CD to MP3: Going Digital Means Missing MusicMP3s, bad because they have less music in them. So much less music, in fact, that your brain loses the ability to feel emotions listening to them. Okay, sure, over-compressed MP3s sound awful, especially at lower bitrates. But get ready for some strange psychoacoustics here, folks.

Producers howl over sound cut out by MP3 compression (and I see, while I was sitting on this, it got slashdotted, though no one took the bait

As Joel Selvin writes for the The San Francisco Chronicle, MP3s have less music:

…the music contained in these computer files represents less than 10 percent of the original music on the CDs.

Wow, I knew that compressed digital audio files contained less data, but less music?

In its journey from CD to MP3 player, the music has been compressed by eliminating data that computer analysis deems redundant, squeezed down until it fits through the Internet pipeline.

Of course! If they didn’t, we might stop up the tubes that make the Internet — or … um … one tube, apparently. (No wonder congestion is bad if we have just one pipeline! You need it to fit!) And there’s more:

When even the full files on the CDs contain less than half the information stored to studio hard drives during recording, these compressed MP3s represent a minuscule fraction of the actual recording.

The humanity! All those years when we were buying CDs, we were only getting half of what was recorded in the studio?! Why, that must mean they’re recording, say, four whole tracks when they record the album. And one take. (Okay, I’m assuming they somehow got this statistic by assuming 96kHz sample rates … except that’s not really half the amount of data … and that would still require 16-bit … and I don’t know who told them that, anyway.)

There are the obligatory and predictable quotes from Phil Ramone and others. I can understand engineers being squeamish about someone listening to a low-bitrate MP3 on iPod earbuds, though I wonder how they missed people taping pennies to their turntables in the 60s. (Scratches and dust, I suppose, just give you more music!)

You’ve read these kinds of articles before. They’re not entirely wrong, they just struggle to explain what lossy compression is. A journalist, I can imagine, would do that easily; I haven’t written any compression algorithms this morning so I’ll admit my own understanding of data compression is rudimentary. But, of course, what a journalist should do is talk to experts, and you hope they’ll tell you something that makes sense. In this case, they seem to explain away our ability to hear music at all. Get ready for — experts gone crazy!

Gasp at speaker designers who call into question our ability to hear digital audio — AT ALL! John Meyer, from Meyer Sound Labs:

“It turns you into an observer,” Meyer says. “It forces the brain to work harder to solve it all the time. Any compression system is based on the idea you can throw data away, and that’s proved tricky because we don’t know how the brain works.”

Go ahead. Listen to a digital audio file — right now. I don’t understand it. I have no idea what’s going on. Are you even hearing music? Who knows?

What’s missing here, of course, is the acknowledgment that the digital audio file itself also “throws away data”, because the real world doesn’t have samples. For that matter, this whole argument seems to result from a century of gradually losing the ability to distinguish between a recording and live sound. Part of the reason people are so upset by digital audio compression, even if it’s lossless, may be because people have an artificial attachment to the recording itself. But don’t tell that to people who engineer records and design speakers.

People are so desperate to prove that digital recordings are somehow evil that they even turn to research to prove that your brain on digital files stops feeling emotions:

It could be that MP3s actually reach the receptors in our brains in entirely different ways than analog phonograph records. The difference could be as fundamental as which brain hemisphere the music engages.

“Poorer-fidelity music stimulates the brain in different ways,” says Dr. Robert Sweetow, head of the University of California-San Francisco audiology department. “With different neurons, perhaps lesser neurons, stimulated, there are fewer cortical neurons connected back to the limbic system, where the emotions are stored.”

I expect that’s true — of 64kbps MP3s played through, say, a tin can. Welcome to the age of recordings: we sacrifice superior formats for inferior formats, yes, as higher-fidelity CDs give way to copy protection-laden, incompatible, inflexible, lossy compressed files. But then people get so upset that they throw out any understanding of the actual music in favor of the recording, artificially elevate old records without any real basis, call into question the basic idea of data compression even though its popularity demonstrates that it can work.

Here’s a suggestion: what about what gets thrown out when you go from a live performance to a recording? I love listening to records. I have strong feelings listening to records. (Erm, digital files.) But I’ll bet most people’s limbic systems get the biggest rush when they hear something live.

Oh, well, at least all of this should make Bob Dylan happy. If “new records have sound all over them”, and MP3s “take out some of the music,” does that mean the resulting record has the right amount of sound on it? You know, like taking your finger and wiping extra jam off of toast?

  • My mind is blown.

  • In related news, you damn kids should really get off my lawn.

