Is it wrong to just extract the tracks from an audio CD to my hard drive with Isobuster and than use lame to convert them to a mp3? Are the tracks digital and when I use Isobuster do they stay digital?

I tried CDex and it works pretty good but is it really necessary to use a program like that to get good results?

CDex and EAC both support some technique of reading skethed CDs. Otherwise all other rippers should be the same.....

There's nothing at all wrong to extract on your hd and then encode. I've found that to be more convenient. Example: You've just ripped and encoded in one pass, but you realize that you didn't have setting x the way you'd like it. With the one pass method, you've now got to not only encode again, but another rip.

Just rip to wav and don't delete your source on an encode till you're sure that you've got the quality you want.

Oh, bye the way, I don't know anything about Isobuster, but EAC and CDex are the standard around here.... At least on Windows.

I couldn't find skethed in the dictionary. I was wondering what you meant by that. Are you referrring to damaged or scratched? Isobuster might help with that.

I don't know of the complete usage for Isobuster but you can extract data from Cd's. I think that although you are in windows you some how bi-pass windows to extract the data. Very rarely I'll burn a CD and my drive can't read the disc but Isobuster can and will extract the data. You might want to check out the website for Isobuster. The description from the website:

IsoBuster : The Ultimate CD/DVD data recovery tool !

IsoBuster features :

- Better Error handling and several retry-mechanisms to aid you in getting the data anyway.
- More CDs stay 'readable' after problems (such as Buffer Underrun).
- Read and extraction of files, tracks and sessions from CD-i, VCD, SVCD, CD-ROM, CD-ROM XA, DVD, DVCD, ...
- Mpg (*.dat) Extraction and dat2mpg 'in one'.
- ISO9660, Joliet, Romeo (Short File-names <-> Long File-names on mastered CDs)
- Big Endian (Motorola), Little Endian (Intel) (The File System Windows sees vs. what Unix, Mac and other systems see)
- UDF 2.01 but also : UDF 1.02 (e.g DVDs), UDF 1.5 (e.g. Packet writing on CD-R and CD-RW)
- Rock Ridge Support (e.g. for Commodore users)
- Single sector extraction
- File system properties (must for FS developers)
- Neat features the OS doesn't offer
- CD-Text support
- ...

On top of this, Isobuster interprets CD image files, such as :

- *.DAO (Duplicator), *.TAO (Duplicator), *.ISO (Nero, BlindRead, Creator), *.BIN (CDRWin), *.IMG (CloneCD), *.CIF (Creator), *.FCD (Uncompressed), *.NRG (Nero), *.GCD (Prassi), *.P01 (Toast), *.C2D (WinOnCD), *.CUE (CDRWin), *.CIF (DiscJuggler), *.CD (CD-i OptImage), *.GI (Prassi PrimoDVD).

It seems to be aimed at data rather than audio. Audio error detection uses completely different algorithms. I don't know Isobuster, but I'm quite sure that EAC is the best ripper for audio error detection.

http://www.exactaudiocopy.de

Exactly! That's why I made my post, I don't know anything about this subject and wanted to understand more about the extraction of Digital audio and if it's necessary to use a special program or process for extraction to preserve the digital audio. I don't even know what the term "ripping" really means. Do you have to follow a certain process to call it "ripping"? I'm pretty sure you just can't go into windows explorer and copy the digital tracks to your PC, is that correct? Isobuster probably wasn't designed for audio extraction but it might do that also.

I hope this clerifies my original post and the purpose for it. I'm not sure if the posts that were made actually answered my question since those people might not know anything about ISObuster.

Ripping or grabbing is the process of getting the contents off an Audio CD.


Yes, that's correct. You would end up with a bunch of "shortcuts".


I know Isobuster and use it myself sometimes but I wouldn't recommend it for audio extraction either. I don't know how it rips audio tracks, you would have to ask the author. But it probably doesn't have the advanced options that EAC has.

Thx for the info. I used Isobuster to extract 10 audio CD's to my drive and ran a bat file that converted all 124 songs non-stop which took 100 minutes. I'm very new to this but I used this command line:

"C:lamelame.exe" -v --vbr-new -V 2 "CD-foldersCD1Track 1.wav" "mp3-foldersmp3-CD11.mp3"

I was very happy with the results but like you mentioned Isobuster probably doesn't have the advanced options that programs like EAC have.

