Log in

View Full Version : How do you encode your cd audio rips?


tim
2004-05-28, 01:55
________________________________
move along nothing to see here folks!

eclectica
2004-05-31, 08:55
I use CDex (http://www.cdex.n3.net/) to create wav files from my audio CDs. Then I use RazorLame (http://www.dors.de/razorlame/index.php) to encode my mp3s between 128 to 256 bitrate. I make the mp3s CBR and joint stereo.

eclectica
2004-06-05, 19:18
I find that most mp3s are either 128 bitrate and then 192 bitrate, though there are few at 160 bitrate. It goes against what I would expect statistically. There is no bell curve in the statistical distribution of the bitrates or otherwise 160 kbps would be the most popular.

I think of 256 as the platinum standard, 192 as the gold standard, 160 as the silver standard, and 128 as the bronze standard. So what is 224 then?
:confused:

I don't like VBR mp3s because I feel that there are problems with compatibility and they do not play well.

-GS-
2004-06-16, 00:25
The Quintessential Guide To Creating The Highest Quality MP3s On Earth

(original document was posted on zeropaid.com)


The Life Cycle Of An MP3

A Metallica CD is inserted into a CD-ROM drive.

The ripping software is started from the desktop. (What's ripping software? A ripper is a piece of software that lets you extract audio data from a CD)

The CD-ROM drive starts reading the CD and the ripping software saves the extracted audio as pure, uncompressed .WAVs on your computer.

STOP! Ok, at this point your audio is a virgin. It's pure. It's untouched. It's a bit-for-bit digital copy of the audio that is on your CD, and is now saved on your computer in .WAV format. The .WAV files are referred to as being uncompressed, because they're huge files.

Why don't we all trade .WAV files? The only thing stopping people from trading freshly ripped .WAV files is the fact that they're damn huge. An average song in .WAV format is about 40 megabytes, which is about 10X as large as the average song in .MP3 format.

Trading .WAVs wouldn't be any fun because they'd take 10X as long to download, and your hard drive would fill up 10X as fast.

I Think I Know Where You're Going With This...

You're such a keener. Yes, the whole idea behind the .MP3 format is to take those uncompressed 40 megabyte .WAVs and compress (squish) them down to 4 megabyte .MP3s.

Can you guess what this process is called? Anyone? Yes, it's referred to as compression. You will also hear the term encoding associated with this process. Encoding is actually a better name for the process, it refers to the actual process of the .WAV being turned into a .MP3. The result? Compression!

Quick Recap Of The MP3 Lifecycle:
Audio CD is placed in CD-ROM drive
Ripping software extracts audio from the CD in uncompressed .WAV format
An encoder takes that .WAV and compresses it to .MP3 format so you can trade it with your friends on Kazaa Lite

You'll recall, we had a virgin .WAV file, which was rather large, and along comes this MP3 encoder. And the encoder compresses the .WAV into a relatively small .MP3 file, perfect for trading. And yet it's the same music? How can this be?

You see, our pathetic human ears don't NEED all the audio data that is in the large .WAV file in order for our brains to decide "Hey, I like this song man......". In fact, this is what the MP3 compression format is banking on.

MP3 encoders know that as a pathetic human you won't miss really high frequencies that you can't hear. So what do they do? They cut 'em out, and serve you up an MP3 that to YOU sounds just as good as the CD. There is a little more to it than this. The decisions by the encoder on what to keep and what to throw away are referred to as the psycho-acoustic model of the MP3 encoder. Yes, go on, impress people with that word. The psycho-acoustic model is really what makes one encoder good, and another one crap. For instance, the Xing encoder is terrible because it has a horrible psycho-acoustic model. Which encoder is the best? Keep reading.

This type of compression is known as lossy. Why is it called that? Because you lost something, audio data. Yes, it really is that simple. You know when you use WinZip to compress something to a .ZIP file? At any time you can come back and UnZip that file, and everything will be inside it just as it was, because that's lossless. When converting from WAV to MP3 you lose something, and you DO NOT recover it when converting back from MP3 to WAV. It's gone forever. Lossy. Bye bye audio data.

