mp3s' rotten fidelity - the proof

September 22, 2009 12:03 PM

Researcher John Meyer has devised an objective index demonstrating that mp3s offer far lower fidelity than either vinyl LPs or standard CDs. And yet this eight-year study at Stanford University shows that prolonged exposure to mp3s leads young listeners to prefer the format. No wonder record producers are despairing.
posted by Paul Slade (98 comments total)

I really hope he did all this research using music from John Mayer.
posted by olinerd at 12:04 PM on September 22 [9 favorites]

Your favorite sound encoding format sucks.
posted by Joe Beese at 12:05 PM on September 22 [2 favorites]

LOLaudiophiles.
posted by fourcheesemac at 12:08 PM on September 22 [6 favorites]

Not clear on why "record producers are despairing" moreso than in the 1950s - 70s, when "young listeners" typical exposure to new music was through a shitty ass one-speaker transistor radio.

The listener controls how he listens to the music. Deal with it.
posted by drjimmy11 at 12:08 PM on September 22 [24 favorites]

I also like the distortion an amp produces when you run a guitar through it and the sound of Darth Vader's voice as it sounds coming through his mask.
posted by Astro Zombie at 12:09 PM on September 22 [2 favorites]

MP3s are adequate in terms most people care about.

The perfect is the enemy of the good.
posted by rokusan at 12:10 PM on September 22 [10 favorites]

So, he took the bitrate (or bit encoding capacity of an analog format) and divided it by 100,000, without accounting for psychoacoustics whatsoever? Unless I'm missing something, that's a pretty trivial result. Well, the accompanying table is kinda neat, I guess.
posted by blenderfish at 12:10 PM on September 22 [7 favorites]

Some prefer Razor Point and some prefer Sharpies.
posted by bz at 12:10 PM on September 22

Although I'm OK with high bitrate MP3's, listening through earbuds is an awful way to absorb music. I hate those things.
posted by davebush at 12:11 PM on September 22

As long as record producers are fighting the Loudness War, I don't think their first concern is the quality of the audio.
posted by Plutor at 12:11 PM on September 22 [12 favorites]

Perceptual encoding is pretty much what MP3's are about, and that John Meyer article makes no note of that whatsoever. Humans don't hear "bit depth", we hear sonic events, MP3 is designed to replicate the perceptually indistinguishable sonic events with less information. They are not perfect at this, but to not even address it, and be an audiophile - well, it makes you look about as ignorant as the standard audiophile.

Confidential to the author: I have a rock I will sell you for $500, if you put it under your amplifier it will make your music sound much better, I have graphs and everything.
posted by idiopath at 12:14 PM on September 22 [11 favorites]

His 'fidelity potential index' (which sounds like something out of an eHarmony ad) also says that CDs sound better than vinyl. There's something for everybody to get mad about here.
posted by box at 12:15 PM on September 22 [8 favorites]

All things being equal, the more information a format can transmit, the better the sound will be. So here are the formats broken down into their bare bit potential some with high and low ranges.

This is a completely ridiculous argument. MP3 format is based on a lossy compression algorithm, the entire point is to use less information per unit of time. This is like saying that a zip file with a text document in it is a lower fidelity format than a plain text file, because the plain text file uses more bytes per character to express its content.
posted by burnmp3s at 12:17 PM on September 22 [9 favorites]

NiN has the right idea, here... two different versions of the album, one designed for the average listener with ear-buds who'd just as happily listen to the same music on FM radio, and one designed for the audiophile with nice headphones or a decent home stereo. MP3 or AAC files with the loundness cranked for the buck-a-song downloader, and FLAC or something similarly nice, properly produced for best audio effect, for the enthusiast. You can probably justify charging more for it, or selling it in an album bundle, too, and the music fan will eagerly pay for convenience and quality. (Note how the Beetles mono box set sold out, despite being stupid expensive.)
posted by Slap*Happy at 12:18 PM on September 22 [1 favorite]

The Good Enough Revolution: When Cheap and Simple Is Just Fine in the September Wired talks about this.

Bottom line: Low-fi audio you can download for free and listen to anywhere will always be more popular than more expensive audio you need more technical equipment to listen to.

