CDEX & Lame 3.96+ Tutorial or How to

Hi Folks,

I’m looking to backup my music collection soon and I’ve decided on CDEX to help me out with that using a LAME encoder. EAC looks really good as well, but my ears are not so good and I doubt I would hear the difference (as I’m happy with even 128’s sometimes). I hear that vers. 3.96 is the best version of the codec out there - is this true? I would like to find a tutorial that shows me how to configure it in CDEX to get excellent 192 or 224 VBR files converted to MP3. Any ideas? I’ve tried searching, and it seems there is only basic CDEX tutorials or Advanced EAC ones. Please help.

Thanks. :slight_smile:

This comes in handy:
http://www.free-codecs.com/guides/CD_Ripping_Encoding_Guide.htm

FYI, Hydrogenaudio now recommends the latest LAME 3.97b. But I guess that I’d be never able to hear the difference… :disagree:

Regards, :slight_smile:

ET

here are some encoder settings that you can choose from, just click the images to enlarge, and check the same options and just try them out.

I have provided 3 different options: VBR-ABR, Classic VBR and old time favourite CBR, i have checked the options that give the better quality to file size ratio. It’s up to you and your PC about ripping and encoding on the fly.

BTW my encoder is not updated to the latest release, but these settings are valid for all Lame engines.

01 - VBR-ABR settings 128=224 kbps stereo VHQ (q=0)

02 - VBR settings 192=224 kbps stereo Normal (q=5)

03 - CBR settings 192 kbps stereo VHQ (q=0)

From the CDex Manual

Quality (default value Normal):

With the LAME encoder, you can specify the output quality; thus you can trade off encoding time against sound quality. The default (normal) is recommended for the lower bitrates (<160 kbps), high quality for bitrates >160 kbps.

VBR Method Setting

The VBR method setting allows you the change the VBR algorithm which is used for the encoding (detailed information can be found in the LAME user documentation - see http://www.sulaco.org/mp3/doc/html/index.html). The following selections are available

Disabled: Don’t use VBR; instead encodie with a Constant Bit Rate (CBR)
VBR-Default: Use the default VBR method (currently set to VBR-MTRH)
VBR-Old: LAME’s first functional approach, based on masking, bisection in the bit domain.
VBR-New: LAME’s second approach, based on masking and direct noise allocation.
VBR-MTRH: a combination of old and new (VBR) routines
VBR-ABR: The Average Bit Rate (ABR) setting, the encoding principle is similar to what AAC uses as VBR encoding, it is based on perceptual entropy, but more like CBR than VBR. When you select the ABR option in the VBR Settings box, the ABR edit box will be enabled. In this edit box you can specify the target average bit rate. Of course, a larger bit rate will yield generally better-sounding (but larger) MP3 files.

ABR Settings

VBR Quality Setting

This option allows you to set the Variable bit-rate option. Variable bit-rate encoding will enable dynamically determined bit-rates that depend on the music content of the current frame. This improves the overall quality of the encoded file without increasing the file size. This option sets the criteria used to determine when to increase the bit-rate for a frame. The lower the number, the lower the criteria will be. Thus VBR_0 will yield the best quality (but the largest file size) while VBR_9 will have less sound quality but the file size will be the smallest. The recommended variable bit-rate option is VBR_4. If you don’t want VBR encoding, set this option to None.

Thanks everyone so far for the replies!

Hardplay, I’m interested in the #1 setting, how come the ABR is set to 140? I though it was usually around 192 for the average bit rate of a high quality VBR?

Also, what is this paranoia mode in CDEX?

Thanks again :slight_smile:

how come the ABR is set to 140?

This is a setting that i usually use for my portable players, it produces decent sized files, but of course you can increase that value to what ever you want.

Consider that this is VBR-ABR, it doesn’t have much sense bringing the value to 192 kbs. Many users split the min. and max value, so that would be 160 kbps.

You can also try and give the max. option a greater value like 256, in this case 185 to 195 kbps would be a reasonable value as the averge fixed bitrate.

I though it was usually around 192 for the average bit rate of a high quality VBR?

ABR and VBR are not the same thing, VBR will scale the total range from 16 to 320 kbps. ABR (average bitrate) as it’s called will work inside a fixed minimum and maximum bitrate range. In the case of the first setting the user determines the lowest and the highest encoding setting and also indicates the average bitrate for the whole file.

