[LAU] Decent and attractive audio player

[Steve Fosdick]

On 27/04/08 23:49:22, hollunder at gmx.at wrote:
> 
> On Sun, 27 Apr 2008 23:35:00 +0100
> Steve Fosdick <lists at pelvoux.nildram.co.uk> wrote:
> 
> > On 27/04/08 20:51:11, Dave Phillips wrote:
> > > IIRC gapless playback can be activated from the Preferences 
> dialog
> 
> > > but it is not on by default. Sorry, I didn't test it. :(
> > 
> > Dave was talking about rhythmbox.  I have checked this out and
> > rhythbox does not support gapless playback instead it supports
> > 'crossfading' which is not the same thing at all.
> > 
> > See http://wiki.hydrogenaudio.org/index.php?title=Gapless
> > 
> > Regards,
> > Steve.
> 
> The gapless-problem is the same for all gstreamer-based audioplayers,
> as far as I remember I read that it's by design.

I stand corrected on this point.  I checked out the rhythmbox-dev 
mailing list and it turns out that although the new rhythmbox output 
engine is called the crossfading backend and can do crossfading it can 
also do genuine gapless playback.

To make it work in gapless mode you have to enable the crossfade 
backend but set the crossfade duration to zero.

Technically it works by using a gstreamer adder in the output and, as 
one track is about to finish it builds a new gstreamer decode pipeline 
for the new song and pre-rolls it so it is ready to supply the PCM data 
into the adder the moment the previous track ends.  I have tested it 
with the samples refered to in the hydrogen audio page in my original 
post and it played them flawlessly. 

> Another reason for me not to use rhythmbox is that it relies on mono,
> something I don't know what to think of.

AFAIK mono is an implementation of .NET which in turn relies on a 
virtual machine language (like Java does).  The combination of the 
inefficiency of that approach with a single company (Microsoft) being 
in control of the specification makes it seem like a poor choice of 
technology.

Fortunately though rythmbox does not appear to be implemented in mono.

Regards,
Steve.