And, frankly, through the tiny, tinny speakers of my PC, .mp3 or not isn't going to make much difference.
posted by anastasiav at 12:18 PM on September 22

Does this guy have any credentials beyond working for (or perhaps owning) a high-end speaker company with the word "Research" in its name?
posted by He Is Only The Imposter at 12:18 PM on September 22

Well, now I feel justified lugging my turntable, speakers and a portable generator on a cart behind me as I run. Still can't figure out how to stop the records from skipping though...
posted by sciurus at 12:18 PM on September 22 [2 favorites]

Previously: MP3s vs $1,,400 (£1,000) hand-made headphones

So to put the evolution of music into perspective and evaluate the stages, it is important to compare the fidelity potential of the various formats whether iPod, mp3, SACD or DVD-Audio.

I got worried when he listed iPod next to MP3. Is he comparing MP4 vs MP3? Or the sound processing in iPods? What about the creation of the media (some cheap labels recycled records, and you might even see bits of old labels in the new pressings)? And there was no mention of tuners, speakers, wires or any of the "sound carriers" involved.
posted by filthy light thief at 12:19 PM on September 22

From a subjectivist standpoint, isn't it illogical to say that people think the worse one sounds better? Whatever people think is better, is better, in their world.
posted by HotPants at 12:20 PM on September 22 [3 favorites]

box: "also says that CDs sound better than vinyl."

vinyl? haha. that's... that's not even a thing, is it? ha. what are you talking about, crazy person?
posted by shmegegge at 12:20 PM on September 22 [1 favorite]

burnmp3s: "This is like saying that a zip file with a text document in it is a lower fidelity format than a plain text file, because the plain text file uses more bytes per character to express its content."

No, because zip is lossless. I am with you on the speciousness of his argument, but unlike a zip archive mp3 does lose data, but it can, within a margin, do this without losing audible signal.
posted by idiopath at 12:21 PM on September 22 [2 favorites]

Of course, this is all mathematical and therefore useless. Do vinyl LPs actually meet their "fidelity potential index" specs for all time (wear) and listening conditions (jogging)?
posted by DU at 12:21 PM on September 22

Well, now I feel justified lugging my turntable, speakers and a portable generator on a cart behind me as I run. Still can't figure out how to stop the records from skipping though...

I don't know what any of those words mean.
posted by Astro Zombie at 12:21 PM on September 22

This is perhaps the worst "research" I've ever seen.

By this methodology, HDTV and analog TV have equivalent fidelity. (HDTV is actually worse, if there are subchannels.)
posted by smackfu at 12:24 PM on September 22 [1 favorite]

Because it's hard to strap a portable cd player to your arm and even harder to strap a big-ass turntable.
posted by moonshine at 12:28 PM on September 22

When I was a kid everything I heard was on AM radio, through a single speaker with a frequency response approximately equal to the telephone. And yet I wouldn't prefer it to my current audio system.
posted by tommasz at 12:33 PM on September 22 [2 favorites]

I am with you on the speciousness of his argument, but unlike a zip archive mp3 does lose data, but it can, within a margin, do this without losing audible signal.

Yeah, I completely agree with that. I'm just saying that counting bytes per unit of content as in the article does not in any way measure fidelity, or even measure the potential for a given format to be high fidelity. Fidelity just means how close the sound produced is to the original, which is related but not directly proportional to the bitrate of any digital format.

If you want to listen to an old Amiga song that was created with a tracker, for example, the highest fidelity format you can store it in is a MOD file. But the chart would list the "Fidelity Potential Index" of a MOD as being extremely low because it is a very efficient way of expressing the song in terms of bitrate, even though the result when played on an Amiga machine with the same hardware would be exactly the same as the original.
posted by burnmp3s at 12:34 PM on September 22

devised an objective index demonstrating that mp3s offer far lower fidelity than either vinyl LPs or standard CDs.

Mp3s have lower fidelity than CDs? No shit, sherlock. Standard CDs have the highest fidelity available short of getting the studio masters (DTS audio blah blah). They are raw, uncompressed audio. MP3's are lossy compressed versions of what is on CD.