Quick Recap Of MP3 Encoding:
Your ears suck, MP3 encoders take advantage of that.
Xing is a horrible MP3 encoder.
MP3 is a lossy compression format.

Ok, Got It, So What's The Best Software?

All rippers and all encoders are not created equally. Whether you're aware of it or not, the ripping software you're using right now to make MP3s is using one of the following MP3 encoders (in order of quality):
Xing (worst MP3 encoder of all time)
Blade
Fraunhofer
LAME (best)

One thing you have to remember is that an MP3 is only as good as the WAV it came from. And where do WAVs come from? They are ripped from CDs. And is all ripping software created equally? It sure as hell isn't!

Have you ever heard a 'crackle', or 'pop', or a short 'blip' in an MP3 you were listening to? This was due to the fact that someone used bad ripping software to rip the .WAV file. That .WAV file was compressed to .MP3, and the defect was still present.

It's like the old saying goes, garbage in, garbage out.

I can tell you unequivocally, absolutely, and without hestitation that there is only ONE piece of ripping software that a serious MP3 creator would ever consider. It's name is Exact Audio Copy (http://www.exactaudiocopy.de) and it's freeware!

Why Only Exact Audio Copy? Why Is [Insert Your Favorite Ripping Software Here] Inferior?

EAC is the only software that offers an option known as 'secure mode'. What is secure mode? In layman's terms it means that your drive reads each section of the CD at least twice. If there are any differences, EAC will attempt to re-read those sections until it finds a match.

Why is this good? Because it means you are assured a 100% perfect digital audio extraction of your CD (in .WAV format naturally). It also means if you've got some CDs that have been scratched, EAC will re-read the scratched section of the CD until it is confident the rip is perfect.

As I mentioned above, there is no reason to bother using a good MP3 encoder on a WAV that is not a perfect copy of the audio on the CD. There is no excuse not to use EAC, it's free.

And no, no other ripper is as good as EAC. Trust me. This is the one you want to use.

Alright, So What's The Best MP3 Encoder?

Oh there's so many choices aren't there? Well, it would be more accurate to say there's many options, but really only one choice you should consider.

You probably already know what a bit-rate is, most of the MP3s you listen to are probably at a bit-rate of 128kb/s, 160kb/s, or 192kb/s. You probably also know that the higher a bit-rate goes, the more disk space you are eating up. (Once again, you see how you don't get something for nothing).

So what's the best bit-rate? Well hold on there, not so fast. We've only talked about constant bit-rates so far. When you have an MP3 that plays at a constant bit-rate of 160kb/s throughout the song, you can say "This MP3 was encoded at 160kb/s CBR." (CBR is an acronym for constant bit-rate).

But what if the bit-rate could fluctuate throughout the song? Perhaps during complex parts of the song it could rise way up, all the way to 320kb/s, to retain all the quality of the original CD sound. And then during 'simple' sections it could lower back down to save disk space. Wouldn't that be the ultimate MP3?

Well luckily other's thought that would be ideal as well, and that's the whole basis of 'Variable Bit Rate' (VBR) encoding.

Let's say Johnny encodes a Metallica song using a constant bit-rate of 160kb/s, and the size of his MP3 turns out to be 4 megabytes.

Along comes Susie, and she decideds to encode the same song, using variable bit-rate. During a heavy guitar riff the bit-rate zooms up to 320kb/s, because that can be difficult for an encoder to encode properly at lower bit-rates. Then during some quiet vocal sections, the bit-rate drops down to 128bk/s, because you don't need a very high bit-rate to maintain the quality of simple vocals.

It turns out Susie ends up with a file that is also 4 megabytes, so who's MP3 is better? Why Susie's of course! She's managed to maintain all of the quality throughout the song. During the heavy guitar riff's, Johnny might find that 160kb/s wasn't enough, and if he listens carefully it won't be as 'crisp' or as pure as the original CD audio.

Continue on the next post....

-GS-
2004-06-16, 00:27
You may have heard that VBR MP3s had some problems, and a long time ago that was true. However, VBR is now accepted as the best way to encode, and there is only ONE option you need to consider, LAME (http://mitiok.cjb.net).

Yes, that's right, the highest quality MP3 encoder you can get is called LAME, and it's free. Maybe you've heard of LAME, maybe you haven't, but either way it's what you'll want to use!

