> X-Authentication-Warning: geek.rcc.se: majordom set sender to
>[EMAIL PROTECTED] using -f
> From: "Joshua Bahnsen" <[EMAIL PROTECTED]>
> Date: Thu, 30 Mar 2000 16:00:17 -0500
> Content-Type: multipart/alternative;
> boundary="----=_NextPart_000_0007_01BF9A61.0B308E60"
> X-Priority: 3
> X-MSMail-Priority: Normal
> X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600
> Sender: [EMAIL PROTECTED]
> Precedence: bulk
> Reply-To: [EMAIL PROTECTED]
> X-UIDL: -(b!!cXD"!nOS!!2;X"!
>
> I just want to make sure I understand what using a filter actually
> does. This is what I think it does, tell me if I'm right. Say you
> want to encode a file at 128 Kbps, if you had a filter that cut off
> at 16 Khz, then that would leave more bytes open for the range
> between the low cutoff and 16 Khz, right? If you didn't have the
> filter, then there is less available for the midrange because you
> are also encoding higher rates that the average person can't hear,
> am I getting this at all? I've read many places that 16 Khz is the
> normal cutoff of human hearing, so I assume that encoding at 160 and
> 192 with recent versions of lame, it just gives the opportunity for
> higher samples because there is more to work with, right?
>
> Joshua Bahnsen
that sounds pretty good to me. The most optimal solution
would be some kind of dynamic filter that adapts to the
type of music. If the high frequencies actually contain
more energy than the mid frequencies, we probably shouldn't
be filtering. (by far the most common occurrence of this
computer generated wav's designed to test a mp3 encoder's
lowpass filter :-)
Mark
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