And the argument about vinyl is idiotic. Studio records are recorded digitally. They are mastered digitally. CD's are digital reproductions. The CD is the most accurate reproduction of what the artist wants you to hear.

Vinyl at best has a 20-25khz frequency response after the first play. But according to RIAA standards, it is acceptable if the frequency response of a vinyl record declines to 10khz after 35 plays. Furthermore, the mechanical operation of a turntable necessarily introduces artifacts into the audio that were not intended to be there by the artist. You might like the way it sounds, but that is not "fidelity", i.e. a true representation of the source.

Comparing vinyl to CD's is like comparing a UHF over-the-air broadcast with Blu-ray. The Blu-ray is a more accurate representation of the source material. But it is not an accurate representation of how you watch the Partridge Family on your parent's Zenith console with the rabbit ears, multiple ghost images, and the dicey vertical hold.

It would be much better if audiophiles simply admitted that they liked the ritual surrounding playing vinyl records, and the retro sound (hiss, crackle, whatever) the turntable makes.
posted by Pastabagel at 12:35 PM on September 22 [19 favorites]

No, because zip is lossless. I am with you on the speciousness of his argument, but unlike a zip archive mp3 does lose data, but it can, within a margin, do this without losing audible signal.

Parts of MP3 are lossless (it uses huffman encoding.)

An even better question is FLAC. By his stupid stupid reasoning, a FLAC-encoded CD has half the audio fidelity of the original CD.
posted by blenderfish at 12:36 PM on September 22 [3 favorites]

when you do double-blind tests with mp3s people can't tell the difference.

c't The Magazine for Computer Technique: english translation (original in german)
But when deciding between 256 kbps encoded MP3s and the original CD, no difference could be determined, on average, for all the pieces. The testers took the 256 kbps samples for the CD just as often as they took the original CD samples themselves.

maximum pc
in fact, that in many cases, we were unable to tell the difference between an uncompressed track and one encoded at 160Kb/s, the bit rate most of us considered the absolute minimum acceptable for even portable players.
posted by bhnyc at 12:39 PM on September 22 [1 favorite]

If you play two CDs at once, does that give you double the fidelity?
posted by smackfu at 12:41 PM on September 22 [4 favorites]

in fact, that in many cases, we were unable to tell the difference between an uncompressed track and one encoded at 160Kb/s, the bit rate most of us considered the absolute minimum acceptable for even portable players.
posted by bhnyc at 3:39 PM on September 22

The track they used was 4'33".
posted by Pastabagel at 12:42 PM on September 22 [2 favorites]

Back in high school, my friend got some kick-ass JBL studio-quality speakers from an auction at a radio studio. They were so damn heavy we could hardly lift them up stairs to his room. The casing was some sort of super-dense particle board. He got them going through a big 200 watt amp (sorry, I don't have more detail than that) and played "Sultans of Swing" by Dire Straits (on CD).

The quality was freaking amazing. You could hear the nuances of the guitar solo and tiny details in the drumming you might not otherwise notice. It was like the difference between regular television and High-def. Not something you pay attention to after a while, but noticeably better.

This was around the time MP3s were starting to catch on. Later he got some 128kb mp3s going through the same system, and you could here the flaws. Sort of a swishing, slightly-out-of-tune A.M. radio background sound. A fluctuating hiss, for lack of a better description. There was no depth to the sound.

Another time I walked into a store and heard some Van Morrison playing. But the music sort of made me stop and say - wow, that sounds good. It was so good, I started looking around for the radio. It was one of those white BOSE cd players you see flyers for all the time.

These experiences didn't lead me to become an audiophile, but if the day ever comes when I have some disposable income, I'm going to buy some kick-ass speakers and a record player.
posted by Brodiggitty at 12:42 PM on September 22 [1 favorite]

Did you seriously just post a link to some table on an audiophile site? C'mon. There's no there there.

I mean, there's a piece of methdology described thusly: "Since the sampling frequency for the CD format is 44.1 kHz -- roughly double the highest frequency (20 kHz) it can reproduce, the analog equivalent sampling frequency is calculated to be double the highest frequency that medium can deliver."

