > Hi,
> >
> > several times on Internet and this mailing lists archive I found that
> > people recommend setting parameter "turbo" in 1st pass encoding while
> > omitting it in 2nd pass.
> 
> That's silly. turbo option as an effect only if it's specified in the
> first pass. Otherwise isn't a non-op.

Of course, turbo parameter is used in 1st pass only.

> > Or even, to define only subset of parameters in
> > 1st step and put several additional parameters (like mdb=2:vhq=4) in 2nd
> > pass. Generally speaking, is this correct?
> 
> Well, the whole purpose of turbo is that you keep the same options for
> all passes, and let turbo remove the most expensive ones that have a
> very little or no impact on the resulting 2-pass encode.
> 
> > In my opinion 1st pass log is
> > used for better redistribution of bitrate in 2nd pass encoding but 1st
> > pass log should be created with exactly the same conditions (encoding
> > parameters). Isn't it?
> 
> That's correct, but it's all a matter of trade off IMHO. I prefer to
> speed-up the first pass, and invest the time I saved in the first pass
> to use higher quality encoding options. In the end, for given the same
> overall time, I get better quality.

I understand now this turbo parameter; with turbo one will get almost
the same log file (bitrate redistribution) like with all other
parameters but gaining the execution speed. So do I have to explicitly
omit other parameters (like mbd=2:vhq=4) when using "turbo" in 1st step
or turbo parameter will order mencoder to ignore these sophisticated
parameters while creating 1st pass log?