Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Music Books Media Book Reviews

MP3: The Definitive Guide 97

It may have taken a semi-obscure German professor, the profit motive of the world's CD-ROM drive producers, ingenious hackers and aesthetically gifted interface designers, but the simple fact is that MP3s are out there, and they're everywhere -- every major desktop OS comes with players for your easy listening. Chronic book review madman chromatic points you to a fount of knowledge for anyone who needs more than "click here to play."

MP3: The Definitive Guide
author Scott Hacker
pages 388
publisher O'Reilly & Associates
rating 8.5
reviewer chromatic
ISBN 1-56592-661-7
summary Everything there is to know about MP3, as it stands right now and might be in the near future.

*

The Scoop

Judging by the way even my non-technical friends are talking about MP3s, digital music is on a lot of minds. As usual, O'Reilly has published the definitive guide to all things MP3. Computer and music guru Scott Hacker takes you through the codec, the software, the controversies, the competition, and building your own equipment. Though it's aimed at end-users, the book is still accessible to the do-it-yourself weekend hardware wizard.

What's to Like?

Hacker's writing is simple and not-too-technical. In places, it's even informal. Sure, there are plenty of gory details, but you won't miss anything essential if you skip over the sidebars now and then. An average computer user could probably create his own MP3s while reading chapter five, for example. Power users aren't left out, though: Audiophiles, hackers and tweakers will benefit from the extensive comparisons of players, encoders, hardware, and competing codecs.

No stranger to alternative operating systems (he also wrote the BeOS Bible), Hacker takes pains to be cross-platform, covering Windows, Mac, Linux, and BeOS. This isn't limited to playback options, though that's the most extensive discussion, but includes serving files over the Internet. Of special consideration are quality issues. The author's perspective as a sound connoisseur comes in handy while discussing optimal (and affordable) recording and playback equipment.

As per the title, the Guide completely covers the subject. If you're interested in collecting MP3s, creating them, playing them back with software, with portable hardware, car hardware, building your own hardware, making music available to others, discovering alternate means of delivery and other codecs, or just want a broad overview of all things MP3, you'll find something of immediate interest. If Hacker whets your appetite for more information, follow one of his references to the source itself. (That's especially nice in his treatment of the more exotic hardware players.)

What's to Consider?

Though the chapter on legal information and MP3 is excellent, and among the most extensive treatments of the issue lay readers are likely to encounter, it's U.S. Centric. Also, it should be noted that the digital music debate is undeniably fuzzy, so any interpretations are open to correction. Though he debunks the common disclaimers found on shady MP3 sites, the author wisely sidesteps copyright arguments by explaining the relevant laws, and allowing his readers to come to their own conclusions in the gray areas.

People who've been tracking the scene for a while know how fast things change. Information on specific programs or hardware players could become obsolete quickly. (That's noted in the text.) For the most part, Hacker prefers to explain concepts and trends rather than the fine details of any particular implementation. For items still unresolved, such as the eventually supported ID3v2 specification, he provides caveats regarding compatibility issues.

The Summary

Catch up to the digital music revolution with MP3: The Definitive Guide. It's packed with information, yet easy to read, and stuffed with links to satisfy your appetite for up-to-the-second information.


Purchase this book at ThinkGeek.

Table of Contents

  1. The Nuts and Bolts of MP3
  2. How MP3 Works: Inside the Codec
  3. Getting and Playing MP3 Files
  4. Playlists, Tags, and Skins: MP3 Options
  5. Ripping and Encoding: Creating MP3 Files
  6. Hardware, Portables, Home Stereos, and Kits
  7. The Not-So-Fine-Print: Legal Bits and Pieces
  8. Webcasting and Servers: Internet Distribution
  9. Competing Codecs and Other File Formats
  1. ID3v1 Genres
This discussion has been archived. No new comments can be posted.

MP3: The Definitive Guide

Comments Filter:
  • Mp3 is old hat...soon Mp4 or some other standard will take over and the book will be out of date.
  • If you looked at the facts, you'd have noticed that Napster has violated the Sherman Antitrust Act. Maybe you should think before you support such fragrant violators of U.S. law.
  • I think MP3 as a file format may slowly die out, but not because of its borderline legality or the current playback restrictions (ie. primarily only on a PC).

    No, I think compressed digital music as a whole will slowly take over the industry, with the exception of few "old-format" CDs that will still be printed.

    After a standard for packing encoded audio onto CDs is established, Sony and Co. will start supporting it on their home stereo systems, integrated with your CD player.

    These home players will be able to tell the difference between a CD-Audio and CD-EA (Encoded Audio), and will be able to tell the difference between MP3 and other encoding standards.