[img]6ea3c00449[/img]

Encoding will take the same time, regardless which ripper you use as you have the files on your harddisk and then start lame or your batch file or whatever. Only to your information, you can also setup EAC to encode your songs on the fly.

QUOTE

Do you know what your command line does? -v is not needed, and --vbr-new doesn't mean it's better only because it's new! Rather try --alt-preset standard which gives you better quality and has a good quality/size ratio.

Since nobody knew how IsoBuster extracts audio, I decided to give it a try.
It seems to use normal burst copy, it's fast and doesn't detect errors. Unrecommended.

Does "on the fly" refer to extracting from CD directly to mp3? It would seem to me that there's a better chance of gettting good results when extracting to your hard drive first.

I did try --alt-preset standard before I made my initial post. I thought I had to install the entire lame program "Lame 3.90.2-ICL" for that preset to work, which I did install. Now I know that it's built into lame.exe. It's pretty amazing what that little lame file can do. I think I decided to use --vbr-new -V 2 because it was much faster.

I did a comparison using a Sting album
--vbr-new -V 2--------------8 minutes to encode - 63.8 mb's
--alt-preset standard----31 minutes to encode - 63.12 mb's

Why would --alt-preset standard take 4 times longer? I also noticed when using --alt-preset standard it came out joint-stereo, with --vbr-new -V 2 it comes out stereo.

I guess I had good results because my CD's are in perfect condition. I didn't get any pops. I'm not sure how else to tell if the quality is actually good except by listening.

Well the best one to answer that would be the writer of the presets, Dibrom. All I know is that the alt-presets do a better job, and good things take time

If you use --alt-preset fast standard (which I will abbreviate as "--ap fast standard"), the encoding job will take about half as long. I could be wrong about the details, but I think the "--ap standard" preset uses the vbr-old model, while "--ap fast standard" preset uses vbr-new. The only reason for the two distinct presets is because vbr-new implements the preset's code-level tweaks differently than vbr-old, and in some cases "--ap fast standard" won't sound quite as good as "--ap standard".

Two short remarks :

If I remember well, Alt Preset Fast Standard is not necessarily worse in quality. I think it is very slightly more unstable : sometimes a tiny bit worse, sometimes a tiny bit better than the regular APS.

As everyone will tell you here, joint stereo is better. The idea of regular stereo being better is a old myth started with a bug in the Fronhaufer encoder, and now perpetuated like a dogma by some traders who require 192 CBR stereo MP3s (while Alt Preset Standard, or even Alt preset 192 are much better).

More info : http://www.hydrogenaudio.org/forums/showth...s=&threadid=759

(hey ! This is in the beta FAQ ! When will it be online, and how will we Moderators know it, since for us it is already online ?)

Audio extraction basics

Computers CD ROM drives are mainly made for reading... CD ROMs !
A CD ROM contains files, like the hard disc of the comlputer, and they can be copied from the CD ROM to the Hard Disc.
Audio CDs don't contain files. The data is stored a bit differently. A computer can recognize an audio CD. When you play an audio CD with Windows Cd Player, Windows doesn't read the CD, it just sends an internal "play" command to the drive. The the drive, plugged into the soundcard, plays the CD by itself. The audio data are not read by the computer in this process.

But for audio processing purposes, programs were made that can perform a "Digital Audio Extraction" (DAE). They read the audio data, and convert it (without changing it) into files for computers. These are WAV files.

Any burning program : Nero, EasyCD, etc can perform a DAE. At the beginning, not all CD ROM drives were capable of DAE ! Now, more and more multimedia players (Windows Media Player 7, Winamp 3 ...) play audio CDs performing a DAE and sending the data to the soundcard internally, without cable, instead of asking the drive to play the CD with a sound quality that could vary from drive to drive.

But for a reason I couldn't explain because I don't know it (Laziness ? Patented algorithms ?), Audio extraction programs don't reproduce the playback of CD Players.
During this playback, three levels of error correction occurs :
The C1 level is quite active, and corrects the erroneous raw data, that is not readable error free. The presence of this error correction level allows to store much more data on a CD than if the raw data was required to be read reliably without error correction.
The C2 level is less active, it corrects the errors that couldn't be corrected in the C1 level.
Last, when the C2 level fails, the original data can't be guessed any more. Then the player "fills the gap" with some interpolated (=guessed) data.

