Hi

I've been reading articles here and in at other boards for about an year and, if I'm honest, I'm a little bit fed up with all that technical stuff. I really like it but it's quite hard to understand all that error correction algorithms in detail as I do have just little basic knowledge about all that routines. And now, after buying all the equipment necessary for ripping I'd like to start to rip. I just would like a final affirmation so that I can be sure I've done everything right.

My goal:I want to be able to produce bit identical copies of CDs.

I decided in favour of "Image+Cue" as it is one way to preserve the pregap before track one. I know there is also the method to add the pregaps at the beginning of the following track but I consider this method as uncomfortable.

Once again I want bit identical rips. If I've forgotten anything, tell me immediately. In addition: with bit identical I refer to the audio data and the CD layout. Whether there is CD-TEXT or not doesn't matter.

My drive: Plextor Premium (internal version)--> because of its capabitlity to get back the data ususally lost because of the offset by overreading into the lead I/O. (Is it right that I won't lose any data using this drive?)

Used format: *.wav

Here is my configuration: (just the points that have influence on track quality)


Drive options:

Extraction method:

+ secure mode
+ drive has accurate stream feature
+ drive caches audio data
- drive is capable od retrieving C2 error information (@ Pio2001 I've read some of your posts dealing with C2, what is the result of your discussions? Is it better to turn on the C2 option or to leave it unticked?)

I got the read command, gap detection method and all the other setup informations from uli-76's offset-DB.

The compression options:

Do I need to enter any compression options if I always extract to wave files?

The question is because I noticed some difference. If I use the MS PCM Compression at 44,1khz 16bit stereo (this pcm compression is just used when I order EAC to extract compressed, isn't it?) then the file is slightly different in size to the file I extracted uncompressed although they should be of the same size. Shouldn't they?

compressed: 36,4 MB (38.191.834 Bytes)
uncompressed: 36,4 MB (38.191.820 Bytes)

What shall I do?

And now the last register card, Eac options:

exctraction:

- fill up missing offset samples with silence (no, as I shouldn't have missing samples because of my drive.)
? no use of null samples for CRC calculations
? synchronize between tracks (yes/no? I've heard it's just necessary if your drive doesn't have accurate stream?!)

Finally: I use the ASPI Interface

So, thanks a lot for reading.

Will I get bit identical, absolute perfect audio data and structure when working that way? Every sample has to be like it is on the original cd!

so whish me the best for my first rips.

bye Fabian

PS: If you have got any improvements, please tell me. I really want perfect copies.

This has been discussed multiple times, you should search on the forum.

Very true.



Fabian, I'm a little confused. You say you want bit perfect, but then say that you don't want cd text or whatever. Well now you've lost the concept of bit perfect. Another thing you don't take into account is that the cd format is made to use error correction.

PCM is not a compression format, it's just a lossless format. About the difference in size, it comes from the wav header, in front of audio data. It does not change the content. The raw stream has no header, because it's just plain raw data.

Indeed. I find it hard to believe ff has been reading the forum for a year, yet missed my recent posts on this very subject =).

As him, bit identical copies of audio CDs is what I am after, though the lengths you have to to achieve the goal sometimes leaves me wondering if it is worth the time, effort and money.

I found EAC produce inconsistent results. Some CDs will copy perfectly, both the layout and the audio data; while others will have some tracks with mismatched CRCs. Therefore I recommend PlexTools Pro instead. This software comes with the Plextor Premium CD-RW and the Plextor PX-712 DVD±RW drives; it automatically applies the correct read/write offsets when used with these drives (in EAC you have to specify +30 and -30 manually).

However, it is important to note that in versions starting with 2.13 the Copy CD/DVD function is broken and the copy will not have the same CRCs as the original. So v2.12 would be the recommended version for the task.

It is also worth noting that I've had zero success with PTP's Audio CD Maker function in this matter. Either the layout or the CRCs will be correct, but never both. For those willing to experiment I can provide the results of my findings; but in short , it is little gain for a lot of pain.

If EAC is used, be advised that it does not detect the read command correctly for either of the above-mentioned drives. Its guess (D8) will often result in a sync/read error on a good CD. Instead, I found MMC1 to work just fine.

Other points of interest re: EAC
- the Premium C2 reporting is fully trustworthy and should be acknowledged accordingly in Drive Settings
- the other options mentioned seem to be irrelevant when using the Img+CUE button (or Action->Copy image & create CUE sheet->uncompressed..., or Alt+F7)

HTH

hi

I absolutely didn't want to annoy anyone by asking a multiple discussed topic once again. I'm nearly sure I've read most of the articles yet as I'm lerning about EAC for more than a year.

