Sign in with
Sign up | Sign in
Your question

Digitizing my CD Collection w EAC: Advice Please

Last response: in Home Audio
Share
Anonymous
July 28, 2005 7:28:54 PM

Archived from groups: rec.audio.tech (More info?)

I am beginning to archive or digitize my music CD collection (~2500 albums),
and after some research, settled on Exact Audio Copy to handle the ripping
to wav. files. At this time, I am not concerned with mp3's or playback; I'm
just interested in getting at least most of the collection onto a few large
hard drives as backup to my physical collection. Eventually, I would like
to be able to play the music back through a dedicated computer as another
source for my audio system. I'm about mid-fi level in terms of equipment
and consider myself to be a mild audiophile.

I chose EAC to ensure accurate digital copies of the CD's even though my
discs tend to be in very good condition. Anyway, I've begun using EAC in
secure mode along with freedb for identification and all seems well.
However, I would like to make sure I'm doing everything properly before I go
much further with this effort.

On Hydrogenaudio forums, I found a posting that made recommendations for
settings in EAC, and I have used those for the most part. I have included
that text below:

========================================================

From Hydrogenaudio thread "Newbies guide to bit-perfect CD copying", May
2004

Required EAC Settings:

Menu Action:

* Select: append gaps to previous track (default)

EAC Action


Required EAC Options:

Extraction tab:

* Check: Fill up missing offset samples with silence
* Check: Synchronize between tracks
* Select: Error recovery quality - High
* Rest: leave unchecked

EAC Extraction


General tab:

* Check: on unknown CDs, select automatically access freedb database

EAC General


Tools tab:

* UN-Check: retrieve UPC/ISRC codes in CUE sheet generation
* Check: Use CD-Text information in CUE sheet generation
* Check: Create '.3mu' playlist on extraction
* Check: Automatically write status report after extraction
* UN-Check: Activate beginner mode, disable all advanced features

EAC Tools


Normalize tab:

* Do NOT use Normalize

Filename tab:

* Naming Scheme = %N - %T

EAC FileName


Required Drive Options:

Extraction Method tab:

* Secure mode - NOT Paranoid, Synchronized or Burst mode!
* Check: Drive has 'Accurate Stream' feature
* Check: Drive caches audio data
* Do Not Check: Drive is capable of retrieving C2 error information
EXPLANATION

Drive Extraction


Offset/Speed tab:

* Select: Use read sample offset correction. Please test and check your
offset! HERE is tutorial
* Check: Overread into Lead-In and Lead-Out (if EAC will generate [Sync
Error] on last track - uncheck) Updated
* Check: Allow speed reduction during extraction



=========================================================

I use a home built machine that includes a Toshiba SD-M1612 DVD-ROM drive
and a Sony DVD burner. After testing the drives, EAC recommended the
Toshiba for reading the CD's. The ripping goes smoothly, and I probably
average 12x to 16x overall, so it takes about three minutes to rip an album.
When the ripping is completed, I save the log, and get a message that says
there were no copy errors. (I've converted about 60 albums so far and only
two have shown any errors at all.) However, while ripping, the "error
correction" block stays blood red with some cross hatching and doesn't show
anything. Is this normal?

Overall, everything seems to be doing what it's supposed to. I haven't
tried to do any serious listening to the encoded material because I have no
way to play it back at a high quality level. (I believe this would take a
computer with an exceptional sound card, right?). In any case, over my
low-fi computer speakers, the tracks sound fine with no glitches, etc.

If anyone has significant experience with this software and procedure, I
would like to hear some thoughts and advice on what I am doing. Thanks in
advance for any help.

- Magnusfarce
Anonymous
July 28, 2005 11:42:13 PM

Archived from groups: rec.audio.tech (More info?)

"Magnusfarce" <magnusfarce@adelphia.net> wrote in message
news:a6ednfhDKqeLwXTfRVn-uA@adelphia.com

> Overall, everything seems to be doing what it's supposed
> to. I haven't tried to do any serious listening to the
> encoded material because I have no way to play it back at
> a high quality level. (I believe this would take a
> computer with an exceptional sound card, right?). In any
> case, over my low-fi computer speakers, the tracks sound
> fine with no glitches, etc.