    With all this technology for digital music around, the recording industry will have to convert over to encoded music or be lost. You'll be able to purchase the songs you want individually at a reduced price -- either via the Internet, where you download the music, or at a music store, where your selections are burned in a matter of minutes in the format of your choice using quick, high-quality audio encoders and ultra-fast CD-R drives.

    How's that for a future?

  • by epeus ( 84683 )
    Chapter 3 mentions skins which is weird as they have nothing to do with MP3 per se, they're just an annoying trend in UI design that some MP3 players have adopted.
  • Well, looking at my bookshelf, there's a gnu on the cover of Learning GNU Emacs (the connection being obvious) and an appaloosa on the cover of Apache: The Definitive Guide (being either the native American connection or just the similarity of the name). No exact matches, but pretty close. Now, if they really wanted to do some appropriate book titles, they'd get these bugs off the Oracle book covers and put 'em on their Red Hat Linux books. :)

    Cheers,
    ZicoKnows@hotmail.com

  • When a company smells, I consider that a violation of my 8th Amendment rights.
  • If we keep annoying them, soon enough they'll bein Chapter 11 as well.
  • I'm in the process of converting my own 150+ cd library to 192kbps MP3, so I can push the "shuffle" button in Winamp and get a random song from my FULL library.

    Why? Wouldn't that be like converting all of your DVDs to digital files on your computer, so that you could hit "shuffle" and get a random movie scene from your FULL library? That would be crazy. Or getting all of your books in some e-book format and hitting "shuffle" to get a random 20-page excerpt?

  • I prefer Terry Pratchetts concept of 'Hermit Elephants' (from the Discworld Companion).

    "These poor creatures lack the thick skins of normal Elephants, and use abandoned mud huts to provide camoflage and protection instead. They have not problem obtaining abandoned mud huts since the previous occupants move out very rapidly when a hermit Elephant moves in."

    This is accompanied by a hysterical picture of a mud hut with two mad eye's and a trunk peering out the window.

    EZ
    -'Press Ctrl + Alt + Delete to log on..'
  • What a load of garbage.

    If Sony had its way no matter what, we'd all be listening to MiniDisc players and popping Memory Sticks into our systems.

    MiniDiscs, despite billions of dollars of promotions, remain firmly entrenched in the theatrical design ghetto.

    Do you really believe that the Memory Stick has a future?

    wug
  • So...Informative. Hmmm. The information presented (that isn't Redundant) is:

    Overall, Scot is a cool guy.

    Nice.

  • I doubt it, but here's a little test we could run to find out.

    1. Obtain a pile of money. Good methods include crime and inheritance.

    2. Run around pulling at your hair screaming, "I HATE THE .BMP FORMAT!!" I'd just like to point out that I do hate the .BMP format. My friend and I were arguing about this the other day. I said, "Any modern computer will have no problem decompressing a .GIF or .JPG on the fly." He replied, "Yeah, but drive space is hardly an issue anymore." "Yes", I retorted, "But bandwidth is - and not everyone has direct access to a T1." At this point he hit me.

    3. Where was I? Oh, yeah, lobby Congress to try to stamp out the .BMP format in its entirety. Print T-Shirts that say "Down with .BMP!" Threaten foreign countries who use .BMPs. Try to buy the .BMP format outright and claim that you invented it.

    Will these steps cut down on the number of .BMPs on the internet? Maybe. But for every ten you get rid of, six slashdotters will embed them in their comments using MIME just to piss you off.

    In conclusion, .BMP sucks, and I like Mountain Dew.
  • ... has got to be "porn".
  • The blurb for this review says "every major desktop OS." Then how come I still can't find an MP3 player for my Windows 3.1 machine (whaddaya mean win 3.1 isn't major?)? WinPlay3 refused to play my mp3 files, saying they were an unsupported format. Every other player I've found is for win 9x. Anyone have any suggestions?
    ===
    -J
  • Signal 11 has posted 0 comments (this only counts the last few weeks)

    That's weird. I could have sworn I posted hostile responses to Signal 11's crappy comments at least ten times or so in the last few weeks.
  • "Too" lossy is a matter of individual taste.

    How much you'll accept is a matter of taste, but the fact that loss is bad isn't. Something that can lose less information but take up the same amount of disk space is just better, regardless of your taste.

    The bus came by and I got on
    That's when it all began
    There was cowboy Neal
    At the wheel
    Of a bus to never-ever land

  • I'd just like to point out that you are the lamest, whiniest little bitch in history.

    You may have a point about VQF, but you don't actually give any information on VQF. You appear to be name-dropping to look cool. Most of us probably either A) are hearing about VQF for the first time; or B) have sort of heard of VQF once in passing. Instead of acting ridiculously elitist, why don't you make yourself useful and DESCRIBE the VQF format and its comparative advantages?

    Also, MP3 does *NOT* sound like an old, scratchy 78. It's obviously not quite CD format (because it's compressed), but in a blind "Pepsi challenge" type test through headphones, I could probably not distinguish the two.