The goal of this thread is just to get a last affirmation before I start to extract audio data. It would simply make me feel protected from any mistake and as a consequence a rerip. But I'm absolutely sure you know what I mean.

Once again sorry if someone feels irritated. I never meant to do this.

I just want a yes, you can be sure or a no, you've forgotten something.

Concerning your questions:

Well, maybe the phrase "bit identical" is a little bit confusing.

What I want are bit identical tracks, and a bit identical layout (track, gap pattern)

What I do not care about is CD-Text and such things.

I just care about the audible date even if it's just a single sample as well as about the correct track marks.

Is it more comprehensible now?

bye, fabian

appendix:

Are there any other things I have to care about like gapless playback (I know I needn't worry about this when using lossless code, it should just be an example)

Hello, Festenflieger



I've just bought it today. It's still inside its box. I'll test this this week. I can already tell that only Plextool Pro itself might use overburning when you burn the copy. EAC won't overburn on a non-Teac drive.



The result of the discussions is that no one conduced yet the necessary tests in order to answer this question.



Compare then with EAC (Tools\Compare wavs) in order to ensure that they are bit for bit identical. The header is always 44 bytes. I think that the compressed version has 14 bytes of metadata at the end of the file.


Unchecked.



Yes, it doesn't hurt.



That's not sure. Some isolated cases of consistent errors have been reported (same wrong CRC). It doesn't seem to occur more than once in 1000 tracks. Maybe less.
Accurate rip, if your CD is present in the database with the same pressing as yours, would guarantee you 100 % perfect rips. It's still rare to find a CD in the database, but the more people use it, the more CDs will be recognized.

But accuraterip only works when ripping tracks as individual wavs, not image+cue.

Hi

Now I have to go to school again so I just have very little time. (It's my last year at grammar school so I have to work quite hard.)

@ nerver_again

Sorry, I've searched the forum once again but didn't finde any propper articles! Maybe you can give me a link.

What I'm not searching for is tech stuff. (I can care about that stuff later when I'm already extracting.)

For now I just want to get the setup for EAC so that I get best possible results concerning Audio Data and Layout (gap/track pattern). What I'm not caring for is CD-Text, ISRC, .... and such additional information.

I was really searching hard but couldn't finde any!

I'd really like to buy the CD-DATA from the lables. I'd save so much time. The only download service that offers *.wav is allofmp3.com but as far as I know they rip theirselves. So the quality won't be better than EAC, right?

Back to my configuration:

Apart from the consistent errors (they are called that way, aren't they) that can occur with my setup+drive combination I will get all audio data (every bit) exactly extracted?
Right?

Can you give me any usernames other than never_again that are on the search for the perfect 1:1 copy? so that I can contact them?

bye, fabian

(I've just bought the equipment for a 0dbA Computer, you just can hear the the hdd running) maybe I can finisch my 19" rack by christmas!

I think most of your questions are answered in the FAQ. A search on "offset" will probably also give you some good info. But from what I can see your setup and drive seems to be as good as it gets.

fensterflieger,
you're on a good way.


Why ? I think checking it also doesn't hurt. I've always enabled it and it doesn't cause any problems, but it makes the copy more exact.

fensterflieger,
do you plan to write your images back as audio CDs (a good plan) ? Then I have this and this thread for you.

The EAC tutorial
http://users.pandora.be/satcp/tutorials.htm
will have answers on many of your questions. It does a much better task than the faq, since they are only links to discussions. The tutorial is a very good guide, compressed information.

That option doesn't make copies more exact in any way. It just tells EAC if it should count zeros (digital silence) when calculating CRC or not. Therefore is useable only if your drive has a negative offset (and doesn't overread into lead-in) or has a positive offset (and doesn't overread into lead-out) AND if "fill up missing offset samples with silence" is checked. Otherwise it can result in wrong CRC calculations.

hi

it's weekend and i've some time to do some reading so my first task for tomorrow is to read coaster factory once again. it some time ago when i read it the last time so maybe i skipped some important notes 'cause i didn't understand at that time.

have a nice day! fabian

edit: yes i want to write them back to cd ..... perfect copy ..........! (except of cd-text and so on)

PS: as my drive supports overreading into lead I/O I needn't check the null samples option right! the only function is to get the same CRCs with different drives! right?