One technique for checking the reliability of ripping is to
rip the same CD track twice, and compare the two files. They
should be exactly the same, but if your ripping procedures
are flakey, they won't be.

> If anyone has significant experience with this software
> and procedure, I would like to hear some thoughts and
> advice on what I am doing. Thanks in advance for any
> help.

Sounds like you are doing well.
July 29, 2005 3:18:45 AM

Archived from groups: rec.audio.tech (More info?)

I am also going to rip my CDs. I will use EAC and FLAC. I chose FLAG
because of the lossless compression and because it includes meta data.
My big question is about cue seets and normalization.

Regarding normalization:

a) It does not alter the actual data, right? It just adds some data
that some players will take into consideration and others will ignore
it.

b) What are the advantages/disadvantages to to normalization?

c) If I do normalization is it per album or song? I prefer it per song
because for some albums songs (should) have different volumes.

Regarding cue sheets:

a) Most likely I am not interested in creating exact copies of a CD as
I will stream songs in arbitrary sequences over a network. Should I
still care about cue sheets?

b) There are not too many programs that can convert a full CD into
individual songs. Therefore, I wonder if it is possible to store cue
sheets in each individual file?

c) If cue sheets can be stored in each individual file (i.e. not one
cue sheet for a whole CD) can I still create an exact copy of the
original CD?

Rob
Related resources
Anonymous
July 29, 2005 3:42:25 AM

Archived from groups: rec.audio.tech (More info?)

Hi there, I'm the OP. After reading the initial responses, I browsed
through the EAC settings and found that, although I was convinced otherwise,
it was still in Burst mode, not Secure Mode. However, after I stopped
throwing things and ripped a few CD's in secure mode, I noticed that the
file lengths are exactly the same as for those CD tracks that I had ripped
in burst mode (and from which I got "zero error" reports).

For those albums/tracks that were ripped in burst mode that had no reported
errors, can I assume that they are as clean as ones I rip in secure mode?
In other words is there any reason to re-rip the burst-mode-zero-error CD's
again in secure mode?