    I did notice you posted something pretty awesome earlier, although I forget what it was, so I'll give you props for that, but in terms of this last post, you suck.
  • Is it any good? Anyway, for those (like myself) still listening to MP3s you will not find finer than those of EleMenT [mp3.com] . There are six tracks to download and I can GUARANTEE you will not here another band on the net that are anywhere near as diverse and original as these guys!
  • by jd ( 1658 ) <imipak@ y a hoo.com> on Thursday June 22, 2000 @08:39AM (#983835) Homepage Journal
    The websites I can suggest are:

    There may be more websites with useful information out there. I'm collecting all the MP4 code I can, to see what the specs, in reality, mean. Oh, and to see if there's anything worth filching.

    Video, I'm not sure. Again, MPEG-4's video layer is said to be very good, but I'd have to see some videos in that format to be convinced. QT4 is OK, but there are no (en|de)coders for it for Linux, which limits it a bit.

  • I agree. How about an encoder which asks the listener to sit some tests when first used so that it can optimise the psychoacoustic model for that listener?
  • If it takes someone the teeniest bit of effort to type in "http://www.vqf.com" to go to their web site, I am NOT going to be terribly sympathetic.

    I've posted a list of websites which document the various standards. I mean, mpeg-4.com for mpeg-4 - how obscure can you get? Really.

    As for me being whiny... I can see that someone who fussed and moaned, but did nothing, might deserve the label. Difference is, I'm ripping apart over a dozen MPEG-4 audio and video packages to see how they work, to write an encoder/decoder for the Free Film Project. A high-quality encoder is essential for such a project, for distribution.

    Talking of people complaining and whining, what are YOU doing for the Free Software community, today? Besides making use of it?

  • Thank you, everyone, for your support. The Album is an artform that will never die.
  • Surrounding the format as a whole? Come off it. Most of the legal trouble is about copyrights, not the format itself.

    Sorry, most of the legal trouble is about PATENTS, not copyrights. Ie, even if you develop the MP3 encoder totaly independently, you may be forbidden to use and copy it. If you are lucky, Franhauffer may allow you to use it for a price.

    It is as bad as with .GIFs...
  • MP3 is old, too lossy, sounds bad, and has way too many legal messes surrounding it. I'm sure that eventually all of these can be worked around (MP4?), but why bother? Vorbis is here, and is free (speech) and free (beer). Oh, and It's Just Better (tm), too.

    Great! Where can I download free upgrade for my Rio to play Vorbis ? What do you mean I can't ?

    Backward compatibility is BIG factor. Not only there are tons of mp3 already around, but there are plenty of hardware for it.

    While I eventually could be bothered to reencode hundred or so CDs I own (to get somewhat better quality and get rid of patents and other issues), I will not do it unless I can upgrade/cheaply replace expensive (it was and still is for me - but I just love it, so it was worth it) hardware that does MP3s just fine.
  • he's a *cough*"good hacker"*cough* --

    Hacking cough?
  • Know one thing tho - without having to even see this book - that it doesn't mestion all decoders that there are for mp3.

    Reason is that I have it released the Black Omega MP3 decoder engine yet. It still going through its alpha testing stage - the Windows version works and still got to put in the audio object for the Linux version.

    Also still want to do several major changes to it.

    1. Recode the Windows audio interface object and then from that develop the Linux audio interface object. Such that it will process MP3 frames in groups in larger groups in one process run rather than using wait conditions causing a large overhead in process switching.

    2. Increase the fault taleraunce on the decoder such that if headers are wrongly positioned our header information is wrong then it can reseek and find the next header.

    3. Improve the API so that full playback control is integrated into it.

    4. Ring out more bugs from it and get MPEG-2 LSF extension parts working. Well namely only got get the LSF scalefactor part debugged. MPEG-1 playbacks near perfect now.

    5. Do much optimization. Get it working then optimize once it works is what me say.

    Anyway me going to be moving house shortly and my new flatmate and me going to get cable modem permenant connection so once got that I will open up a CVS tree on it and do initial early alpha releases onto freshmeat.

  • Actually, CDs were invented by Phillips...
  • If it explains a way to insert VBR headers into an MP3 that was created without 'em, I'm there, dude.

    Most annoying thing I've seen - only once or twice out of $BIGNUM MP3s - was an MP3 that wouldn't report its time correctly in WinAMP.

    I'm not talking about "normal" VBRs, which are pretty close to accurate on the track length, and which show the bitrate changing from 128/160/224/226/whatever as the song plays.

    I'm talking about a VBR MP3 where the track length is totally misreported. As in, it varies by 20-30% depending on where in the song you happen to be playing at the moment, and the bitrate indicator shows an oddball number like 147kbps.