Christ. That's just idiotic.
posted by mr_roboto at 12:43 PM on September 22 [4 favorites]

I wonder if this has less to do with quality than with familiarity with that quality of sound. Younger people are most likely not going to have heard much music on CD or even at a high enough encoding rate to have no degradation. To them, that's what music is supposed to sound like. If you give them a CD to listen to, if they hear a difference at all it's going to sound oddly 'full'. I can imagine them preferring something a bit flatter because that's what they know. Arguments about which are 'better' go out the window.

I know that when I hear an MP3 or a CD of music that I listened to on cassette or LP when I was younger, it sounds wrong to me. My copy of 'Siamese Dream' by Smashing Pumpkins got something like 400 plays in my crappy car stereo, and eventually the wobble and warmth of the worn-out tape became part of my experience of the album. Similarly, my LPs had a familiar pattern of crackles and pops that were there every time I played them, and when I hear the same music without it now, it feels wrong, even though it's 'objectively' better sounding without the noise.

That argument applies a lot less to music that the listener isn't familiar with (which I assume the studies used) but still I could see a preference for familiar sound quality.
posted by Barking Frog at 12:51 PM on September 22 [2 favorites]

You can fix the loss of fidelity by playing your mp3s out to good speakers through Monster-brand cables. They are the best-- they're very expensive.
posted by Mayor Curley at 12:59 PM on September 22 [20 favorites]

I highlight all my MP3s in green because it helps the processor track the bits and improves the sound.
posted by entropicamericana at 1:02 PM on September 22 [13 favorites]

I don't know who is more annoying, audio nerds or wine nerds. Both strike me as people with two much time and money on their hands who want to cultivate sophisticated, uber discerning taste so they can look down their noses at the commonfolk because they have nothing better to do than to claim to be able to taste the oyster shells in the soil that grew the grapes or some such horseshit.
posted by nathancaswell at 1:13 PM on September 22 [2 favorites]

No, no, the audio nerds do things like inventing MP3. I think you are thinking of audio poseurs.
posted by idiopath at 1:21 PM on September 22 [6 favorites]

"too", something about this gets me worked up :/
posted by nathancaswell at 1:21 PM on September 22

I put all of my MP3s in a special folder named 'Super High Fidelity Version.' They all sound much better that way.
posted by metagnathous at 1:22 PM on September 22 [1 favorite]

'Audiophile' is a lot like 'hipster,' isn't it? It's a word that you use to refer to other people.
posted by box at 1:22 PM on September 22

the two aren't really all that comparable. wine nerds don't just straight up talk out of their ass about how wine is made or how your taste buds work. (at least, if they're actual wine nerds. there's no accounting for people who are actually just ignorant and spouting bullshit.) when they say that they can taste fennel or fruit or earth in a wine, it's just a way of cataloging and describing what a thing tastes like so others can understand you. they're not saying they can taste what's in the soil, they're saying they can taste what the growing process has done to the flavor of the grape, which makes sense. that's why the grapes are grown in this or that climate, soil, whatever.

music nerds, though. as some people have hinted, there's an entire set of music nerds who will specifically spout total nonsense about how lasers work or how electricity is conveyed over wires in order to justify buying a marker for 20 dollars or a cable for 1000. to say nothing of people who believe you get better audio fidelity from 60 year old worn out vinyl.
posted by shmegegge at 1:24 PM on September 22 [3 favorites]

It would amuse me to no end if the folks at Vice magazine were to offer a custom branded hipster web browser.
posted by idiopath at 1:26 PM on September 22

shmegegge: "music audio nerds, though"

my bad.
posted by shmegegge at 1:26 PM on September 22

I don't know who is more annoying, audio nerds or wine nerds. Both strike me as people with two much time and money on their hands who want to cultivate sophisticated, uber discerning taste so they can look down their noses at the commonfolk because they have nothing better to do than to claim to be able to taste the oyster shells in the soil that grew the grapes or some such horseshit.

Here here! A healthy mix of people who have deluded themselves into believing there is a difference and people who have convinced themselves that there is something wrong with them for not noticing the difference is all that's required to keep everybody from breaking through the bullshit and engaging in worthwhile pursuits.
posted by HotPants at 1:28 PM on September 22

I hate the word "hipster" too.

All "hipster" means is "somewhat fashionable urban dweller between the age of 17 and 35 that I am not friends with."