- Magnusfarce



"Magnusfarce" <magnusfarce@adelphia.net> wrote in message
news:a6ednfhDKqeLwXTfRVn-uA@adelphia.com...
> I am beginning to archive or digitize my music CD collection (~2500
albums),
> and after some research, settled on Exact Audio Copy to handle the ripping
> to wav. files. At this time, I am not concerned with mp3's or playback;
I'm
> just interested in getting at least most of the collection onto a few
large
> hard drives as backup to my physical collection. Eventually, I would like
> to be able to play the music back through a dedicated computer as another
> source for my audio system. I'm about mid-fi level in terms of equipment
> and consider myself to be a mild audiophile.
>
> I chose EAC to ensure accurate digital copies of the CD's even though my
> discs tend to be in very good condition. Anyway, I've begun using EAC in
> secure mode along with freedb for identification and all seems well.
> However, I would like to make sure I'm doing everything properly before I
go
> much further with this effort.
>
> On Hydrogenaudio forums, I found a posting that made recommendations for
> settings in EAC, and I have used those for the most part. I have included
> that text below:
>
> ========================================================
>
> From Hydrogenaudio thread "Newbies guide to bit-perfect CD copying", May
> 2004
>
> Required EAC Settings:
>
> Menu Action:
>
> * Select: append gaps to previous track (default)
>
> EAC Action
>
>
> Required EAC Options:
>
> Extraction tab:
>
> * Check: Fill up missing offset samples with silence
> * Check: Synchronize between tracks
> * Select: Error recovery quality - High
> * Rest: leave unchecked
>
> EAC Extraction
>
>
> General tab:
>
> * Check: on unknown CDs, select automatically access freedb database
>
> EAC General
>
>
> Tools tab:
>
> * UN-Check: retrieve UPC/ISRC codes in CUE sheet generation
> * Check: Use CD-Text information in CUE sheet generation
> * Check: Create '.3mu' playlist on extraction
> * Check: Automatically write status report after extraction
> * UN-Check: Activate beginner mode, disable all advanced features
>
> EAC Tools
>
>
> Normalize tab:
>
> * Do NOT use Normalize
>
> Filename tab:
>
> * Naming Scheme = %N - %T
>
> EAC FileName
>
>
> Required Drive Options:
>
> Extraction Method tab:
>
> * Secure mode - NOT Paranoid, Synchronized or Burst mode!
> * Check: Drive has 'Accurate Stream' feature
> * Check: Drive caches audio data
> * Do Not Check: Drive is capable of retrieving C2 error information
> EXPLANATION
>
> Drive Extraction
>
>
> Offset/Speed tab:
>
> * Select: Use read sample offset correction. Please test and check your
> offset! HERE is tutorial
> * Check: Overread into Lead-In and Lead-Out (if EAC will generate [Sync
> Error] on last track - uncheck) Updated
> * Check: Allow speed reduction during extraction
>
>
>
> =========================================================
>
> I use a home built machine that includes a Toshiba SD-M1612 DVD-ROM drive
> and a Sony DVD burner. After testing the drives, EAC recommended the
> Toshiba for reading the CD's. The ripping goes smoothly, and I probably
> average 12x to 16x overall, so it takes about three minutes to rip an
album.
> When the ripping is completed, I save the log, and get a message that says
> there were no copy errors. (I've converted about 60 albums so far and
only
> two have shown any errors at all.) However, while ripping, the "error
> correction" block stays blood red with some cross hatching and doesn't
show
> anything. Is this normal?
>
> Overall, everything seems to be doing what it's supposed to. I haven't
> tried to do any serious listening to the encoded material because I have
no
> way to play it back at a high quality level. (I believe this would take a
> computer with an exceptional sound card, right?). In any case, over my
> low-fi computer speakers, the tracks sound fine with no glitches, etc.
>
> If anyone has significant experience with this software and procedure, I
> would like to hear some thoughts and advice on what I am doing. Thanks in
> advance for any help.
>
> - Magnusfarce
>
>
>
>
>
Anonymous
July 29, 2005 7:38:27 AM

Archived from groups: rec.audio.tech (More info?)

Magnusfarce <magnusfarce@adelphia.net> wrote:
> I use a home built machine that includes a Toshiba SD-M1612 DVD-ROM drive
> and a Sony DVD burner. After testing the drives, EAC recommended the
> Toshiba for reading the CD's. The ripping goes smoothly, and I probably
> average 12x to 16x overall, so it takes about three minutes to rip an album.
> When the ripping is completed, I save the log, and get a message that says
> there were no copy errors. (I've converted about 60 albums so far and only
> two have shown any errors at all.) However, while ripping, the "error
> correction" block stays blood red with some cross hatching and doesn't show
> anything. Is this normal?

? Error correction blocks should only turn red when it's correcting
errors. Make sure you are ripping in 'high secure' mode.

> Overall, everything seems to be doing what it's supposed to. I haven't
> tried to do any serious listening to the encoded material because I have no
> way to play it back at a high quality level. (I believe this would take a
> computer with an exceptional sound card, right?).

"Exceptional' sound cards are rather common, actually. Or you could
bypass a card and output digitally to a receiver with a optical/coax/USB
input.

> If anyone has significant experience with this software and procedure, I
> would like to hear some thoughts and advice on what I am doing. Thanks in
> advance for any help.

I archived ~1000 CDs this way myself, to
lossless FLAC. You might consider using a lossless compression format
like FLAC, to save space. EAC can convert your waves to FLAC as you rip
them (or rather, it can pass your .wav on to a FLAC encoder, as soon
as they are ripped, and then delete the original .wav).
Anonymous
July 29, 2005 10:03:52 AM

Archived from groups: rec.audio.tech (More info?)

"Magnusfarce" wrote ...
> For those albums/tracks that were ripped in burst mode that
> had no reported errors, can I assume that they are as clean as
> ones I rip in secure mode?