This last step is not performed by audio extraction programs !

This step occurs very rarely on brand new CDs in good players. Usually never. So when you rip a CD, the result is often perfect.
But if your original CD is not perfect, if it is a CDR that was poorly burned, if it has scratches, or fingerprints, or if it's old and no more in perfect state, the last error "concealment" process is required for getting good sounding results.
Otherwise, and it is the case in computers, there are audible clicks in the extracted file.

EAC was developed to detect these clicks during the extraction, and trying to read again and get a proper result for the faulty data. Other programs have some options of this kind, like CD EX, or CDP32, but they don't match EAC, that is very complete and has a lot of advanced options for audio error detection and recovery.

Comparison using a "Sting" album

--vbr-new -V 2-----------------8 min to encode, 63.8 mb's (vbr new model) came out in stereo

--alt-preset fast standard--16 min to encode, 63.7 mb's (vbr old+new model) came out in joint-stereo

--alt-preset standard--------31 min to encode, 63.12 mb's (vbr old model) came out in joint-stereo

Did I get the vbr models correct?

I read in a Besweet document that you shouldn't use vbr new model. Too late I already encoded 10 albums with vbr new model and I'm not doing it again since I deleted the wav files off my drive.

What is the reason that sound quality should vary when a CD is played digitally? I can't imagine any differences. Or do you mean the implementations of error correction?


vbr-old is recommended because it was heavily tested, which was not the case with vbr-new.

That was true in pre 3.90. There was vbr old, mtrh and new.
Now vbr-mtrh is in place of vbr-new. This mode is recomended. (at least in alt-preset fast standard)

The last part of the sentence ("instead of") was refering to analog playback, not digital.

vbr-mtrh must be the old&new merge, it's twice as fast as vbr-old and half as fast as new. I mistakingly used vbr-new but the results were still very good. I actually wouldnb't know what to look for. It took 100 min to encode 124 songs on a 750 Duron. and I was using a file sharing program at the same time.

Just in case somebody doesn't know, vbr-new = vbr-mtrh in latest versions of LAME.

I persoanlly make an image file of an audio cd using Clone CD. Then mount that CD using Daemon Tools, and extract (using EAC in burst mode, but secure gives fast results as well) to wave and convert to whatever format I feel like. This way I only have to touch the original CD once, when making the image file.

I haven't found any quality differences between ripping straight from the CD and ripping from a mounted image file, but I am using new, untouched CD's. I would think that for a "mediumly worn CD" that this method would work just as well (but faster) than using EAC in secure mode. The reason for this is because the image file still needs to have a complete image, so on scratched data CD's someform of filling in the gap has to happen (and whatever method this is works prefectly). On heavily scratched CD's EAC in secure mode would be your best bet.

I still don't quite understand why people fuss so much about scratched CD's. Sure you might have some older ones that you want to rip and encode, but there are ways of fixing that (in all but the worse cases). When I purchase a CD, the first thing I do is make the image file, burn a copy from that image file, then encode with the image file. Then I gently place the CD back into its original jewel case, and stow it away on a self. If I need a "decent" copy of the CD I go to my Ogg -q 5's that I've created (I'd like to move up to flac or MAC, but currently don't have the storage needed), and if I need a "prefect" copy I still have the original untouched CD on the self to image.

But, if your borrowing from friends converting to a lossless format is probably your best bet.

Yes. If I remember correctly, it seems one of the more obvious samples that shows --apfs has lower quality than --aps is "dogwhistle.wav". Other samples don't give big difference. IMO it is quite safe to use --apfs if you want speed as well as quality.

Yes it fills the gaps with read errors, exactly like burst mode. The creation of the image file is no more than an extraction in burst mode. It's not secure at all.

the Only BeSweet document that discuss LAME is this..

It might be an out-of-date document from Doom's site from 12/2001. The heading is called AC3 to MP3 using azid & lame. There is a large picture of BeSweet GUI v0.5 by DD (23.12.2001) [Win2K Mode]. I have a print out but I don't have the url.