Seriously, look in the mirror you vans wearing, plaid shirt clad, track bike riding dipshit, those "hipsters" you're bitching about look and behave EXACTLY LIKE YOU AND ALL YOUR FRIENDS DO*. Living in New York just gives you the luxury of judging people who look 99.9% the same as you incredibly harshly for the .1% they don't look like you. If I transported you both to Lincoln, Nebraska, you'd probably be BFFS you'd have so much in common.

*yes, me and all my friends too
posted by nathancaswell at 1:30 PM on September 22 [2 favorites]

If you play two CDs at once, does that give you double the fidelity?

Depends. Is it a Flaming Lips album?
posted by Iridic at 1:36 PM on September 22 [3 favorites]

shmegegge: "the two aren't really all that comparable."

As much as I love both wine and audio, and the nerdery thereof, I have to disagree; I think they're very comparable. Both subjects have a lot of grounding in actual science and have a significant number of enthusiasts who are interested in learning about those things. But both have become defined in the public mind as pseudo-science and hand waving. Notes of leather and oak! Tube amps!

That's not a criticism of either field, but rather a criticism of people who see winophilia or audiophilia as an end, instead of a means toward the acquisition of facts.
posted by Plutor at 1:38 PM on September 22 [1 favorite]

box: "'Audiophile' is a lot like 'hipster,' isn't it?"

Audiophiles actually use the term. People explicitly produce audiophile products for a market of audiophile consumers. People who are hipsters are so afraid of becoming a demographic that they refuse to self label. Audiophiles are pretty up front about being a demographic, and happen to be a market segment that is willing to spend huge amounts of cash for some big words and flimsy logic.
posted by idiopath at 1:39 PM on September 22 [2 favorites]

If I transported you both to Lincoln, Nebraska, you'd probably be BFFS you'd have so much in common.

For one thing, we'd both bitch about the jackass wizard we'd run afoul of.
posted by Copronymus at 1:42 PM on September 22 [14 favorites]

I'm not defending people who spend $500 on headphones or make up baloney about wine, but I will humbly assert that it is possible to focus on a particular stimulus (taste of wine, quality of music) and become more proficient at discerning different qualities and nuances and still not be a snob. Otherwise we all have to agree that Dan Brown is a "good enough" writer and not bother with, you know, literature lest we be deemed elitist. The tricks are:

* being able to let go of perfection if it's too expensive or unavailable and just enjoy the song for the song's sake, or the wine for the pleasant buzz.
* not looking down on someone who doesn't have the refined tastes you do.

For the record, in my opinion you can get great wines for ten bucks, and nice headphones for 40 bucks. And listening to my old records on vinyl is great as a nostalgia experience but blows as a sonic experience. YMMV.
posted by freecellwizard at 1:47 PM on September 22 [6 favorites]

I mean, there's a piece of methdology described thusly: "Since the sampling frequency for the CD format is 44.1 kHz -- roughly double the highest frequency (20 kHz) it can reproduce, the analog equivalent sampling frequency is calculated to be double the highest frequency that medium can deliver."

Christ. That's just idiotic.

Sounds like someone accidentally discovered the Nyquist-Shannon Sampling Theorem.
posted by albrecht at 1:50 PM on September 22 [1 favorite]

People who are hipsters are so afraid of becoming a demographic that they refuse to self label.

It isn't so much that as it is that to openly embrace the hip is unhip. To say "I am cool" is uncool.
posted by Pope Guilty at 1:54 PM on September 22

I've long suspected that there is an inverse relationship between music I like and audio format. If a song sounds great on a transistor radio, it's probably a great song. I'm less concerned with every nuance of the song than I am with the overall song itself. Am I missing the details? Yeah, sure. Who cares? That's probably horrifying to musicians, and I fear I am contributing to The Loudness Problem by my lack of engagement in issues of audio fidelity, but coming from a musical subculture where often the first time you heard your new favorite song was on a hissing cassette that someone recorded tape-to-tape on a boombox, I just can't bring myself to care. If I want to hear a song the way it's "supposed" to sound, I'll go see the band live.
posted by BitterOldPunk at 1:55 PM on September 22

Hipster audiophiles get a lot of FLAC.
posted by geoff. at 1:58 PM on September 22 [12 favorites]

My copy of 'Siamese Dream' by Smashing Pumpkins got something like 400 plays in my crappy car stereo, and eventually the wobble and warmth of the worn-out tape became part of my experience of the album.