I wouldn't assume that. The file *length* is dependent on the
duration of the song. Just because the two files are the same
length doesn't mean that they have identical contents.

> In other words is there any reason to re-rip the burst-mode-
> zero-error CD's again in secure mode?

I would assume that the files ripped in secure mode are more
accurate representations of the original data.

Of course you could compare some files and establish for
youself how accurate the "burst mode" rips were in your
particular instance.
Anonymous
July 29, 2005 12:04:21 PM

Archived from groups: rec.audio.tech (More info?)

On 28 Jul 2005 23:18:45 -0700, "rob" <rmdiv2000@yahoo.com> wrote:

>I am also going to rip my CDs. I will use EAC and FLAC. I chose FLAG
>because of the lossless compression and because it includes meta data.
>My big question is about cue seets and normalization.
>
>Regarding normalization:
>
>a) It does not alter the actual data, right? It just adds some data
>that some players will take into consideration and others will ignore
>it.
>
No- it alters every single bit of data. It actually multiplies each
sample value by whatever normalization factor you have chosen. Hence
it is a process that requires dither - this raises the noise floor.
Fortunately it raises it from completely inaudible to completely
inaudible plus a little bit, so no damage is done.

>b) What are the advantages/disadvantages to to normalization?
>
The disadvantage is that you no longer have the product that the
artist wanted you to hear. The advantage is that you can arrange it so
that your entire collection plays at the same volume. This actually
sounds pretty boring to me, but hey - whatever.

>c) If I do normalization is it per album or song? I prefer it per song
>because for some albums songs (should) have different volumes.
>
You can normalize exactly as much or as little as you want - your
choice.

>Regarding cue sheets:
>
>a) Most likely I am not interested in creating exact copies of a CD as
>I will stream songs in arbitrary sequences over a network. Should I
>still care about cue sheets?
>
No.

>b) There are not too many programs that can convert a full CD into
>individual songs. Therefore, I wonder if it is possible to store cue
>sheets in each individual file?
>
All CDs are individual songs - no conversion is required.

>c) If cue sheets can be stored in each individual file (i.e. not one
>cue sheet for a whole CD) can I still create an exact copy of the
>original CD?
>
>Rob

If what you want is an exact copy of the original CD, then don't even
think about this stuff - just use EAC and that is exactly what you
will get.

d

Pearce Consulting
http://www.pearce.uk.com
Anonymous
July 29, 2005 5:13:44 PM

Archived from groups: rec.audio.tech (More info?)

Thank you for the help. Based on what you said, I've got a couple of
questions.

First, how would you suggest that I compare the burst mode versus the secure
mode ripped files? I'm not set up to do it by listening, so it would need
to be some sort of software to compare the files. Do you know of anything
that would work on these wav. files?

Second: The results from the burst mode rip logs show "zero errors". Given
that these were all very clean, undamaged CD's, are we confident enough that
such files might be of lower quality to throw them away and start completely
over in secure mode? In other words, does "zero errors" mean the same thing
in burst mode as in secure mode? Clearly, I'm trying to salvage the hundred
or so album already ripped. BTW, is it reasonable to use a lesser mode as
long as I get zero error results, and switch to secure mode for problem
albums? (Just looking for the easy way out.)

- Magnusfarce


"Richard Crowley" <rcrowley@xpr7t.net> wrote in message
news:11eka9psgk227a1@corp.supernews.com...
> "Magnusfarce" wrote ...
> > For those albums/tracks that were ripped in burst mode that
> > had no reported errors, can I assume that they are as clean as
> > ones I rip in secure mode?
>
> I wouldn't assume that. The file *length* is dependent on the
> duration of the song. Just because the two files are the same
> length doesn't mean that they have identical contents.
>
> > In other words is there any reason to re-rip the burst-mode-
> > zero-error CD's again in secure mode?
>
> I would assume that the files ripped in secure mode are more
> accurate representations of the original data.
>
> Of course you could compare some files and establish for
> youself how accurate the "burst mode" rips were in your
> particular instance.
Anonymous
July 29, 2005 7:03:28 PM

Archived from groups: rec.audio.tech (More info?)

