Jeremy,

Is the choice of 0xff best? Could there be some confusion with the sync?

Perhaps a better choice would be 0x55 ?


jeremy brand wrote:

> Hello,
>
> Background:
>
> The idea behind this patch is to keep LAME from putting LAME name and
> versions into each frame of the mp3.  I know that it is nice to have LAME
> information in frames for debugging and testing, but sometimes there is a
> want to have _clean_ (so to speak) mp3s.  I have consulted with other
> collegues that use LAME and one of the repetitive comments about LAME is
> 'do you like that they put their name and version in each frame of the
> mp3'?  Truthfully, I don't.  Besides, soon, I will be building a complete
> (as near as) studio archive of mp3s out of about 250 CD and I do not want
> software name and versions in my mp3s.  This patch is handy for me, maybe
> it will be handy for others.
>
> Implementation:
>
> The default functionality of LAME _has_ _not_ _changed_.  Functionality is
> only changed when the new flag '--no-grafiti-frames' is added.  When this
> flag is added, the LAME name and version _are_ _not_ added into the
> ancillary data in each frame.  Insead, only 0xff's are used.  There is
> absolutely no difference in size of file, or sound or any other functional
> part of the mp3.
>
> Now, I dug this up in the ChangeLog.
>
> ----
>   2000-11-01 17:56  markt
>     * libmp3lame/: bitstream.c, bitstream.h, util.h:
>
>       restored bitstream.c to original.
>       drain_into_ancillary_data was written the way it is
>       on purpose.  dont change it without checking with me first
> ----
>
> I am modifying that function, but I _have_ _not_ changed the functionality
> of drain_into_ancillary_data, but only changed from adding 'LAME version'
> into the frame to only adding 0xff's.  This simply makes it nicer for
> people who want to have an un-tainted (so to speak) archive.  It just
> makes it more flexible for those people who want _clean_ frames.  Nothing
> more, nothing less.
>
> So, now, I'm officially asking to change drain_into_ancillary_data.  Feel
> free to apply the patch and note that I have (hopefully) not missed
> anything.
>
> Thank you,
> Jeremy
>
> Credit for the patch goes to Josh Samuelson <[EMAIL PROTECTED]> who
> found and modified drain_into_ancillary_data initially to only fill with
> 0xff's.  Credit for adding the option, command line flag, struct change,
> and the rest go to Jeremy Brand <[EMAIL PROTECTED]>.  I'm only mentioning
> this in hopes that the patch gets included in CVS and that credit goes
> where credit is due.
>
> --
>      /   Jeremy Brand, B.S.      \     Sr. Software Engineer          /
>     /   phone://39-34-853-23988   \   mailto:[EMAIL PROTECTED]      /
>    /   http://www.JeremyBrand.com/Jeremy/Brand/Jeremy_Brand.html    /
>   /  "LINUX is obsolete"  -- Andy Tanenbaum, January 29th, 1992    /
>  /  Get your own Free, Private email at http://www.smackdown.com/ /
>
>   ------------------------------------------------------------------------
>                              Name: no_grafiti_frames.diff
>    no_grafiti_frames.diff    Type: Plain Text (TEXT/PLAIN)
>                          Encoding: BASE64

--
MP3 ENCODER mailing list archive is at:
http://www.mail-archive.com/mp3encoder%40minnie.cs.adfa.edu.au/

Reply via email to