    I believe that this problem is due to someone encoding in LAME with the -t (disable VBR tag) flag set. But I'm damned if I can figure out how to re-create said tag given only the final MP3.

    Any suggestions? It's the weirdest thing I've ever seen.

  • WHY??? Why waste time with MP3's when VQF is so much better? Why bother with a standard that's all but dead, now that we have MPEG-4?

    Is this some massochistic ritual thing that I missed out on, flooding a hard drive with oversize, low-quality audio? And I bet these are the same people who did stuff on Geek Pride Day, too. HOW can you have pride in using obsolete, badly-lossy, disk-hogging formats?

    IMHO, real pride comes from boldly moving forwards and adopting newer, more powerful audio formats, not stagnating. Sure, it may irritate the RIAA, but that's all it does. Irritate. It's not as if they really care -that- much about something that sounds like an old, scratchy 78.

    Again, if fighting the RIAA or MPAA was a serious goal, you'd use the best compression around, not toy-stuff.

    There's more to audio than tin cans and string, but you won't find it at mp3.*

  • by sander123 ( 120105 ) on Thursday June 22, 2000 @06:06AM (#983847)
    You can read chapter 6 of this book here: http://www.oreilly.com/catalog/mp3/chapter/ch06.ht ml

  • Compact Cassettes were invented by Philips, too.
  • by kcarnold ( 99900 ) on Thursday June 22, 2000 @07:08AM (#983849)
    <rant>

    MP3 is old, too lossy, sounds bad, and has way too many legal messes surrounding it. I'm sure that eventually all of these can be worked around (MP4?), but why bother? Vorbis [vorbis.com] is here, and is free (speech) and free (beer). Oh, and It's Just Better (tm), too.

    </rant>

    Remember the SAT? Windows is to Linux as MP3 is to Vorbis.

  • are you some kind of flaming wierdo who plays an album from begging to end?
    You have such primitive musical tastes and are very accepting of prepackaged predigested content.

    its better to take individual songs you like, and make your own playlist. personally i mix songs from all different genre's into a single playlist.
    it balances the spheres.
  • by Anonymous Coward
    >Reason is that I have it released the Black Omega MP3 decoder engine yet.

    Well, bully for you. Please let us know WHEN YOU HAVE IT DONE. The book was complete when it was published. What, you want them to wait for you? Gimmee a break. Please work on your grammar, too.

    Thank you,
    the caffeine-deprived

  • The Shorten format(.SHN extension), though larger than MP3, is lossless.

    O'Reilly has posted a sample chapter [oreilly.com] of Chapter 2, "Inside The Codec" of the MP3 format.
    http://www.oreilly.com/catalog/mp3/chapter/ch02.ht ml
  • I think if you were actually worried about quality you would be listening to a CD instead, and eagerly awaiting whatever the next-gen super CD thingy is going to be.

    A good reason to get excited about MP3 is the momentum and wide acceptance it has. If you want to get music from other people, you need a common format. VQF hasn't been as widely ported as MP3 either. Unless you are an experienced geek, with access to a Windows machine (surely that's a self-excusionary situation?) you won't get it running on Linux or FreeBSD (via an XMMS plugin that uses WINE to call the VQF DLL).

    Four years ago my college friends and I were really excited by MP3, which sounded pretty good through our low-rent stereos. Today, the only one of us still interested lives overseas and uses it because he can't locally buy the music he likes. The rest of us can't tolerate the quality.

  • Actually, it was a collaboration between Sony and Philips. Hence digital outputs (like a CD/DVD-ROM's CD digital output) are often called S/PDIFs, for Sony/Philips Digital InterFace.
  • Someone's playing games with slashdot's name parsing code. I've already spoken with Cowboy Neal about this and it's a bug in the string-to-database parsing, which is how this guy [slashdot.org] has managed to overwrite the namespace for my name.

    They're working on it though, don't worry.. you'll be back to flaming the original me in no time. (And sorry about forcing the +2 bonus, as I can't login, I can't control the default score)..

  • Oh yeah, big shot? Let's see a link to this supposed encoder-decoder.

    Oh, what? It's not done? You're barely in the conceptual phase? Well, as it turns out, I'm working on a new OS, which I'm going to be calling GNAGA (GNAGA's Not A Good Acronym). It is currently better than your lame-ass encoder, because it has a name.

    Incidentally, how is typing "vi" equivalent to "ripping apart"? Or are you literally looking at the binaries and trying to translate them into source code?