"Magnusfarce" wrote ...
> First, how would you suggest that I compare the burst
> mode versus the secure mode ripped files? I'm not set
> up to do it by listening, so it would need to be some sort
> of software to compare the files. Do you know of anything
> that would work on these wav. files?

Command Line> FC /?

FC = File Compare
/? will print out the full range of command line switches, etc.

This program will do a binary compare on any two files.
I'm confident there is more documentation on this out there
if you need it.

>
> Second: The results from the burst mode rip logs show
> "zero errors". Given that these were all very clean,
> undamaged CD's, are we confident enough that such files
> might be of lower quality to throw them away and start
> completely over in secure mode? In other words, does
> "zero errors" mean the same thing in burst mode as in
> secure mode? Clearly, I'm trying to salvage the hundred
> or so album already ripped. BTW, is it reasonable to use
> a lesser mode as long as I get zero error results, and switch
> to secure mode for problem albums? (Just looking for the
> easy way out.)

I would take the word of the ripping program to report
whether it was encountering data problems that forced it
to make replacements (or not). Likely various ripping
applicatoins are better/worse at this error reporting than
others.

I just use EAC and am satisfied with the default settings.
Perhaps I am not as obsessive about this?
Anonymous
July 30, 2005 10:00:52 PM

Archived from groups: rec.audio.tech (More info?)

"Richard Crowley" <richard.7.crowley@intel.com> wrote in message
news:D ce93g$h3r$1@news01.intel.com...
> Command Line> FC /?
>
> FC = File Compare
> /? will print out the full range of command line switches, etc.
>
> This program will do a binary compare on any two files.

FC is useless for this purpose. What you need is a binary compare that will
ignore leading and trailing zeroes.
I think the compare facility in the freeware program "Audiograbber" may be a
better choice. There are probably others that are better.

MrT.
Anonymous
July 31, 2005 11:09:29 AM

Archived from groups: rec.audio.tech (More info?)

"Mr.T" <MrT@home> wrote in message
news:42eb33c6$0$25201$afc38c87@news.optusnet.com.au
> "Richard Crowley" <richard.7.crowley@intel.com> wrote in
> message news:D ce93g$h3r$1@news01.intel.com...
>> Command Line> FC /?
>>
>> FC = File Compare
>> /? will print out the full range of command line
>> switches, etc.
>>
>> This program will do a binary compare on any two files.
>
> FC is useless for this purpose. What you need is a binary
> compare that will ignore leading and trailing zeroes.

FC does this.

> I think the compare facility in the freeware program
> "Audiograbber" may be a better choice. There are probably
> others that are better.

Both CDEX and EAC both come with file compare facilities
that resynch after finding missing or inserted data.
Anonymous
August 2, 2005 5:10:10 PM

Archived from groups: rec.audio.tech (More info?)

"Arny Krueger" <arnyk@hotpop.com> wrote in message
news:5eSdnezdcqH_LHHfRVn-qg@comcast.com...
> > FC is useless for this purpose. What you need is a binary
> > compare that will ignore leading and trailing zeroes.
>
> FC does this.

Sorry then. What are the command line switches?

> Both CDEX and EAC both come with file compare facilities
> that resynch after finding missing or inserted data.

Since the OP is already using EAC, the "compare waves" option would probably
be his best bet. Easier than using FC surely.

MrT.
Anonymous
August 3, 2005 7:57:43 PM

Archived from groups: rec.audio.tech (More info?)

Don Pearce wrote:
> On 28 Jul 2005 23:18:45 -0700, "rob" <rmdiv2000@yahoo.com> wrote:

> No- it alters every single bit of data. It actually multiplies each
> sample value by whatever normalization factor you have chosen. Hence
> it is a process that requires dither - this raises the noise floor.
> Fortunately it raises it from completely inaudible to completely
> inaudible plus a little bit, so no damage is done.
>

Huh? Normalization doesn't require dither. Bit reduction does.

