On Fri, 27 Feb 2009, David Hall wrote:

> R.L. Horn wrote:
>> Bizarrely, if I encode using --nogap and --nogaptags, I don't, but the
>> splice isn't as noticeable as with tagless (and gapless) streams.
>>
> In this latest experiment are you merging the two MP3 files into one?

No, though I did strip the Info tags and concatenate the streams to
confirm that they were, in fact, gapless.

Repeating with a slightly more complex audio sample (an A-major chord)
brought back the inter-track "tick", with or without Info tags. I'm now
inclined to believe that the supposed seamlessness of the aforementioned
separately-encoded-and-tagged streams (440Hz sine wave) was a fluke.

Further complicating matters is the fact that I'm having trouble making
sense of the tags generated by lame (3.98.2). In particular, they're
showing a 576-sample encoder delay, even for the --nogap streams. At
first, I thought this was attributable to the tag frame itself, but MPEG1
Layer III uses 1152-sample frames.

As this is rapidly spiraling outside of my, admittedly limited, realm of
expertise, I've made my test files, along with a Bourne shell script to
strip the tags from the --nogap streams, available at:

   http://www.eastcheap.org/gaptests/gaptests.tar.bz2

Anyone interested is more than welcome to give them a listen.
-------------------------------------------------------------------
List admin: http://cool.haxx.se/cgi-bin/mailman/listinfo/rockbox
FAQ: http://www.rockbox.org/twiki/bin/view/Main/GeneralFAQ
Etiquette: http://www.rockbox.org/mail/etiquette.html
Received on 2009-02-28