  • Barry Wood

    I had no idea that as a kid I was so deprived of music when I was listening on a little 9v battery-powered transistor radio. I can only imagine the flood of emotions that would have washed over me had I been listening on a higher fidelity device… sigh…

  • A.M. Gold

    Can't be any worse than the frequency "range" on today's FM radio stations, whose engineers apparently never met a compressor they didn't love…

  • In case your mind isn't completely blown yet, I left out this quote because I think I hadn't fully fathomed its shades of meaning:

    MP3s have won the war of the formats because of technology, not because of their audio quality. "It's like hearing through a screen door," says neuroscientist Daniel Levitin of McGill University, author of "This Is Your Brain on Music."

    Yeah, just like listening through a screened door.

    Wait a minute. Doesn't music heard through a screened door sound, um, exactly the same? (Fewer … bug bites?)

  • Darren Landrum

    YAY! I finally get to post a comment.

    I saw the article on Slashdot, and all I can say is that this entire argument is based on a fundamental misunderstanding on what data compression actually is. If I make a zip file of a large text file (say, a dump from a database), the new zip is now about a tenth the size of the original text. Does that mean it now contains only a tenth of the information?

    No, it doesn't. In fact, it's not even a text file anymore. It is now a piece of data which can be used to recreate the original text file when asked to.

    Sentences like this (in reference to FLAC) are the dead giveaway that this journalist has no idea what he's talking about:

    <blockquote cite="Joel Selvin">It reduces storage space by 30 percent to 50 percent, but without compression.

    Reducing file size is compression! When we talk about lossy compression, we talk about what information we're willing to lose in order to reduce the size of the final compressed file even further. Obviously, we want lossless compression in the case of zipping up a text file, but with music and video, we have some more leeway.

    As with ZIP files, an MP3 isn't an audio file anymore. It is a set of data which can then be fed through an algorithm in order to recreate as close as possible the original file. If you want to know how much data is really lost, you need to compress your audio file, then decompress it back to a wave file that can be compared to the original wave. I think you'll find it's a lot less than ninety percent.

  • bliss

    I'll tell you this, I listen to CDs WAY more than I listen to MP3s. And the live experience, for me at least, is more engaging than listening to recorded music. There really is no comparison to listening to live music and recorded music. Now, I don't know about all this less music in compressed music stuff, but I do know that live music seems to be more inspirational than recorded music.

    Thinking about theater versus movies versus DVDs versus torrented compressed AVIs, is there more story in theaters than movies than DVDs than torrented compressed AVIs? I don't know, what I do know is that Blu Ray DVDs and HD DVDs contain LOTS more information than DVDs and the difference is so stark that when looking at images on from those two formats it really connects with one's mind and body in a way that compressed formats can never match. Still though there's for sure WAY more information in a live theater than a movie, but arguably, at least, the two types of storytelling mediums are different enough to make the matter mute.

  • Haha, this article was hilarious, on so many levels! I can see why the website's called Seattle Post-Intelligencer, they're obviously finished with intelligence and have moved onto loftier things, such as fixing the wobbles on the Internet, and the Zen-like occupation of appreciating nothing.

  • bliss

    Aren't screen doors like giant versions of the pop shields studios use with microphones when recording vocalists? Taking care of all those 'plosives coming from the front lawn or the dining room?

  • @Darren: great comments; nicely put. Too bad they didn't call you. (I assume the "finally getting to comment" comment means you were one of the people having problems with our spam filter? Very sorry about that!)

  • Des

    Old producers whining about how a new format cheapens their skills. What else is new?

    (Heh, one of the comments on the article:

    "if there is only 10% of the music contained in an mp3 file, then why shouldn't internet radio stations only have to pay 10% of the royalties?"


  • Darren Landrum


    Naw, it just meant that this is a subject I could really sink my teeth into. 🙂

  • mok

    I think every compression tool has its use – and probably what you wind up with is a lesser quality version of what you started with. Still people will use it if it's convenient and not too bad, so they do some good. That said, it is still important to talk about what is lost, in both technical and emotional terms. Strictly for recorded music, (and strictly IMO), every form of lossy compression (analog generations, digital AAC/MP3/etc.) removes the listener from the original source material by degrees. This is only important if you want to be that close to it. That is why there are people out there striving for the sound of the original masters, and quality mastering engineers dedicating their work to preserving as much as possible from those tapes on down to our ears. You can't just laugh off the importance of this discussion simply because you're not looking for the same things that some of us are from our favorite recordings. It is valid and definitely more research is required to explain why, for example, I feel better when I hear my LPs and some CDs than when I hear the same stuff encoded to MP3. To me, it's a feeling – and I can pass the double-blind tests on it too. To be fair, this is what I do for a living, and I'm very acquainted with the sound of many tracks direct from masters. But it matters to me, and you'll find a huge community of people this matters to over at Steve Hoffman's board. Recorded music changed my life – this is really about trying to preserve that opportunity for people who haven't experienced it.