Immagine a Vumeter with a 60db range, the needle will alway’s be lurking back and forth in the 30 db section.

what is this paranoia mode in CDEX?

Cdparanoia is a Compact Disc Digital Audio (CDDA) extraction tool, commonly known on the net as a ‘ripper’. The application is built on top of the Paranoia library, which is doing the real work (the Paranoia source is included in the cdparanoia source distribution). Like the original cdda2wav, cdparanoia package reads audio from the CDROM directly as data, with no analog step between, and writes the data to a file or pipe in WAV, AIFC or raw 16 bit linear PCM.

Cdparanoia is a bit different than most other CDDA extration tools. It contains few-to-no ‘extra’ features, concentrating only on the ripping process and knowing as much as possible about the hardware performing it. Cdparanoia will read correct, rock-solid audio data from inexpensive drives prone to misalignment, frame jitter and loss of streaming during atomic reads. Cdparanoia will also read and repair data from CDs that have been damaged in some way.

At the same time, however, cdparanoia turns out to be easy to use and administrate; It has no compile time configuration, happily autodetecting the CDROM, its type, its interface and other aspects of the ripping process at runtime.

hope that helps you :wink:

Hardplay you’ve been extremely helpful! I think I have one more question and it is: Joint Stereo or Stereo modes? I notice with the 3.97 codec it defaults to this in alot of settings. Should I keep it in stereo instead? What are the effects on quality, speed and speed of encoding?

Edit: Also, what format yields the smallest filesize and still is excellent quality?

good job hardplay. :iagree: :clap:

ET

next to Quality choose Alt Preset Standard. Done.

Yep, j-stereo is right.

Does checking off “Original” do anyting in terms of quality?

Joint Stereo or Stereo modes?

i have been encoding files since i was 3 years old…kidding, i’ve alway’s used “Stereo”, i’ve noticed that when re-encoding back to WAV from mp3, some frequencies sound really bad compared to the previous mp3 file, so that makes me think that Joint Stereo adds artifacts to the global structure of the sound, that’s why i manually configure it that way, my advice, stay away form anything that’s not STEREO !

Does checking off “Original” do anyting in terms of quality?

No, you can ignore it !

good job hardplay

eltranquil, thank you for the kind words :cool:

What are the effects on quality, speed and speed of encoding?

None at all for speed and speed of encoding !

For 99,9% of times Stereo is usually the original conformation, so why change it if you have the ability to keep that bit in your mp3 copy ! Also read the my previous post about this.

In your case you are striving to keep as much of the original data, so why throw away some of it ? :eek:

Edit: Also, what format yields the smallest filesize and still is excellent quality?

do you mean what other formats other than mp3, like Ogg Vorbis ?

If some of you is interested in a discussion between ‘Stereo’ and ‘Joint Stereo’ mode, here there is an interesting thread about that (it takes quite some time to read some of the posts, so you are warned. :bigsmile: ).

Regards, :smiley:

ET

No offense, but your settings are lousy. Really, I don’t enjoy having to put you down like this, because you’re obviously trying to be helpful, but you are wrong on so many counts.

You’re wrong, and you’re advocating something bad and really quite silly.

Anything that’s in an MP3 will be faithfully reproduced when properly decoded to WAV. Nothing is added nor removed from the sound when decoded to WAV.

So you’re experiencing “placebo effect”, or in other words you’re imagining things. Joint Stereo is not worse than Stereo, not in the way you describe.

There’s only one configuration on CD: Stereo. There’s no Joint Stereo or Mono. When mono recordings are reproduced on CD, it still has to be done in Stereo. Both channels will contain exactly the same signal.

Because the entire point of MP3 is that it changes / throws away information to give you something in the end that sounds the same but is much smaller.

If that’s your philosophy, then you should not be using MP3. You should be using a lossless format such as FLAC.

And FYI, Joint Stereo doesn’t throw anything away.
LAME’s Joint Stereo decides on a frame-by-frame basis whether to use LR Stereo (the “Stereo” you’re familiar with) or Mid-Side Stereo (a mathematical reinterpretation of the stereo signal).