A lot of audio effects were originally a result of malfunctioning equipment or equipment operating outside of it's design parameters.

So it's possible that mp3, even if offering less fidelity, might make particular types of music sound better.
posted by empath at 1:58 PM on September 22

If I want to hear a song the way it's "supposed" to sound, I'll go see the band live.

Unless the song was originally constructed in the studio and they can't play it live worth a damn.
posted by empath at 2:00 PM on September 22 [6 favorites]

I don't know who is more annoying, audio nerds or wine nerds.

Oh man, I spent a long-ago New Year's Eve at a party full of both, and wanted to shoot myself by the end of the night.

Being half-deaf has its advantages; I don't have to buy all kinds of expensive equipment because it wouldn't make it any better anyway.
posted by desjardins at 2:03 PM on September 22 [2 favorites]

Unless the song was originally constructed in the studio and they can't play it live worth a damn.

In which case it is unlikely that they'd be a band I'm very interested in in the first place.
posted by BitterOldPunk at 2:04 PM on September 22

I listen more and more to my MP3s, even when other formats are right at hand. It's just easy.

I've been a music fiend since the early seventies. I've heard great music and shitty music, and this I do know: it is possible to make decent sounding MP3s. It takes some experimentation and attention to detail, but it's possible. The biggest problem I have with Mp3s is that they tend to sound "brittle" and "crashy", but even that can be compensated for most of the time. ( I make most of my MP3s at 320, too.)

IMHO the biggest problem with music today, besides horrible production, is going straight digital. Compare old analog recordings ( or even AAD discs) to new DDD stuff. Maybe it's my imagination, but the digital stuff doesn't have the ass that the analog recordings do. It must be an overtones issue.
posted by Benny Andajetz at 2:12 PM on September 22

When it comes to a discussion of quality in water, wine, audio formats, etc., you have to remember that people can be readily separated into four groups:

1) People who are unable to discern the difference
2) People who can tell if they go looking for it
3) People who cannot help but notice the difference
4) People who would like to think (and most espeially would like you to think) that they cannot tolerate the cruder end of that difference.

Some people in groups one and, to a lesser extent, two, automatically put people in the third group in with the fourth group. People in the third group, irritated by this, dismiss the first group as plebs, and the second as cheap and slumming it. The last group are the ones to whom you may sell the $500 one foot audio cable.
posted by adipocere at 2:14 PM on September 22 [1 favorite]

5) People who will perceive a difference whether it's there or not so they can brag endlessly about it on Internet forums.
posted by y6y6y6 at 2:24 PM on September 22 [1 favorite]

Focusing on mp3 vs. CD vs. vinyl is a distraction from the real fidelity problem of music today: compression and the loudness wars. When everything is mixed to within 95% of peak loudness for the entire song, it really doesn't matter what you play it on.
posted by anthill at 2:25 PM on September 22 [2 favorites]

3) People who cannot help but notice the difference

The thing is, for all of the stuff you list (water, wine, audio) there are good reasons to believe that there are very few people in category 3. Instead, most of the people who believe they do fit there are actually in a category you left out:

5) People who think they cannot help but notice the difference but actually can't notice the difference.

Even people who are supposed to be experts, for example, often can't tell the difference in expensive and cheap wines. They rate wine in an expensive bottle higher than wine in a cheaper bottle, even if the wines have been switched without their knowledge. Or if they contain the same wine. And don't get me started on water.
posted by Justinian at 2:30 PM on September 22 [5 favorites]

I remember reading something (web or dead tree, not sure) about audiophiles - folks who spent thousands on speaker wire and hardwood knobs with magical qualities - and there was this one who absolutely refused to listen to live music. Not enough control over the sonic experience or something.