  • Steve

    I also wonder about the clumsy, lossy mechanism technically known as "the ear". I mean, just think of all the pure, rarified musical data that is covered over by all that other noise out there, plus distorted by the eardrum membrane, then those silly little bones, and finally in the ear canal fluid and by those nasty hair/nerve cells. I mean, can you imagine how different what actually gets to the brain is, compared to what started out? "Do you hear what I hear?" is not just a Christmas song. (I also miss seeing all that infrared spectrum info in my increasingly lossy vision pathway.)

    As the proud owner of 4 (count 'em) turntables, I can say that some of the old "records" can still move me, noisy and lossy as they might be. My limbic system doesn't even seem to require the actual, immediate sound input to create that effect.

    Still, I wonder if "Suite Judy Blue Eyes" would tickle my limbics even more if I upgraded from a penny to a nickel on the tone arm?

  • Hi,

    At least can we all agree on a few things?

    a) mp3s do sound like s…

    b) We do not know much about our brains

    c) Journalism, pretty much like mp3, is not a lossless process!

    But, while I usually tend to fully agree with you Peter, I think this time you may be a tad unfair with our dear west coast journalist.

    And, by the way, I personally feel that nice top end and fast transcients, however you do get these (SACD, DVD-A, live, 192kHz or better recordings), are part of a satisfying listening experience and I definitely feel you do not get that with 96kbps mp3s whatever the techical reasons may happen to be.

    Cheers, Pierre-André

  • Wait…so now they're trying to tell me that today's records DON'T have 'sound all over them'? But instead they actually have LESS sound? Oh, man. I give up. I'm going back to wax cylinders.

  • M Pee 3

    Sheesh! Batch remix your stuff to wav or 320kps MP3 and shut up. As a long time Musician and DJ, i put the "BUMP" Back in everything that goes out through my home studio or live- If it moves the dance floor something must be right.

  • dead_red_eyes

    Sure … a 256 or 320 kbps mp3 rip is pretty close to the original thing. But do this, run a CD thru a spectrometer and then run the same song thru it but this time using an mp3. There's obviously a loss there. Almost all my mp3s on my computer are 256 or 320 because I'm an audio whore … and I can tell a 192 or 128kbps rip because there's some slight artifacts and it sounds like ass.

    That said, if digital music is going to be THE NEW THING … let's move to a good codec for once … like FLAC or something.

  • @Lael (and others)–

    a.) MP3s I think we'll agree are a decent delivery mechanism if the bitrate isn't squashed (so 256/320 is ideal if you can get it). More data is always better. Better playback devices are always better — well, maybe not even that, just as with the transistor radio example. I've loved listening on a mono radio some of the time. I enjoy sound enough that I can enjoy it wherever I hear it. Music ought to be enjoyable when played on sub-standard devices and media just as it does on great media. Having to adapt to different playback mechanisms is nothing new; that's why this ongoing article (which we see in various forms and venues over and over again) is ultimately misleading. It ignores the entire history of recording and playback.

    b.) We may not know much about the brain, but we do know something about how people respond to listening. And since the listener is the object, then if it sounds good, it is good. If that means more data, so be it.

    c.) I'm not meaning to be hard on the journalist — and it's certainly not this one person — but this was badly in need of a technical editor, because what's said is NOT accurate. (Darren put that best.) It's not this journalist, anyway … because we see this article over and over.

  • Just because i haven't seen this mentioned above, one of the main reasons for loss of sound information out of the studio is dithering. Dithering takes 24 bit audio, in a program like reason, and chops off the 8 bits it deems unnecessary. It then squeezes it back into the remaining 16 with a little bit of intelligent noise and light distortion to get everything as close to the original as possible. It's not perfect, and some programs do it better than others (reason does it better than cubase, which does it better than FLStudio, imho)

    Dithering can become a pretty lossy problem, especially when repeated (double-dithering) because it just keeps squishing the sound into itself. think of it as squeezing a warm, fresh, fluffy sandwich. the more you squeeze it, the more you lose what makes it awesome, and while the argument can be made that you're still getting the same sandwich, you know it's the little things that count.

    so dither only one time, at the last minute possible, right before you press the cd.