No offense, but your settings are lousy

No offence, why should i be offended ?, it’s not my problem if you believe in what you say … and while i’m here would you be so kind to tell me and others, by what standards you state what you say ?,

but you are wrong on so many counts.

ok, convince me !

You’re wrong, and you’re advocating something bad and really quite silly.

is that your personal believing ? again convince me, with facts !

Nothing is added nor removed from the sound when decoded to WAV.

I invite you to read my post again, even twice if it can help you understand what i wrote!

There’s only one configuration on CD: Stereo. There’s no Joint Stereo or Mono. When mono recordings are reproduced on CD, it still has to be done in Stereo. Both channels will contain exactly the same signal.

this reply demonstrates that you are missing some important notes, and that you haven’t read my posts thoroughly !
can i ask you what does Joint Stereo have to do with CD’s ?

There are three Audio attributes in Professional recording:

01 - MONO
02 - STEREO
02 - DUAL MONO

i suppose nobody told you ?

Because the entire point of MP3 is that it changes / throws away information to give you something in the end that sounds the same but is much smaller.

thank you for enlighting me about mp3 tecnology, i really needed it !

If that’s your philosophy, then you should not be using MP3. You should be using a lossless format such as FLAC

Is this thread about a lossless format such as FLAC ?

Do you alway’s get to people this way ?

lol…it certainly does, but from what I gather, JS is better than regular stereo because it keeps the same quality as stereo, but uses it better?

Yes.

The gist of it is that Joint Stereo allows the encoder to save a few bits from time to time, while still maintaining the same quality of stereo imaging. These saved bits are used elsewhere for better quality.

Because if you are a “professional” as I presume, then it must be at the very least annoying to be shown you’re wrong like this.

  1. [li]Your ABR setting uses quality setting of q=0. q=0 had been “broken” up until about version 3.96. Nowadays (LAME 3.97b1) q=2 and q=3 are much faster and give about the same quality. As well, it sets a maximum bitrate of 224kbps. Unless something like this is absolutely required for compatibility with some crappy MP3 player, you should always use 320kbps as a maximum bitrate. This will increase the quality for difficult passages. Since we’re using ABR, we know damned straight that the average bitrate will still be the same no matter what max bitrate we set, so it’s best to set it as high as possible for those difficult-to-encode passages.[]Your VBR setting uses a quality of q=5. This is not good enough. Again, you want to be using q=2 or q=3 for the best results, because these offer the best possible quality, at a reasonable speed.[]Your CBR setting again uses q=0 when you might as well be using q=2 or q=3 for the same results at a much better speed. Also, you turned on “checksum”. This wastes bits that could be used for quality, instead using them to put CRC’s in the MP3 stream. Unless you intend to use this MP3 for streaming internet radio, the CRC is completely unnecessary.

[/li]

Okay.

That’s wrong. A WAV made from an MP3 will sound the same as the MP3, unless your decoder is very broken. Also your terminology is wrong. When you go from MP3 to WAV, you’re not “encoding”, you’re “decoding”.

In LAME, Joint Stereo is best for most situations. For those situations where Joint Stereo is not best, you would want to be using Mono or Dual Channel instead, depending on the material.

No. I understand things better than you. It seems to me that you are operating entirely on personal belief and cannot admit when you are wrong.

We’re not talking about CD’s! We’re talking about MP3!

Yes, and in CD’s there is only Stereo. There is no such thing as a mono audio-CD. A CD pressing of a Mono recording will still contain 2 channels, but they will be identical.
Technically, dual Mono and Stereo are the same thing as one another (2 channels). The difference is only in how you use/process them.

If you want me to fit it into that picture, then Joint Stereo is a part of the Stereo family.

You mentioned not wanting to throw anything away. So then you don’t want to use a “lossy” format like MP3. Which leaves you with only two options: Lossless encoding, or leaving the recordings in their native format.

I’ll ignore the sarcasm. :wink: You’re welcome.

You guys have been unbelieveable in your responces! I’ve learned a great deal in such a short time and I feel much more comfortable moving towards my archiving :slight_smile:

I think I have one more question. Should I use the ABR or VBR pattern? Space shouldn’t be an issue, I just want them to sound good and make use of the space.

Thanks so much again! :cool: