On Thu, Aug 17, 2000 at 11:24:14PM +0200, Robert Hegemann wrote:
> Bill Currie schrieb am Don, 17 Aug 2000:
[Xing header garble snipped]
> Hm, I think this is not right in LAME. Maybe there are reasons for closing
> and reopening the stream, but I don't see any.
The only reason I can see so far (I've been browsing the code looking for
open/close calls) is either simplicity, ad hoc modifications, or a combination
of the two. Of coures, I could be missing something (like possibly the
inability to seek under certain circumstances).
> Further I fear that in case
> of ID3v2 tags the filesize is not correctly calculated.
Ouch :/ Well, I'll leave this to those with more knowledge:)
> This needs some attention!
Thanks. Though I've found a work around ("don't do that":), this has been
giving me problems for 2 weeks and has obviously been there for a while. The
only reason I figured out what was going on was I decided to run 16 lame
processes at once (as a lark), remove the .wav files and rename the .mp3 files
to their final names. I wound up with 16 25+ minute songs (according to xmms).
Uh-huh, yeah, right, 400 minutes of music on one cd? I don't think so:).
Now all I need is to find (or write) a program to create a correct Xing
header. I really don't feel like re-ripping all those cds (about 30).
Bill
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