  • Alex

    All these articles express the feelings of high fidelity audio companies that see their products dying because nowadays only a few rich people pay so much money to "hear" super high quality audio. IMO, mp3 "teaches" our brains to perceive sounds differently, but not to loose the sense and the feeling of music..also i believe that in about 10 or 15 years we won't talk anymore about mp3 because, the technology that lead us to compression, will lead us to use uncompressed data again (bigger storage, faster connections and pure or not so pure .wav files)….

  • william e

    "It could be that MP3s actually reach the receptors in our brains in entirely different ways than analog phonograph records."

    This is a howler, no doubt about it. However, I don't think this story is a case of "experts gone crazy," as Peter fears. Instead, it seems a case in which a journalist got in over his head in interviews with scientists and technical professionals. The audiology professor quoted in the story seems sensible enough. The problem lies in the way in which the journalists tries to connect disparate sources to support his point.

    I am especially amused to see FLAC defined in part as the format "favored by Grateful Dead tape traders."

  • Darren Landrum

    I think Alex hit it right on the money. MP3, and now AAC and Ogg Vorbis, are interim solutions for music storage and transmission that can be accommodated by currently available bandwidth and disk space, and they happen to work fairly well. Eventually, Moore's Law will catch up, and such things will no longer be an issue. Personally, I'm hoping FLAC is the one that really catches on.

  • it might be true, but i wonder what effect these tiny in-ear headphones have. i think that even worse…

    i can still cry to instrumental tracks on my ipod, so i dont see a big issue here. most music doesnt use psychoacoustic anyway.

    and sure, mp3s dont sound like a cd or vinyl. so? thats no news to me. go listen to some myspace stream rips with cheap in-ears..THATS what i call bad.

  • I grew up with vinyl popping, 8-track kerchunking, cassette hiss, and Radio Shack cans for my head.

    With that in mind, I'll take today's 160kb MP3s or even satellite radio's 32kb or 48kb aac streams and the internet's greater range of available material any day, especially over the absurdly compressed FM classic rawk that stations have on auto-repeat.

    I also believe taht everybody has a different range of hearing. Audiophiles and expert musicians likely have different brain-wiring/sensitivities/ranges than somebody like me, especially after all the damage I've done thanks to the Radio Shack cans of yore.

  • I personally love FLAC and adopted it as a download option for experimedia releases years ago. We even have a little FLAC FAQ we put together for those who download our free releases. Everytime I purchase a new album I rip it to FLAC, burn to a CD-R, and keep my original tucked away on the shelf and the CD-R in my car. Then archiving the FLAC and original retail CD in case my CD-R copies ever get scratched. Pretty much the only time I pull out the original CD is for home use or to read the liner notes.

    Now I just want to see edirol add FLAC playback (or even recording) support for their R-09 recorder so it can act as my portable media player as well.

    Been thinking about getting one of the COWON digital media players that support FLAC at some point.

    But yes I agree. I really hope to see FLAC get adopted more widely as bandwidth and diskspace increases.

  • since battle with DRM is 'almost' over, the new focus is the old MP3…

  • Nick Inhofe

    Just to set the record straight, its not the dithering that causes the loss- its the truncation. Dithering is just a trick to increase the perceived dynamic range of the audio and prevent (some) unwanted distortions.

    While I would agree that it is best not to reduce the wordlength one's signal until the very last moment, if you have to you should definitely use dither (if possible) every time you do it. This includes plugins that might expand the wordlength while they do processing and then reduce the wordlength at their output, like a lot of the Waves processors.

    This subject is best not summarized in a forum post. For a good, detailed discussion I would like to refer to the Bob Katz book 'Mastering Audio'. He devotes an entire chapter to this subject alone.

  • richardl

    When I first saw Joel Selvin's MP3 article in my local San Francisco paper I got all bothered and was tempted to write a thorough rebuttal. But then I thought about it for a while and thought, "why bother?" I've always respected Mr. Selvin as a music journalist, but his grasp of technology leaves quite a bit to be desired. Oh well.

  • Projectile

    Wow? The media really missed the bandwagon on this one. Isn't it a bit late to finally start criticizing mp3 compression? Are we in the year 1998 again or something?

    Regardless of how much data is lost or the psychoacoustic effects. I do know a few things for certain:

    The bad sound quality of mp3s is highly overstated. I've read accounts of several controlled blind tests over the years and they all come to the same conclusion. When encoded properly, most people cannot tell the difference between a 192kbps mp3 and cd. Those people that can hear the difference find it surprisingly subtle when actually forced to compare blindly. One of these tests even included several self-proclaimed "audiophiles" who insisted that they could easily hear the difference, and they were wrong.