    Bitch!
  • :Well, looking at my bookshelf, there's a gnu on the cover of Learning GNU Emacs (the connection being obvious) and an appaloosa on the cover of Apache: The Definitive Guide (being either the native American connection or just the similarity of the name). No exact matches, but pretty close. Now, if they really wanted to do some appropriate book titles, they'd get these bugs off the Oracle book covers and put 'em on their Red Hat Linux books. :)

    Being a native American, i.e. one born on the continent of America, therefore, native to America, I resent the term native American being used to represent American Indians. Why does the English language have to be corrupted by people? Native Americans are those native to America, I was born in America and by definition am a Native American.
  • Being a native American, i.e. one born on the continent of America, therefore, native to America, I resent the term native American being used to represent American Indians.

    I am a Native American by your definition, but not by the accepted definition (American Indian). But don't you think that Native Americans (in the parlance of our times) might not like to be called American Indians since they really have nothing to do with India?

  • I'm a mp3.com Artist [mp3.com] known as Nemesis 404, We make club/dance music for clubs, works out, but my problem or thought is, that is MP3 the best way to go? I think so, and everything points to it. So why not!
  • by Dungeon Dweller ( 134014 ) on Thursday June 22, 2000 @05:34AM (#983860)
    Doesn't look to bad, but it would be better (in my opinion) if it covered the algorithms used in mp3. They are the most interesting part!
  • by Farq Fenderson ( 135583 ) on Thursday June 22, 2000 @05:38AM (#983861) Homepage
    You know, people have expressed distaste about the seemingly arbitrary posting policies -without- being immature about it. The ones I saw weren't moderated down.

    So, how's puberty going?

    ---
    script-fu: hash bang slash bin bash
  • MP3's are undergoing an assault, led by the lawyers from the RIAA. With Naspter struggling to stay alive, do you think MP3's will go the way of the dodo bird? What happens if the RIAA pressures the makers of MP3 devices to stop making them? I know Gnutella, and other services like it, will survive, but if we have nowhere to play this things, outside of your computer, won't the viability of MP3's be diminished?
  • Scott Hacker... I'm envious. :)

    At least he's a *cough*"good hacker"*cough* -- I wonder if that's what we have to do to get the mass media to realize that hacking is a good thing -- accept defeat on the Hacker v. Cracker issue, and make a point about "good" vs. "bad" "hacking"...

    --

  • by Signal 11 ( 7608 ) on Thursday June 22, 2000 @05:35AM (#983864)
    I hear Chapter 10 didn't make it:

    10. Annoying the RIAA for fun and profit.

  • I like O'Reilly books a lot, but I really can't see how they can write a whole book on MP3s, unless there is a rediculous amount of political insight in what ought to be a computer book.
  • by gavinhall ( 33 ) on Thursday June 22, 2000 @05:35AM (#983866)
    Posted by 11223:

    First of all, I'd like to point out that Scot Hacker is the ultamite BeOS evangelist, who also wrote the BeOS Bible and runs the BeTips [betips.net] site.

    On a more relevant note, the mp3asm tool [tu-clausthal.de] availble from the MPG123 web site is a really neat tool to look at the source to - it'll show you the specs of the MP3 format right there. If you're at all interested in the down-and-dirty of dealing with MP3 format files, that source is a great starting point.

  • Yeah. Do you want to know why? Sony is the largest player in the music industry. They have more recording artists that anyone else. Who invented cassettes? CDs? Minidiscs? Sony. Does this shock me? No. Did Sony invent the MP3 format? No. If you start putting things together, Sony founds and pushes RIAA to attack the MP3 stating that it has security holes (which it does) while they propose a solution that they can collect money on. So yes, MP3s will perish only because Sony has more money that everyone in Milwaukee. ****** The above opinions are my own. Britny Spears may really be a robot with a VAIO computer as a brain. ******
  • Looks a little simplistic, but chapter 2 looks like fun :)
  • Comment removed based on user account deletion
  • by Submarine ( 12319 ) on Thursday June 22, 2000 @05:44AM (#983870) Homepage

    I browsed this book last month and found it informative and quite well-written.

    The author explains the thorny issues of patents pertaining to MP3. Two corporations have patents on MP3 technologies:

    However, as it has been pointed out, the law section is US-centric. It would be interesting to know where in the world the said patents apply. Theoretically, Europe does not allow patents on algorithms and mathematical formulas, but this provision can be circumvented by patenting an overall technique using an algorithm.

  • by Anonymous Coward
    This anonymous Coward thinks one of the best places to look for information on coding schemes and algorithms is going to be in IEEE confrence papers. There is actually a good bit of information on audio masking effects and the real theory behind compression.
    On a more basic level, think of MP3 as a format that takes a recording with everything, then takes out what you can't hear as a human being with ears and at least one cochlea. The remainder (we have found) is much less than we thought it'd be.
  • People! Good citizens of slashdot! It's time that we stopped allowing the wretched Music Industry to sue the good people of Napster. There's only one solution: Let's get rid of democracy and replace it with literature from a Burger King calorie chart. It's time to own up to the fact that it simply has not worked. Napster should be suing the Music Industry for having to trade such bad Britney Spears songs.