>
>>b) What are the advantages/disadvantages to to normalization?
>>
>
> The disadvantage is that you no longer have the product that the
> artist wanted you to hear. The advantage is that you can arrange it so
> that your entire collection plays at the same volume. This actually
> sounds pretty boring to me, but hey - whatever.

I don't think so. You still pretty much do have what the orignal artist
intended. You just turned up, or down, the volume/amplitude.
Normalization is analogous to the volume control on your amplifier. You
turn up the volume and it increases the volume of everything the same
exact amount. Relatively,then, nothing has changed.

>
>
>>c) If I do normalization is it per album or song? I prefer it per song
>>because for some albums songs (should) have different volumes.

Ideally, you should normalize once per album. That keeps the relative
quietness, or loudness between songs the same as the artist or mastering
engineer had intended.
>>
>

> You can normalize exactly as much or as little as you want - your
> choice.
>
>
>>Regarding cue sheets:
>>
>>a) Most likely I am not interested in creating exact copies of a CD as
>>I will stream songs in arbitrary sequences over a network. Should I
>>still care about cue sheets?
>>
>
> No.
>
>
>>b) There are not too many programs that can convert a full CD into
>>individual songs. Therefore, I wonder if it is possible to store cue
>>sheets in each individual file?
>>
>
> All CDs are individual songs - no conversion is required.
>
>
>>c) If cue sheets can be stored in each individual file (i.e. not one
>>cue sheet for a whole CD) can I still create an exact copy of the
>>original CD?
>>
>>Rob
>
>
> If what you want is an exact copy of the original CD, then don't even
> think about this stuff - just use EAC and that is exactly what you
> will get.
>
> d
>
> Pearce Consulting
> http://www.pearce.uk.com
CD
Anonymous
August 4, 2005 2:39:44 AM

Archived from groups: rec.audio.tech (More info?)

Codifus wrote:
> Don Pearce wrote:
>
>> On 28 Jul 2005 23:18:45 -0700, "rob" <rmdiv2000@yahoo.com> wrote:
>
>
>> No- it alters every single bit of data. It actually multiplies each
>> sample value by whatever normalization factor you have chosen. Hence
>> it is a process that requires dither - this raises the noise floor.
>> Fortunately it raises it from completely inaudible to completely
>> inaudible plus a little bit, so no damage is done.
>>
>
> Huh? Normalization doesn't require dither. Bit reduction does.
>
What do you think happens if you normalise the level downwards? You have
dropped the level of the dither below the smallest bit, and quite likely
have a signal quantised with no dither at all. This will cause
quantisation distortion. Any process in the digital domain that affects
the amplitude should be dithered as a matter of routine. If you are
normalising up in level, of course, this doesn't apply, but then since
it can go either way, it is better to dither regardless.

d
Anonymous
August 4, 2005 2:41:51 AM

Archived from groups: rec.audio.tech (More info?)

In article <42f139a1$0$24015$ed2619ec@ptn-nntp-reader01.plus.net>,
Don Pearce <donald@pearce.uk.com> wrote:

>> Huh? Normalization doesn't require dither. Bit reduction does.
>>
>What do you think happens if you normalise the level downwards? You have
>dropped the level of the dither below the smallest bit, and quite likely
>have a signal quantised with no dither at all. This will cause
>quantisation distortion. Any process in the digital domain that affects
>the amplitude should be dithered as a matter of routine. If you are
>normalising up in level, of course, this doesn't apply, but then since
>it can go either way, it is better to dither regardless.

If you're normalizing downwards in volume, or if you're normalizing
upwards by a factor which is not an exact integer, then dithering is
going to be required in order to prevent artifacts from occurring.

--
Dave Platt <dplatt@radagast.org> AE6EO
Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!
Anonymous
August 6, 2005 11:23:04 PM

Archived from groups: rec.audio.tech (More info?)

"Don Pearce" wrote ...
> Codifus wrote:
>> Huh? Normalization doesn't require dither. Bit reduction does.
>>
> What do you think happens if you normalise the level downwards?

In the context of this thread, "normalize" appeared to mean
"upwards" (like to 0dBFS, even).
!