    Yeah, I can hear the difference, but who gives a crap unless you are listening to a great recording through a high end system. Lets face it, probably 95% of the entire listening audience are listening to over compressed music through crappy speakers or headphones with insufficient amplification. And this was the case long before the ipod or mp3s even existed. I thinks it's obvious to anyone who owns a decent pair of headphones or has spent more than $200 on a pair of speakers that there are FAR more important factors in the listening experience than mp3 compression.

    What I would really like to see artists releasing 24bit mp3s for download. That would be a HUGE improvement over the current 16 bit cd format. I can hear a far bigger difference between a 24bit vs 16bit recording than I can mp3 vs cd. So I would imagine that a 24bit mp3 would be a noticable improvement over a 16bit cd. But that wil probably never happen. Is it even possible to encode an mp3 from a 24 bit file without bit reducing first?

  • dead_red_eyes

    Yeah, 24-bit would be nice for mp3s and CDs.

  • richardn

    I'm a musician who has suffered from a genetic hearing loss [otoschlerosis] for about ten years – this gives me hearing that is very limited with a variety of "interesting" anomalies [i used to be able to EQ around it but not any more]… I was also trained as an audio engineer [obliviously prior to the hearing loss].

    The thing that I have learned [slowly] is that, whether it's traffic noise, the breath of your sleeping child or a more conventional music [either live or recorded] – hearing takes place largely in the mind as an active engagement with the world.

    Somewhere along the way the metrics used for basic measurement of sonic phenomena have supplanted the actual experience of human hearing as it exists in life.

    I couldn't give too hoots about analogue/digital lossy/lossless etc – my very best musical experiences always have been in sub-optimal circumstances [even before the tin ears arrived] – whether it was the worn-down tape of Caravanserai playing on a wood-paneled flat-batteried tape player or the first time I heard Gould's '55 Goldberg on shortwave…

    The thing we call hearing/music really does happen mostly as an active process in your mind [heart or soul if you prefer]. I really think it's in our best interests to realize this and get on with the work at hand.

  • richardn

    Or more precisely:

    Now that I am deaf I have found my hearing very much improved.

  • I agree this is mostly a moot point. Doesn't make a bit of difference they way most people listen the equipment they listen on.

    This kind of thing will never die. The whole brain thing is just a silly new twist. I am sorry if anyone reads they are really missing out. That all this emotion has alluded them. I still think that it not the data that is captured but the emotion that is captured which is important.

    <blockquote cite="I had no idea that as a kid I was so deprived of music when I was listening on a little 9v battery-powered transistor radio. I can only imagine the flood of emotions that would have washed over me had I been listening on a higher fidelity device… sigh…">

    I know what you mean. I fell in love with music listening to my father's semi-destroyed vinyl through crappy head phones late at night. Imagine how much more in love I could be…

    Anyone that cares a lot about audio quality and has the equipment to match is not going to listen to lossy anyway.

    I agree that compression lossy or not will be gone as storage and bandwidth grow when there are 30Tb Ipods and super fast wireless internet everywhere. I don't think it will be as long as 10 years either. This debate will disappear (finally.)Then "experts" will find something else to complain about. It pretty much has happened with any new format. I bet people complained when they switched from cylinders to discs.

  • Greg

    CDs may have "sound all over them" and also have "90% less music," but it all ads up to the end user getting remarkably less sound information than was originally recorded. Audio compression removes the dynamic range, which is probably the most important thing for the expressiveness of music, and file format compression removes most high frequencies which drastically affects the timbre, which is the *other* thing that makes expressiveness in music.

    It's the same as starting with an HD video , bringing the contrast down 50%, then bringing the resolution down to 640 x 480.

    No one buys (or even downloads) movies like that, so why do we do that to our music?

    Of course, MP3s can be pretty transparent at higher bitrates, but I think that music-lovers would really want the "HD" version (whether analog or lossless).

  • i think richardn has it.

    i remember a similar point made by a synth professor I had, though not exactly about sound quality. He was discussing the argument often posed that electronic music somehow lacks the same quality possessed by music created for traditional acoustic instruments. specifically, some traditional musicians have posed the argument that because "artificial" processes are used to generate both sound and composition in many forms of electronic music, it lacks the same intentionality and conscious development found in more traditional methods. his response was that (paraphrasing here) "music is an element of culture, and while we can argue over certain qualities of it, its always dubious to suggest that there is an absolute value to any one element of it."

    i think its disappointing that instead of using our ability for rational/analytical thought to temper and broaden our viewpoint, we often use it to segment and isolate ourselves into viewpoints of an increasingly abstract nature. and in this particular case, it certainly seems like ignorant abstraction, which is pretty frightening….

  • I could do ten times the damage with a limiter that I could with an mp3.