    Sincerely,
    The Red Menace (aka Ralph Nader)
    LostBrain [lostbrain.com]

    Here's my wife [lostbrain.com], where's yours?

  • *grin*

    O'Reilly generally doesn't bother matching the animals much to the subject on their books. Generally....(Ie, the Python Pocket Reference has a snake on it, but none of their other Python books.)

  • by Jon Erikson ( 198204 ) on Thursday June 22, 2000 @06:13AM (#983874)

    Try www.mp3-tech.org [mp3-tech.org] or www.mpeg.org [mpeg.org] for more technical information.


    ---
    Jon E. Erikson
  • ...be its legality? MP3 is a nice compressed audio format that is well suited to personal libraries. I'm in the process of converting my own 150+ cd library to 192kbps MP3, so I can push the "shuffle" button in Winamp and get a random song from my FULL library. MP3 is practical, and won't die until someone thinks of something that is smaller and sounds better.
  • While the RIAA may be able to scare small companies, but I think that larger companies will escape their wrath.
  • by emil ( 695 )

    AFAIK, the mp3 encoding technique relies upon the Discrete Cosine Transform (DCT), which is also used in JPEG.

    An implementation of the algorithm can be found in the source code for mpg123. This is available in SRPMS in any Red Hat mirror.

  • I know this comes off sounding like a newbie question (which I am). But can you suggest websites that have news on compression and new file formats? Or sites where I can buy encoders for Mp4, VQF, etc.

    While were "on" the subject, what is the best video decoder?

  • Shame on you, O'Reilly.

    Shame on O'Reilly? What is their purpose for being in business if not to sell books? I don't see the big deal. I have read books about Math which did not get into every gory detail about equations. Stephen Hawking's most popular books have almost no equations. That's not the point. Someone else can put out the technical manual.

  • I got the impression that it was merely a review of different implementations. Sounds good to me.
  • So I would guess that O'Reilly is targeting a more sophisticated audience than the 'for dummies' book is. I have NO idea how they made a whole book about playing and encoding MP3s for the beginner..unless half of the book looks like this

    Chapter 1: turning on your computer

    I find your computer

    II find the power switch

    III try to find the power switch again

  • The problem I have with VQF is this, the encoder(s) I've seen are terribly slow. Granted, it does have better compression and will take more cycles to pack those bits. I've encoded a song at 80kbps with VQF, and was quite impressed with its quality regarding 128kbps. But, it took nearly 10-15 minutes on my K6-2 333mhz w/ 96megs of RAM... so you say my system is too slow? Maybe it is by your standards and many others, but its a perfectly fine machine and while I can always get more disk space at the local computer store/nearest online retailer, I can't purchase time.

    With my CDRW and encoder combo I can rip&encode a full 74minute CD in about 20 minutes. I highly recommend people consider this encoder "Go-Go-no-Encoda", it features 3dnow!/MMX/SSE/SMP optimizations which can really give back alot of time. Check it out: http://homepage1.nifty.com/herumi/gogo_e.html

    Imagine encoding an entire album such as Pink Floyd's "The Wall", a little over 80 minutes, with VQF? Blah, not *all* music lovers have P3-500mhz/K7 Thunderbirds :)

    I'm just saying, the reason _I_ don't like VQF is because it takes way too much damn time to encode things and posesses limited support for encoders/decoders in the way that the two major ones that I've seen are not very mature and will often drain my system resources. I don't think MP3s are to be used for *everything*, but don't exaggerate its faults. It was meant to stream decent quality audio which could be likened to CD on a typical customer's equipment over a network.

    It works, quite well.
  • They did put out a book on PNG, it had a lot of nice technical information on encoding/decoding, development tools, detailed descriptions of the features, comparisons vs. gif - basically, everything this guide has, plus the nitty gritty stuff that we all want.
  • MP3 is old
    So's ASCII. So are internal combustion engines. Way older than mp3, as it so happens. Being "old" is not a reason for something to be discarded.
    too lossy
    All recording methods are lossy to some degree. "Too" lossy is a matter of individual taste.
    sounds bad
    Again, that's a matter of taste.
    and has way too many legal messes surrounding it
    Surrounding the format as a whole? Come off it. Most of the legal trouble is about copyrights, not the format itself.

    Look, you're free to like and use whatever you want, but please don't flog it on the rest of us. Your assignment for today is to go look up the term "marginal utility". It's a useful thing to know, especially in format holy wars and the like.

  • (And sorry about forcing the +2 bonus, as I can't login, I can't control the default score)..