Also- a local AM station (KIDD in Monterey) just switched to a sophisticated oldies format (think Santana, Steely Dan. etc.), and these songs sound great on lo-fi! Just the way we first heard 'em on our cheap radios back in the day.
posted by squalor at 2:41 PM on September 22

There is a difference between $3 wine and $10 wine that even I can discern, and describe. Between any given $10 and any given $100 bottle, all bets are off; I do not have a nose that can pick out a soupçon of garrigue.

Wine skeptics seem unaware that there are several axes on which to asses wine. You don't need a dog's nose to tell if a wine is balanced: not too sweet, not too tannic, not too acid, not too burn-ey. That's what separates many $3 bottles from many $10+ bottles.

The well-known other axis, the subtle notes that we label "leather", "hint of berries", "neiges d'antan", whatever, that's dog's-nose territory, and subject to much malarkey, I think.

Oh, is this an audiophile thread?
posted by everichon at 2:53 PM on September 22 [1 favorite]

This whole discussion reminds me of this young guy who used to be a neighbor of ours. We were talking one day when he mentioned that he'd put a new sound system in his truck. I walked over with him to check it out. He'd hooked up a single speaker, looked like what you used to hang off your car window at the drive in movies, metal case and all. He cranked this thing up and out came the most horrendous garbled mess you could imagine. He was really proud of it though, I think because it was louder than hell.

That was his main concern: that it be loud, and so it was. It was pretty hard to even tell what was playing through the thing, but he was happy with it. So yeah, perception is not to be underestimated (up to a point.) The thing I've noticed about MP3, and I guess this is due to the number of variables that go into producing one, is that a really well made file done at 192 kb/s can sound better than a crappy one done at 320, depending on the hardware configuration, software settings and the care and patience that the person making the thing puts into it.

I've also bought CD versions of some of my favorite albums that ended up having some pretty severe glitches on them. Early on I'd return the sucker for another copy with an identical fault, and it was pretty apparent that the glitchy CD was produced in mass quantities, and that I'd have to shell out for an expensive import to find a decent version.

Also, my hearing has gone to shit in recent years so I'm not as sensitive to the weaknesses that MP3* might present.

Eat it, Young'ns!

(What's that you say there?)

*.ogg's a good format that I wish was more prevalent.
posted by metagnathous at 3:04 PM on September 22 [1 favorite]

adipocre, you left out one group: People who can't tell the difference and have some sort of wierd rage-hate going on for anyone who can. Sort of like people who last read a book in school and now have a chip on their shoulder about anyone who reads for pleasure.
posted by rodgerd at 3:08 PM on September 22

Yeah I was going to say, I was pretty sure this had already been debunked.
posted by delmoi at 3:14 PM on September 22

> It would be much better if audiophiles simply admitted that they liked
> the ritual surrounding playing vinyl records, and the retro sound (hiss,
> crackle, whatever) the turntable makes.
> posted by Pastabagel at 3:35 PM on September 22 [5 favorites +] [!]

I play vinyl because it's what I've got, the outcome of being a Classical freak since about 1975. I couldn't possibly re-buy it all on CD even if I thought CDs had a future, which they don't. And besides, Furtwangler can't remake his stuff for digital due to being dead and all.

I'm ripping it (to FLAC) as fast as I possibly can, though I won't get through all of it unless I live to 120 or so.


> His 'fidelity potential index' (which sounds like something out of an
> eHarmony ad) also says that CDs sound better than vinyl. There's something
> for everybody to get mad about here.
> posted by box at 3:15 PM on September 22 [2 favorites +] [!]

Sounds better than which vinyl? We'll have to be a bit more specific here. Would that be the DGG European pressings of the Bartok string quartets, that I've played five times (twice to transfer to 1/4" tape, twice to cassette, once to PC) or the Zappa disks that I played to death on the turntable instead of copying due to impatience and stonedness and now wish I hadn't, or the Lou Reed record (Metal Machine Music) somebody left on my floor out of its sleeve during a party and I inadvertently stepped on three times?
posted by jfuller at 3:21 PM on September 22 [2 favorites]

The orgasmic, populist rush of deriding experts for making non-existent or blatantly false distinctions is a lot of fun, but the subtext that such distinctions can't exist or are never perceptible is wrong. For instance, here's a link to the study Justinian was maybe talking about (PDF):