    Many people do.

  • boxguy

    I don't think there's any argument that on any $400+ pair of speakers, a high quality recording as a 16-bit WAVE file sounds better than as a 320kbps MP3.

    I would like it if they came up with a 24-bit lossless format at a reasonable cost.

  • Cut

    @boxguy: You may think that the difference on a $400 set of speakers is obvious between a 16-bit WAVE or AIFF file, and a compressed 320kbps MP3.

    Not so, though. You would be very hard pressed to spot any obvious differences, I think – at least that is the case on my $3500 amp+speaker+DA-converter setup. It also presupposes that you know the music you're listening to intimately, but even then, through A/B testing, your mind WILL play tricks on you.

  • I love how these articles get a lot more comments in response then the others 😛 Even CDM'ers love a bit of controversy and sensationalism now and then.

    As a masssssssive leap to another tangent, and without an apt segue whatsoever, im reminded of the 4 cd release by The Flaming Lips. It contained 4 cd's which were meant to be played on 4 cd players at the same time. Despite planning to, my previous housemate and i never got around to playing his copy of this. I think deep down inside, The Flaming Lips knew all along that cd's were only supplying 25% of the actual music.

  • The last cries of a crumbling empire.

  • @Dave: That's not a massive leap. You've hit the secret! Now, where can I buy three more CD players… hmmm…

    Well, yes, the CDMers enjoy a little controversy, but I'm glad to see mention of Bob Katz's mastering book — good stuff. And let's not start fearing compression in general — just the awful brick wall compression, as heard on the larger radio stations. (My public radio sounds just fine, thank you!)

  • bliss

    The thing is that the brain filters out information all of the time because we humans need to focus. So it's not necessarily a bad thing that technology is also capable of reducing the amount of information that gets to the brain. Sounds like the brain has less work to do. If someone is listening to MP3s on an iPod® while riding his skateboard and eating a Popsicle® on Mass. Ave in Boston, then it might be a good thing that he has less music to listen to. It might be a good thing for office workers listening to MP3s in their cubicles, too. After all, they have to get those Excel® sheets in on time. Anyway, sitting comfortably in the La-Z-Boy® and actively listening to MP3s, one may actually be listening to more music than when she was passively listening to MP3s while getting her workout on the StairMaster®.

  • Good point bliss.

    I listen to a lot of mp3, but it will never replace my audio cds or my own recordings.

    It's background music I enjoy listening to while doing other stuff. If I care enough about the music that I want to sit down and really have a good listen, it won't be in the format of a 128kbit mp3.

  • Sam

    Probably too late of a comment to matter, but here's a much older article as a point of historical reference on this kind of complaint.

    Digital Recordings, A Tragedy Unveiled in the History of Art

  • I was thinking of a way around this "less music" problem and the missing "emotion" of the music. If we get the original cd, and buy the mp3 album on iTunes, we can in theory sum these two recordings together and the phase cancellation will leave only the "lost" "music" from the "recording". I call this the LMFC factor. IF we market the LMFC's right, we can ship "patches" to all the music releases on iTunes to fill in the missing sound holes, and we can even deliver this via the Tube of the internet. I do think we would need to include a safety pamphlet though. I shudder to think what a world with 25% more "emotion" put back into its music might do. Hasnt anyone seen Equilibrium???

  • Michael

    This is a wonderful article, and it leaves me wondering only one thing: where can I go to get some of this music they cut out and throw away–do I find it in a dumpster behind MP3.CORP, or at a thrift store? I want some of the reject bits, and I bet if I get enough, I can stitch it back together and get something really great!

  • Michael

    OK, more seriously now, I found this article very interesting:

  • matt

    By MP3'ing the audio you do loose detail. Reverbs, transients, noise, depth is all missing or effected so yes Id agree half of what is recorded is actually missing.

  • an expert

    I think you guys don't really understand the brain. The brain is like a stomach and the ears are like a mouth, and you are hungering for data. A compressed mp3 gives you less data therefore it is less satisfying. The music is just like a flavor to make you consume more data.

    But when you are poor, you take what you can get. I think that is why I listen to so much more music since mp3's came around. I'm trying to make up for the loss of data.

    However, I don't know if it's true that mp3s make me feel less emotion. I must admit that I've been much happier ever since music got cheap.