    But isn't there a little checkbox for disabling +2 on a post by post basis? Just like the Post Anonymously checkbox...
    --
    Chris
  • MP3 -or- MPEG 2[.5] Audio Layer 3 is very similar to JPEG as is first uses a "Discrete Cosine Transform" or DCT then Quantization (the real lossy part), then Huffman encoding (pretty sure it's not LZW).

    The interesting thing about audio compression is that, unlike video or image compression many of the techniques have been around for a long time. Many of the same methods carry over from traditional DSP in EE. Things like quadrature mirror filters and such.

    To date, I still havn't found a good book on MP3's. This book gave you the skinny on the file format, which is trivial. Then goes on to talk about how he things X encoder is better then a BeOS Y encoder, etc... It's almost entirely an end users book. I would not recommend it for anyone who cares about the algorithms in MP3.

    If you are really interested, I would recommend you talk to Monty from the OggVorbis project, he defiantly knows his shit; also you might want to look through the ISO demonstration source encoder for MP3. It's a lot cleaner then lame, or mpg123.

    -Jon
  • A major flaw of the "perceptual codec" philosophy used in MP3's is that it focuses on what the average human can hear, both in frequency range and psychoacoustics. Most audiophiles have developed the ability to hear more than the average human. The MP3 codec analyzes the "spectral energy distribution" of each frame, and compares it to a mathematical model of human psychoacoustics. But that model is predicated on the average, and music hobbyists are typically far above that average, "trained ears" if you will. The MP3 codec discards some overtones and harmonics, and also "accidents" like fingers squeaking on a fretboard or the breath intake of a flautist between notes. It also discards high and low frequencies that the "average human" can't hear.

    Because the MP3 algorithms have made many such compromises, it is not generally accepted by audiophiles or music hobbyists. My earlier message that made the same point was somehow judged as flamebait by the moderators here.
  • are you some kind of flaming wierdo who plays an album from begging to end?

    Yep. A good portion of the CDs I own are continuous and you get the full experience listening to them straight-up. MP3s are nice, but I only use it for encoding tracks that exist on their own...

    Also, MP3 is notorious for a flat low end. I keep my CDs for when it's time to bring the noise.
  • For the past few years, Scot has been a big BeOS advocate, and he even wrote the BeOS Bible.

    This MP3 book is a new and different product from him. I know that he has been working on it for the past several months, so it should be very good if it took all that time.

    Overall, Scot is a cool guy.

    "...we are moving toward a Web-centric stage and our dear PC will be one of
  • Which format is smaller and with better sound quality?
  • Hmm, I'll probably have to browse this in the bookstore first. I can't imagine what could fill a whole book about mp3's. At least at an interesting level...

    As far as good or different players, rippers, etc,... couldn't that information be found online?

    Oh well. It's better than hearing about another Britney Spears release.

    Rader

  • While I respect the lowly hermit crab for its plight in having to find a new home every time it gets to big for its old one, what does hermit crabs and mp3s have to do with each other?

    Mp3 usage seems to have grown explosively, but it hasn't moved out of its old home, so that can't be it...

    --
    Eric is chisled like a Greek Godess

  • Where can we find a site to cover this? I'm interested in learning if the curve isn't to steep.
  • Just say no to: lssy cmprssin

    Sheldon
  • 256kbps MP3 is nearly indistinguishable from CD audio, even on high grade equipment- you're thinking of 96-128 kbps 22khz Napster-traded junk.
  • I'd really like to find a book that gets into the theory of these compression algorithms. I want something more technical than "type bladeence blah.wav". Does anyone know of any good references on MP3 compression? I know that Vorbis uses discrete cosine transforms and I think MP3 does too. I'm trying to get a handle on how these things work and I'm not having much luck.

  • Umm, no, that IS his name... He also wrote the BeOS Bible, and is sorta kinda like the ESR/RMS of the BeOS world (obsessive BeOS freak ;)
  • If you looked at the facts, you'd have noticed that Napster has violated the Sherman Antitrust Act. Maybe you should think before you support such fragrant violators of U.S. law.

    Yeah, those damn smelly law-breakers. They are fragrant indeed.

  • You are right, I refuse to support trusts unless they are sticklers for corporate hygene. Nothing worse than an overly aromatic monopoly.

  • The animal on the cover of MP3: The Definitive Guide is a hermit crab (Coenobita perlatus). The hermit crab is commonly found in the Indian and Pacific Oceans, and inhabits the areas surrounding the Islands of Aldabra, Mauritius, and Samoa. Despite the name of the hermit crab, which alludes to a solitary lifestyle, Coenobita perlatus are very social creatures. They characteristically travel in groups of 25 or more, and have been found living in colonies of over 100 in the wild. Hermit crabs make their homes by occupying the discarded shells of gastropds in order to protect their soft, coiled abdomens and inner organs. They prefer shells that fit snuggly in order to prevent evaporation of moisture. Most hermit crabs carry water in their shells, which they use for breathing and a water source when they are far away from the sea.

