I want to archive my CD collection by putting everything on a hard drive at
the best possible quality. Does this have any hardware requirements either
in the recorder or soundcard areas? Is there a recognised 'best format' to
use? And is there any truth in the argument that quality does deteriorate
when transferring from one format to another i.e. CD to xyz, then xyz to
MP3?
K

Ken wrote:


I use EAC 0.99 prebeta 4 and encode to FLAC format which offers very
good quality at the expense of file size, say 330MB per ripped CD.

http://www.exactaudiocopy.de/en/inde...rces/download/
http://www.rarewares.org/dancer/danc...f=lame-current (Lame
compiled for Win32)




Generational loss does happen with lossy codecs. IOW FLAC to 128kbps MP3
format is of better quality than 320kbps MP3 to 128kbps MP3. OGG to 128
128kbps MP3 would be about the same as the last example.

MP3, VBR or OGG are good choices where compromises must be made. FLAC is
better for archival purposes.

"Ken" <1n32b@optus.net.au> wrote in message
news:ejfQ7OEzIHA.4808@TK2MSFTNGP02.phx.gbl...
If you want to preserve the sound quality of the original CD then a .WAV
file sampled at 44.1 Khz using 16 bit resolution is the only way to go.
This records the CD data in the same format as is on the CD. Using one of
the many codecs that are available to compress the data into a smaller space
(e.g. you mentioned MP3) also looses some of the data on the way. Thus when
you uncompress the data, the result is not the same data that you
compressed. This is because the compression is what is known as lossy.
Many compromises are made and some codecs are often acceptable for most
purposes. There are compression algorithms that are not lossy (for example
compressing all the .WAV files into a ZIPped folder), but the nature of the
data that forms a sound .WAV file invariably means that the degree of
compression achieved rarely makes it worth the effort.

M.I.5¾ wrote:
But if he is willing to suffer a slight reduction in quality, he can sure
gain a tremendous savings in disk space. For example, I think a 128 kbps,
Joint Stereo, mp3 file is pretty good, and is roughly 1/10 the size of the
corresponding WAV file.

Hi - and thanks for some very useful advice. HD space is not an issue. The
goal is to have a digital copy on my PC as close to the original quality of
the CD as possible so that, should the CD be damaged at some point, I can
burn another with no loss of sound quality.

The other question related to using my new iPod. Unless there's a better
format I'm going to make MP3 copies of some of my music for use on the iPod
but would never be thinking of making CDs of MP3 copies. The question was
put because I'd read somewhere that it's best to go straight from CD to MP3
because there is a quality degradation if , e.g. you go from CD to format A,
format A to format B and then format B to MP3. (I don't pretend to
understand why anyone should do that but it was discussed in a book I read
on iTunes, iPods or whatever.

BTW I take it that MP3 is the best format for iPods? Again storage space
isn't an issue - it's a Classic with 80GBs - and I know that there are
obvious limitations in listening to music on a iPod. I'm certainly not
intending loading it up with every piece of music I've got but I just feel
that I might as well try and get the best sound out of it as I can.

I had a look at EAC 0.99 prebeta 4 and the idea of encoding to FLAC format
and need to look into it more closely. But I'm a newbie to audio and that
program seems to pre-suppose a degree of technical know how I don't have. I
have Nero and Roxio. Would they be adequate for making WAV copies?

Regards,

Ken

Ken wrote:
Just do keep in mind how huge those files will be. As a rule of thumb,
it's in the order of 44,100 samples/sec X 16 bits per sample X 2 for
stereo = 1,411,200 bits per sec. (OR: 1,411,200 / 8 = 176,400
bytes/sec). So for a minute, multiply by 60, to get: 10,584,000 bytes
per min, or 10 MB a minute!! One hour is 600 MB, or 0.6 GB!!

Every time you make a conversion there is a slight (but often minimal) loss.
Generally speaking, it's often pretty small, and is often pretty hard to
hear. But compress it enough times and you will. :-)

FLAC is lossless, and (consequentally) can't compress music very much, so
you don't really gain much in terms of file size reduction (as contrasted to
mp3s, for example). I'd say, skip it.

OR go to mp3's, where you can choose the compression level in the software.
Note: The 128 kbps, Joint Stereo (or Stereo) mode is the most common mode
used for mp3's, where you gain a 10:1 or better filesize reduction (as
compared to the WAV format). You might want to read up on some of this,
either in the Nero or Roxio help files, and/or at the Wiki site, for
example.

They should be. And check out their help files, too.

"Bill in Co." <not_really_here@earthlink.net> wrote in message
news:O1cKPoPzIHA.5716@TK2MSFTNGP04.phx.gbl...
Something wrong with your theory somewhere. A 650 MB CD holds 74 minutes of
audio data, and that includes an overhead for error correction. That means
that one hour of data occupies 527 MB. Take out the error correction and
and the PCM data is down to 465 MB. Since a .WAV file is in the same format
as a CD, the data occupies the same space: that is one hour occupies 527 MB.

I'm not disputing your arithmetic, but it doesn't match reality.

M.I.5¾ wrote:
Is the overhead for error correction that large? (62 MB)

I don't know why there is that apparent discrepancy. Maybe you do.
Maybe in the assumption of a (standard) 8 bits per byte being used above?

Does it really?

JohnO wrote:
Is that the fallacy? I'm still trying to figure out what is errant in
what I wrote out. The math seemed right to me (for computing the size of a
wav file). :-) But clearly something was off.

"Bill in Co." <not_really_here@earthlink.net> wrote in message
news:%23ky2I8YzIHA.2384@TK2MSFTNGP02.phx.gbl...
Helluffino.

IME 1 minute of PCM from CDs is very close to 10 MB, so your math looks
close. I'm really not sure, but the 650/74 is marketing, so it automatically
gets my skeptic eye. ;-)

-John O

"Bill in Co." <not_really_here@earthlink.net> wrote in message
news:%23fErGqSzIHA.1236@TK2MSFTNGP02.phx.gbl...
Yes.

So much so that white book Video CD found it very advantageous to drop it in
order to gain a significant increase in data rate.

"JohnO" <johno@!NOOSPAM!heathkit.com> wrote in message
news:u6dAe7azIHA.6096@TK2MSFTNGP06.phx.gbl...
650 Mb CDs really do hold 74 minutes of CD data. This was the maximum size
that was originally specified in the red book (CD-A) specification. In
practice Audio CDs appeared that hold much more audio than technically they
should have done and the players had no problem. The largest pre pressed
audio CD that I am aware of plays for 96 minutes.

Most blank CDs that you buy today are 700 Mb and they hold 80 minutes of
audio. 900 Mb disks are available, but few burners will recognise them (and
those that do have usually had a firmware upgrade).

However, now I think about it, the quoted capacity *excludes* the error
correction data so the actual total data quantity is higher than this at
about 740Mb for a nominally 650Mb CD.

Bill in Co. wrote:
I still don't know why there is this seemingly apparent discrepancy, per my
calculations above. Is it due to some marketing as JohnO suggested, or
what?

The way I computed it, the wav file (not taking into account any overhead
for error correction, etc) comes out to 635 MB for 60 mins (using 10.584
MB/min X 60 min).