  • Devil

    yeah, this article is all anti-mp3 PR, for sure… but the fact is, digital formats (including CDs) really DO sound like shit compared to analog. if you can't hear the difference, it just means your ears either suck or else, more likely, they just haven't been trained to listen critically. mock it if you want, but even the best, most hi-fi, fastest sampling rate digital recording sounds thin and cold when compared to a great, old-school analog medium like 2-inch tape (or even an LP, really). that's just a fact, apparent to anyone who really listens.

    i remember when CDs first came along– i bought sgt. pepper's the day it was released (6/1/87) and took it home and compared it to the LP, side by side, and even on my crummy system i could tell that the CD sounded lifeless and cold. i admit it's pretty great to get rid of scratches and pops so that you can hear the music clearly, but i agree that CDs definitely lack the same power to move the listener emotionally that analog mediums possess.

    if you talk to some industry people, you'll find that pro recording engineers will generally tend to look down on digital, because their professional experience has shown them how much better analog sounds. i'm thinking that if the reporter talked to any pro engineers or producers, then he/she probably picked up on the "digital is inferior" vibe without really understanding how to explain WHY it's inferior… maybe? i don't know… but seriously, talk to some people who record sound for a living (and who have been around a while) and try to get them to laugh along with you at the notion that "digital sucks." they will not laugh with you. they will laugh AT you.

    (of course, virtually all of them record digitally now, because it's cheaper and easier (and if you go 24-bit/96KHz it sounds OK– not great, but OK), even though they constantly bitch about how much they hate it. but the industry has gone digital, and that's not gonna change. like they say, progress makes things easier, not necessarily better…)

    oh, and for the pinheads making fun of the idea of emotion in music– wtf do you even listen to music for? to feel hip?? to fit in with the cool kids at school?? because you obviously don't actually LIKE it, if the idea of it moving a listener emotionally is a big joke to you. grow up and get some ears, kids. the reality is that digital sucks ass compared to analog. people who have grown up on CDs have basically had their ears dumbed down so that they have no idea what they're missing– as for me, i'll take 1/4-inch tape over 24/96 digital any day of the week (and i have to believe that all of you would too, if you ever got the chance to actually listen to both side by side).

  • damn, i wanna get in on this comment fest, but i kinda think everybody has valid points! even 'brain is like a stomach' guy. yeah – mp3s suck, but they are also super cool, that neat watery, digital flangy sound you get on the high hats with a 128k file? love it!

  • I can see it now: a VST plug-in to simulate badly-compressed MP3 files because they're "vintage" and "cool."

  • Best comment here yet has to be "An Expert" with his revelation that mp3's are perhaps the audio equivalent of crack cocaine! Since the scourge of MP3's, kids are now consuming more music then ever and we took a sneak peak into his very shattered soul with:

    "But when you are poor, you take what you can get. I think that is why I listen to so much more music since mp3’s came around. I’m trying to make up for the loss of data."

    This would happen to be the ONLY reason im sure. Listening to more music TOTALLY has no relation to the emergence of a myriad of alternate and rapid methods to purchase music, let alone the increase in global distribution on all levels. No it is only due to there being less MUSICAL NUTRIENTS in the emm pee threez. The loss of data is driving the kids crazy!

    Beam me up Scotty.

  • an expert

    Dave Dri is nice.

    The Devil is evil. As an expert, I find things that can be described as "sucking" or "good" are usually not facts. Also, I find that "difference" is not bad unless I am being a bigot.

    Finally, much music has been created, recorded, encoded, and distributed digitally and what was left of the music moved my emotions quite vigorously. Perhaps emotions resonate to the song itself (good ol' harmony, rhythm, melody, lyrics even) and not only the clarity or "goodness" of the sound that carries it. That would explain why our measurements of emotion are so varied.

    And no format is perfect, even live. Unless you like destructive interference. And Jimi Hendrix's guitar is kinda fuzzy.

  • richardn

    I now have this vision "Devil" [see above] sitting in an anechoic chamber listening to "St Peppers" over and over again on a turntable while wearing the latex-molded ears of Sir George Martin.

    [if such a product exists I'll have some too]

  • richardn

    oh and yes, I, of course, meant "Sgt." not "st"

  • This whole thing again? The last time this controversy floated through the papers [1998?] I felt it was timely and totally agreed that mp3's sucked a whole lot of goodness out of music… but that was mostly because the bulk of what i "found" in mp3 was between 96-160kbps.

    Now, and already somebody has mentioned this, but i'll state it again: 320kbps rips sound better than most CD's, and often times converting a CD to 320kbps via LAME will even enhance the material. YES; the process of conversion from any format to mp3 does tend to "up the contrast" a bit — long, spacey reverbs do sort of fade away longer and fast transients are cleaned up where in the original they felt a little thick. The dynamics of mp3's are often slightly more stepped, and the stereo field feels pushed apart… These are mostly good things! I watch my movies with higher contrast for greater impact, too.