    That's the colophon info from O'Reilly. I think it makes sense... the whole community vs. loner idea... whatever.

  • Can I be picky?? Can we please get our words correct - too instead of to??
  • A little insight here. MP3s are split into frames. In a VBR MP3 different frames are encoded at different rates - dependant on how much saveing there is compression to signal or spectral error from the orginal (psycho-aucostic model depending).

    But MP3 knows only set encoding frequency e.g. 64,96,112,128,160,256 etc. So essentially what changes is the bitrate. So likes of 147kbps as you see is probably an average. The actual VBR header is encoded as what is called a Xing header and when decoded provides a lookup table onto which a decoder can jump to the different header locations within the file instead of calculating the bit position of the a header through an equation.
  • Posted by 11223:

    And for some silly reason, the Camel is perfectly matched to Perl...
  • Not if you're not logged in when you go to post!
  • Chapter 2 of the book, "How the Codec Works" tells you about a lot more than just the file format -- it's a pretty in-depth examination of the technical details of the MP3 algorithm / codec.
  • "Our" streets? How are they ours? Since when we are we entitled to steal just because it's suddenly so easy? Napster is neutral technology, but most Napster users are thieves. It's really simple. Listeners don't have some inalienable right to free music. If a musician or label wants to give away music, that's fine. Very few do.
  • Chapter 2 is very good, and can be found here [oreilly.com]

    However, One chapter makes one an expert on mp3, not. The book talks NOTHING about how each of the filters and compressions levels work, how one implements audio masking, DCT etc... The only thing that chapter gives you enough detail to actually work with is the file format, which is more or less trivial.

    The rest of the book is not technical, and at least for me completely useless.

    -Jon

  • There's always the option of plugging your lap top into your car stereo or piping your music through your sterio with a couple of cables. As computers become smaller and smaller the fact that Mp3s could be played just on them becomes lass and less of an issue, at least that's my take
  • by Jon Erikson ( 198204 ) on Thursday June 22, 2000 @05:51AM (#983909)

    Does this book actually cover algorithms and techniques for encoding MP3s? I know there are algorithms and information [mp3-tech.org] on the net, but its one of those areas where how the algorithm is implemented makes all the difference - you really can hear the difference between say BladeEnc and the Fraunhoffer codec.

    From the sounds of it, it doesn't which is a pity really as the rest of it is the sort of thing which will become out of date very quickly - the legal situtation is coming to a head and the technology is rapidly advancing. As such, I doubt this will be a huge success with those already familiar with MP3s.


    ---
    Jon E. Erikson
  • There's not a chance that MP3 players (at least, the variety that run on computers) will disappear. Even if the RIAA, Metallica, et al sued every software/hardware vendor out of existence, the specs are out there. The same bunch of libertarian/anarchist/hackers/whoever will carry on with the development and promulgation of new MP3 players. You only have to look to the examples of DeCSS and Gnutella to see the futility of attempting to stamp out a technology whose time has come.
  • While I doubt that Mp3 will been totally blown away as a file format. i gaze into my crystal ball (pause for dramatic effect) and predict that some form of digital music (perhaps one that can't be traded illegally) will prosper.
  • by imac.usr ( 58845 ) on Thursday June 22, 2000 @05:53AM (#983912) Homepage
    from the colophon:

    The animal on the cover of MP3: The Definitive Guide is a hermit crab (Coenobita perlatus). The hermit crab is commonly found in the Indian and Pacific Oceans, and inhabits the areas surrounding the Islands of Aldabra, Mauritius, and Samoa.

    The MP3 is sometimes found on overseas sites where it can evade prosecution.

    Despite the name of the hermit crab, which alludes to a solitary lifestyle, Coenobita perlatus are very social creatures. They characteristically travel in groups of 25 or more, and have been found living in colonies of over 100 in the wild.

    MP3s are commonly found in groups of thousands on servers worldwide and typically downloaded in multiple batches.

    Hermit crabs make their homes by occupying the discarded shells of gastropds in order to protect their soft, coiled abdomens and inner organs.

    MP3s hide by taking the name of Metallica songs to piss of companies who are trying to search for them in order to prosecute downloaders.

  • Odds are they have more money than everyone in Wisconsin, but that's not saying much.. *grin*
    (Oh, and that's not a dig at WI, I'm from Milwaukee and a Marquette University Student.. there's just not much money outside of Milwaukee and Madison here.)
    Oh, and to keep this ontopic, I don't think that MP3's will die because the RIAA can pressure the big player companies, but the fact is that they can be downloaded and burned onto audio CD (or, with some new players, burned in mp3 format to maximize storage) and listened to that way.

Our OS who art in CPU, UNIX be thy name. Thy programs run, thy syscalls done, In kernel as it is in user!

Working...