All,

I'm looking for a way of re-encoding my all of my MP3's (folders and
directory structure) from 320kbs to 192kbs in one pass. I currently use
Musicmatch as my encoder but it will only allow me to re-encode individual
files in a specific folder and not all of the files in all of the folders at
one time. I would like to keep them as MP3 and not convert them to another
format so Windows Media Player is out - Is their a program that will do
it???



Thanks in advance.

CES

In a previous post, CES said...

Get dbPowerAmp and the <File Selector Add On> option. They are both free
and it should do exactly what you want.

http://www.dbpoweramp.com/dmc.htm

I hope you have just changed your mind about the bitrate you want these
files to be and don't think that somehow the process of making them at
320 and then converting them to 192 is going to improve them over just
making them 192 kbps in the first place. If you have the CDs making 192
files direct from the source and deleting the 320's would be the way to
go.

HTH

--
CQ
More MP3 utilities and links >
http://www.cappycue.com/MP3Links/index.html

Thanks for your suggestion...the answer to you question is no I don't think
that it will improve the quality. The problem I have is that my music
library is 50+gb and when ever I need to copy files to my IPod or my laptop
I need to downsize the file to save space on the destination drive. I
realise the quality is going to be less then if I encoded directly from the
cd's but It took me 9 months to encode and verify the library at 320kbs -so
thats not an option.

Thanks again,
- CES


"CQ" <me@privacy.com> wrote in message
news:MPG.19bc52b31b669fc3989897@news.cis.dfn.de...

search on the net for "batch encode mp3"

-dj


"CES" <None@NotAnEmailAddress.com> wrote in message
news:99v4b.73524$Sq.13780823@twister.nyc.rr.com...

"CES" <None@NotAnEmailAddress.com> wrote in message
news:99v4b.73524$Sq.13780823@twister.nyc.rr.com...
CDex will also do what you want and batches, free and available from
http://www.cdex.n3.net/ .

As being an old command-line fan I use lame.exe for this...
It has the switch --mp3input which means it takes an MP3 file
(decodes it internally) and encodes it in the new bit rate.
This can be quite easily done for whole directories with some
batch programming.
If anybody is intersted in details - contact me:
xA.xFreiberg@xcenit.xde
(remove all x)

Regards
Armin Freiberg


"CES" <None@NotAnEmailAddress.com> wrote in message news:<99v4b.73524$Sq.13780823@twister.nyc.rr.com>. ..

A.Freiberg@cenit.de (Armin Freiberg) wrote:

I was under the impression that doing this will degrade the quality.
ie. 320kps -> 192kbs will result in worse quality then initially
encoding at 192kbs.

IIRC OGG Vorbis doesn't suffer from this problem though.

--
DS

In my opinion the loss by going down to 192 kbit from 320 is far
larger
than that from re-encoding. I may be wrong but I wonder why taking
away
nearly half of the information (320/192=1.72) should result in a worse
result than encoding directly from wav.
It would seem that since there is so much more "musical" information
im 320 kbit than in 196 kbit that it should suffice to give a good
result.
Hmmm ... anybody has done any ABX tests or similar on that?

Regards
Armin Freiberg


DaveS <davls@telus.net> wrote in message news:<uad9mv3ur.fsf@telus.net>...

In a previous post, DaveS said...

I think you are absolutely correct. I believe the difference will be
small but it will exist.

I've never heard that and really can't imagine why it would be true. It
seems to me that the above would be true for any lossy compression
format.

--
CQ

On Wed, 03 Sep 2003 08:24:46 -0400, CQ wrote:

I think it is (true), but its not *fully* implemented yet. I seem to
recall a phrase like "BitStripping" or "BitStreaming" used. I also thought
it seemed impossible, but theres a visual analogy in progressive JPEG.

I can only speak on this subject by using video as an example:
you loose about 15-20% of the quality for each generation that is created,
the reason being that when you resample your encoding a compressed image and
not a raw file. So with each generation you lose clarity, definition nad
mostimportanatly color. It's more like coping a video tape then a digital
tape...

But I've used the resampling technology in Musicmatch when I export to my
Ipod and with a few exceptions I can't tell the difference, but I'm also
partially def.

CES


"seani" <seani@attic.localdomain> wrote in message
newsan.2003.09.03.20.13.44.349746@attic.localdom ain...

On Wed, 03 Sep 2003 05:28:03 GMT, DaveS <davls@telus.net> wrote:

There are at least 2 problems with encoding. One is the MDCT performed
on the audio signal...I believe the closest analog to this is the
video example given by someone else. This is where the "copy of a
copy" type degadration occurs although for very different reasons.

The second problem is with artifacts. When you encode the original
wave file you almost always introduce artifacts. The second generation
encoding then accepts those artifacts as part of the sound to be
encoded and could create artifacts on the original artifacts,
especially when going to a lower bit rate.

You could try making several generations of a brief sample and then
compare the last generation to the first generation. I'd bet that
most of the problems would be barely, if at all, noticeable to most
people, at least for the first couple generations anyway.

On Thu, 04 Sep 2003 05:09:55 +0000, CES wrote:

As I understand it, no resampling is necessary. The original high
resolution version is comprised of several layers of information, each
succeeding layer serving to improve the accuracy of the previous less
accurate approximation.

This means that you start off with (say) the 64kbps version in one block,
and successive blocks add the extra information required to boost the
bitrate up to 96, 128, 160, 192kbps etc.etc. Downsampling is a simply a
matter of discarding the extra information without resampling the
original. I say "simply". I haven't the faintest idea how its
*practically* achieved

Sean

seani <seani@attic.localdomain> wrote in
newsan.2003.09.03.20.13.44.349746@attic.localdom ain:

Maybe you are thinking of "bitpeeling"
http://www.xiph.org/archives/vorbis/200304/0042.html
From reading the above, it would seem that bitpeeling is still very
experimental and only oggs encoded with a new (unwritten) version will be
peelable.

--
fred

<snip>

Ah yes, so it was, thanks. Thats been driving me mad