So What's The Best Setting To Use With LAME?

LAME can be scary. It has so many options, and most people who use it have no idea what half of them do. Luckily there is a group of individuals who have simplified it all for us. There is a community of audiophiles at Audio-Illumination.org who have played around with, tweaked, and tested all the settings more often than most of us ever will in 3 lifetimes.

They also came up with a switch (also known as a command line) that LAME recognizes. When you use this switch you are creating MP3s that sound just as good as the original CD, even to an audiophile using equipment that costs as much as your next car.

So what is this magical switch?

--alt-preset standard

Sound it out. Dash Dash Alt Dash Preset Space Standard.

To LAME, this simple command line represents thousands of hours worth of work by audio freaks, so respect it. :) What this switch does is create an unbelievably high quality MP3, using variable bit rate. Most music you encode with this switch will end up at an average of about 190-200kb/s.

Don't worry if you have no clue how to use LAME, or command line switches, or anything else, as it will all be done for you. LAME integrates nicely with EAC, so it feels like you're using an 'all-in-one' package.

OK, Best Ripper EAC, Best Encoder LAME. Time To Get Started.

Both EAC and LAME can be scary for some people. They were for me too. They each have so many options, and although EAC is getting better, it's still not as user-friendly as say, MusicMatch Jukebox.

However, from my years of experience I've learned all the ropes of each, and I've created a process to help people get up and running within just minutes, using the best possible options.

This way, everyone can create MP3s with the confidence they're doing it correctly, and then learn what each of the options in EAC does as they go along.

--------------------------------------------------------------

Anyhoo, thats a bit old post i platantly quoted form another forum, but,
it shows great deal of info about making mp3's.
A Good Start atleast, maybe bit too much on the freaky side?

:blowjob: ( i just luv teh smileys here)

Well, that EAC dont really like to rip my Old and Used CD's.
It just dont work, gets read errors etc.

I think what is important is the encoder itself.
What ever app i use to rip, i try to make it to use LAME encoder.

Im currently using dbpoweramp (http://www.dbpoweramp.com/) with Lame.
Why? Because its Free and has shitload of support for different formats,
just check out the downloads sections.

Maybe it isnt the best, but it works.

I do my rips atleast 192k because i just think thats good enough for my
gear what i use for listening.

I dont like 128 cuz i dont like the way it sounds, too much audio is missing,
especially in the High End area.


Havent been on here too much lately as logged cuz i forgot my password,
but i remember it today. Been lurking on the background thought...

yay


:jerkoff:

Oh yeah, had to cut it to 2 posts, didnt fit on 1. Sorry about that, but its teh Board Software.

eclectica
2004-06-16, 12:41
Music which has high notes is affected most by encoding. mp3 encoding does the same thing that DSL does, by cutting out the high frequencies. With DSL the phone can be used along with the internet. The way it works is that a filter allows you to speak on the low frequencies, while the data travels on the high frequencies. The low frequencies are used for normal hearing while the high frequencies have the smallest wavelength and contain the most data.

If you have a song with high notes then you need those higher bitrates. I've run tests myself using glass harmonica music, which is shrill with high notes. I've found that there is a slight difference that is very hard to hear between 192 and 256 kbps. I think anything higher than 256 is unnecessary.

A song that is compressed too much will have a watery sound that is similar to the sound of a cell phone. The watery sound of a cell phone is due to the provider compressing your signal in digital mode and thus sounding like an overly compressed mp3. Regular phone lines have a bandwidth capability of about 64 kbps, but cell phones are compressed and take up something like 10 kbps, giving a poorer sound. A cell phone sounds better in analog than digital mode, because it uses more bandwidth, but it is insecure and it uses more battery power that way.

-GS-
2004-06-19, 03:13
Also, we humans lose hearing on the higher frequencies the older we get.

So, if a person is over 60 y.o. does 128k mp3 be just about good enough?

Im sure there is somewhere a study going on about it, he he...


I dont know about my hearing but it cant be 100% as
i've played too loud too much over the years, teh guitar.
I've learned to use earplugs now, thought.
And, im not a teenager anymore.