Right, you can leave both "fill up missing offset samples with silence" & "no use of null samples for CRC calculations" unchecked.

hi

first of all thank you for your answers.

so now i got affirmation that i don't need to check the null samples option as it would be nonsense with my drive (does it just depend on the drive or are there any other factors, i can't think of any)

but: one day a user on digital-inn.de said that even with the plex premium i'd lose some samples as it doesn't have the capeability to overread the whole lead out.

but 30 samples are not that much and the plex premium should do that job, shouldn't it?

then he added that this is just when music is coded far way into the lead out. but how can that be? the lead out is after the last music sample. so usually i needn't overread the lead out to get the music data the only reason why i have to do so is to correct the offset!

wait a minute i'll search the link!

bye, fabian

dear note, please remember me asking about the technical background of offset mechanism.

The FAQ does link to this tutorial...

In fact, it is not the real lead-out of the disc ... imagine the term 'lead-out' as the area that is recognized by the drive as lead-out due to its read or write offset ...

dealing with a drive that should overwrite into the lead-out is quite simple ... just add a second of silence to the end of your wav-image and you'll be fine ... this won't be bit-identical, though ...

Please make sure which values you're talking about. Your sentence is only logical if the values you mentioned are correction values, not the actual read offsets. This "CRC with null samples"-option of course makes the copy more exact.
If a drive
-either has a negative actual read offset value and the ability to overread into lead out (my Plextor)
-or has a positive actual read offset value and the ability to overread into the lead in,
there won't be offset samples missing, so the option is irrelevant. "CRC with null samples" then ensures that the correct number of null samples at the beginning and the end of each track, especially at the beginning of the first and the end of the last one, is extracted.




Music data (non-null samples) can often be found after the actual end of the CD, several thousand samples. This is especially on later pressings of older CDs. This hidden part should contain only noise.

That's what I meant.

Can somebody confirm this? ...Pio?
As to my knowledge, "No use of null samples for CRC calculations" option doesn't affect extraction in any way, it just tells EAC how to calculate CRC, which is done after ripping. It doesn't ensure number of samples to extract nor anything like that. I think you confuse that option with "Fill up missing offset samples with silence".

Let me paraphrase it.

You don't need to muck around with EAC settings if you have a Plextor Premium. Use PlexTools Pro v2.12 or earlier and its function CD/DVD Copy. Under Preferences be sure to have Audio Read Error Recovery for AudioCD set to 5. Recover the best bytes (least sectors) per sector.

The above method produces CDs that are bit-identical copies of the original as far as the layout and the audio data proper are concerned.

Now I absolutely don't want to offend anyone <wink>, but which part of the above is hard to understand or is bullshit? Anyone with a Premium (or a PX-712), PTP ≤v2.12 and EAC can easily check it out for themselves.

edit: HTML tags and a typo

Quite correct.

I have a PlexWriter Premium and that is my method, though I am using the latest version of PlexTools 2.16. No bullshit.

Thank you!
But when you copy an audio CD with PTP v2.16, do the copy's CRCs match those of the original CD? I found that starting with v2.14 (2.13 for the PX-712A) they do not, though the layout is reproduced faithfully.

CRC OK, while "don't use null samples" is unchecked, means that the same amount (not necessarily the correct one) of null samples was ripped during the test and the copy.
If you don't use null samples in CRC calculation, you can have CRC OK with a different amount of null samples in the test and the copy.

I've never paid any mind to the CRCs to tell you the truth. I didn't think matching CRCs was very important so maybe Pio2001 can share some of his wisdom about them...?

Sorry, my Plextor Premium is still in its box

I just tested this with a PX-708-A and Plextools Pro 2.16 using the CD/DVD copy feature.

Original Rip


Copy


The copies are bit identical.

Very interesting!
But it is not quite clear how you get your CRCs here: do you rip the original and the copy with ? and then run CRC32 on them? I was talking about EAC's internal CRC checking.

Do both discs produce identical CRCs when read in EAC?
Thank you.

For that test above I used Plextools Pro to rip the tracks aswell. I don't see why EACs rip should produce anything different but I tried it again to satisfy our curiosity.


For the one disc I just tried, yes.

That is exactly what I wanted to say, thank you Pio. I'll do some CRC-calculating tests, though.

2 fensterflieger: No one noted for you, that you must have checked the option "add-gaps to the next track" (not the default option), because there are cases (CD's obviously) that have 32 or 37 frames of silence added to the basic 2 seconds pregap before the first track. So If you want a bit by bit identical copy you must go this way to extract even the pregap before the first track! There are CD's, I experienced only two, that have a secret track before the real first track (Silent Stream of Godless Elegy - Themes '00 and Winds - Reflections of the I + Of Entity and Mind (MCD) '01) so that's it to clear it up a bit...

hi neighbour

thanks a lot for your answer. answers like yours are the one i'm searching for 'cause they aren't mentioned in most tutorials as many people don't bother abour a couple of samples but i want them. (is this case mentioned on coaster factory?). Although this is an important note i needn't care about this case since i use images! then i needn't care about the pregap, need i?

another question about the 2secs. necessary gap before the first track:

is is a gap like each other track on the cd or is it a different one?
why is it not shown in eac? when i extract the gap before track on do i also extract this 2 secs. or just the additional time so x of 2secs.+x???
i've once read about non standard 2.x secs of silence? what about this case? does eac show the gap-lenght additional to the 2 secs.?

bye, fabian

whatever happned to CloneCD? Or was that meant for Data Copying?

It is a bit different, because the audio sector number zero is always exactly at the end of this pregap. It is considered differently by the software at least.



Because EAC doesn't consider it to be part of the audio data.



Only the additional time.



Yes.

CloneCD can copy audio. If faithfully reproduces all errors that happens when the original is read.

This is another interesting question: What does EAC exactly extract using the "copy image and create cue sheet" function ? I'll check if the pregap of the first track belongs to the image. The pregap of the audio session surely not does, because EAC normally can't accsess it during ripping.

Here's a post by Pio regarding the issue (this one and the last two posts in this thread by myself and Pio)



Pio, I can't believe that you've never looked closely enough at EAC. After using the "detect gaps" function, the pregap of the audio session + pregap of the first track is shown in EAC's main window as being both together the pregap of the first track. In its "test gaps on silence function" only the pregap of the first track is analysed and shown (in fensterflieger's terminology: x).

Yes that's right, oddly enough, it is shown in the gap lenght, but not in the start time.

Example :
Track : 01
Start : 0:00:00:32
Lenght : 0:03:29:65
Gap : 0:00:02:32

So the gap extends before the beginning of the CD .
Actually, when you enable "overread into lead-in", what EAC calls "lead-in" is this 2 seconds stadard gap.

I didn't noticed it, because I nearly never detect gaps. Only if I am not sure if the CD has a hidden track.

I compared the CRCs of 2 test&copy rips of the same track, both CRC OK, first CRC with, second without null samples; they were different. So 'use null samples for CRC calculation ?' affects the CRC.


Due to my analysis, EAC (correct offset settings), using the copy image function, rips the pregap of the first track, but not the pregap of the audio session.

Sure it does. When "No use of null samples for CRC calculations" option is enabled, the track with digital silence at the beggining and/or end will give the same CRC as the same track without that silence. When that option is disabled the same set of such a 2 tracks will give different CRC. That's all that option does, it's as simple as that, glad you got it at last

Yes! That's absolutely right! BTW: Why are you trying this? When you have a correct offset correction already set-up and your drive is able ti access into lead-in, there's no need to do such an experiment...

Another thing is that not all of the drives which are said (by EAC) that they can read into lead-in can do it! I have 2 drives TEAC - My burner CD-W552E can do it and my DVD DV-516E not. That's my observation...

And another thing I've noticed... There are CD's which are strating and ending with (white or another) noise (that can't be heard) no metter the offset and overreading into lead-in/out, that's fact. But it's very rare and it seems that this is maybe a manufacturing problem...

Sometimes there are CDs that I want to rip to an image. Perhaps this CDs contain a hidden track before the first track in the pregap of the first track. Thus I get this hidden track without checking.

Hi

@precisionist

so EAC also gets the gap before track 1 but doesn't get the 2 secs. of session pregap! right?

can there also be a hidden track in the session pregap or just in the gap before track one? so in my mystic "x"?


@ pio2001

Have you already tried your premium?

bye, fabian

The two hidden tracks that I've got do extend into the session pregap. The two first seconds are thus very difficult to retrieve.



I plugged it last week, when my Yamaha CRW3200 died. I didn't test its features yet.

When ripping an image using proper offset correction and maybe 'suitable overreading capability', yes.

If I remember correctly, I've found only one hidden track in the pregap of the first track so far and it didn't extend into the pregap of the audio session...