    As a producer, I'm happy to be able to rip work to 48khz, 320kbps mp3's. They're not nearly as bad as these weirdos with their broad generalizations seem to paint. Maybe they don't realize they can actually have a bit more of the audio spectra recorded in an mp3, frequency-wise vs. CD's?

  • Pingback: micablog » shut your precious whining analogue mouth()

  • Clydicus

    MP3s have won the war of the formats because of technology, not because of their audio quality. “It’s like hearing through a screen door,” says neuroscientist Daniel Levitin of McGill University, author of “This Is Your Brain on Music.”

    That's why they put speaker grille fabric on speakers – to pre-slice the audio, so it will fit through screen doors.

  • Funny, when I saw this on /. and read the 'screen door' comment, my first thought was 'grill cloth'…

    LAME also has optimizations in VBR mode to deal better with 'transients' btw. These are not new either, but were originally in 3.90.3, although limited to that version until the last year or so; as of 3.96.x they're in by default and the –alt-preset (or any other preset) switch will simply default to whatever the comparable V setting is (V0-V2 are all acceptable replacements for 256-320k). LAME is and always has been suboptimial for CBR (320k CBR in other words) as most of the work that's been put into tuning it over the last several years has been on its VBR mode.

    Fraunhofer's "improvement's" are less well documented (and in many cases trade secrets) but CBR fares better there and is its continued claim to fame (especially for streaming data). AAC and other less known wavelet methods are again making strides over the 'older' tech, although modern mp3 codecs referred to here (not sure about Xing) are coming alone fine too. And then there's Vorbis, which also sounds a lot better than it did several years back (Ogg is just a wrapper btw, like mp4).

    CD's and vinyl are largely fixed standards where the actual implementation is fixed (RIAA curve, redbook audio etc). Although the production and playback mechanisms are subject to refinement, outside of rediculously overpriced audiophile gear you're unlikely to find anything available new that actually is refined.

    Compare a higher end CD player from the early 80's to one from the same (or similar company) today. You won't even find a product category for that outside of an audiophile shop as it's all based on the same highly integrated reduced set of cheaply made circuits, regardless of whether it winds up in a boom box, car stereo or standalone player for a component system. You'd have to go to the audiophile shop to find anything that has any real design put into it and that will last longer than a few years. The same thing for turntables, except you're setting the benchmark for a 'good' mainstream turntable further back (late 70's).

    And their replacements will also be fixed, which gives the 'industry' a nice 10-15 year stepping between everyone 'rebuying' their favorite music in the 'new' accepted standard. Although I'm not sure that anyone really cares about DVD-A or SACD. Tape shouldn't even be introduced in 'sound quality' discussions imo, at least not the consumer implementation.

    Mp3's (and other digital audio standards) are not subject to these limitations and will steadily improve over time, while older versions will play back just fine without having to dig out that old 8-track or tape deck. Any attempt to compare a digital file format to a physical format is only relevant for as long as it takes the programmers to release another major stepping on the algorithm(s) they're working on for the digital format.

    Incidentally imho vinyl's continuing existence is 2 fold, on one hand its lifespan goes back to the early spread of recorded commercial music and the physical form will hold up reasonably well as long as it's properly stored. So my mom searches 2nd hand stores for listening to her 'christmas library' etc. On the other hand it's the preferred method of distribution for indie punk/hardcore/rawk etc (7") and dance music (12"). In the latter case certainly mp3's would be preferred for 'promotion' and general (ipod etc) listening but most independant artists I know consider a vinyl release to be a certain benchmark. This is why they're still done even though they're quite often done at a loss of break even financially. Some will argue the 'sound quality' point still but I always find that funny, considering that a 1200MkII isn't a standard due to its sound quality. It's built like a tank though for sure (and pitchable)…

  • Pingback: Going Mobile Means More Static()

  • Marcus

    <blockquote cite="myself">MP3s are the GIFs of the ears.

    Audio compression isn't much different then image compression. Sometimes a compressed picture can be a fraction of the size and look as good as before, and sometimes a more colorful picture will compressed the same amount and look like crap. We have to realize that every beat doesn't have any real significance it's the bigger picture that counts. When we look at a web page we don't take a moment to look at every pixel, we scan it for what we want. The same concept can be applied for music. I can honestly say I can tell a difference between 128kbs and 256kbs (65% of the time) but unless I am in a quite dark environment I couldn't tell the differcne. I think that a better quality copy of a recording will amount to a better experience if the recording is of high quality, if the audio equipment is of quality, and if the environment and the mindset of the listener are in the right condition.

  • Pingback: Mike Wilson » Blog Archive » the cost of compressed music()