Do More Expensive Wines Taste Better? Evidence from a Large Sample of Blind Tastings

As you'd probably expect, the vast majority of the individuals in the study didn't find the expensive wines to be superior to the cheap ones in a blind taste test. Clearly, our sensory experience is easily colored by our preconceptions. Still, that doesn't change the fact that experience promotes the capacity for discernment. In music, for example, listeners without musical training will tend to regard a chord as a monolithic entity, where those with training will be able to pick out its constituent notes (see Bregman, A.S. (1990) Auditory Scene Analysis. The Perceptual Organization of Sound. Cambridge, MA: MIT Press). Those people aren't snobs, they just have the experience necessary to make the distinction between a chord and its individual pitches.

This bears out in the study I linked to, which finds a correlation between price and rating of the wine by those individuals who had wine training. It's pretty easy to see that people who invest in things like this are irremediably deluded, but to heap everybody who claims to be able to detect a difference between, say, a 128kbps bit-rate MP3 file and a lossless audio format into the category of "hoodwinked snob" seems like a pretty clear case of normalizing your own perceptions.
posted by invitapriore at 3:24 PM on September 22 [1 favorite]

I want to add that John Meyer is pretty clearly not one of those experts, and I agree with everybody upthread who's said that he's assessing MP3s on entirely the wrong criteria.
posted by invitapriore at 3:25 PM on September 22

Sounds better than which vinyl? We'll have to be a bit more specific here. Would that be the DGG European pressings of the Bartok string quartets, that I've played five times (twice to transfer to 1/4" tape, twice to cassette, once to PC) or the Zappa disks that I played to death on the turntable instead of copying due to impatience and stonedness and now wish I hadn't, or the Lou Reed record (Metal Machine Music ) somebody left on my floor out of its sleeve during a party and I inadvertently stepped on three times?

jfuller, that was already tried by Christian Marclay in 1989.
posted by metagnathous at 3:28 PM on September 22

I mean, the numbers he has are interesting on some _level_, and it's cool he made the chart. but trying to interpose the notion that these metrics even remotely represent some sort of human-appreciable "fidelity" or "quality" level is ludicrous.
posted by blenderfish at 3:29 PM on September 22

The flip side of the fidelity thing is a bizarre rating system I have for music. In ascending order of quality, if it sounds good when:

1) Performed by the band live
2) Heard over a crap AM radio station
3) Performed by the band live, twenty years later, after the singer has blown most of his pipes out and the drummer has a wicked barbituate habit
4) Performed as some else's cover (and it is still a hit)
5) String quartet version of the tune, when played through a twenty-five year old paper speaker at the top of an elevator
6) Performed by a band of slightly-older kids for some other kid's bar mitzvah
7) The MIDI version, embedded in someone's webpage, played through your ancient 16-bit off-brand audio card through battered speakers you dug out of the Dumpster

... then that is a quality tune. I think it's survival of the fittest.

But really, there are some people in the third group I mentioned earlier. It isn't like they are mutants or anything. Some can even do it in blind taste tests. I'm not fussy about water. Sometimes (and this is weird) I accidentally let water sit out in a glass for a few days and I'll drink it anyway. Yes, I can tell that it's sat out, but that's fine. Personally, I don't get how people cannot tell the difference between Coke and Pepsi, but I accept that some people just don't, just like I know that my mom always had crap color sense and I'd end up being the one who had to match the carpet to other things in whatever new apartment we landed at.

I'd love to see a study done on the psychoacoustics. I've often wondered, are the frequency ranges we "miss" the same for everyone, or does that vary by person, and, if so, how much? Would there be a correlation between people who varied from the usual psychoacoustic profile and their enjoyment of MP3s? Would listening to a lot of MP3s perhaps make someone pay less attention to the missed frequency bands, and how could you separate the chicken from the egg there?
posted by adipocere at 3:49 PM on September 22

the Lou Reed record (Metal Machine Music) somebody left on my floor out of its sleeve during a party and I inadvertently stepped on three times?

I do believe that is the official audiophile procedure for optimizing the sound quality of MMM.
posted by mr_roboto at 4:03 PM on September 22 [3 favorites]

Also, to be fair to the guy, I think the OP is a bit of an exaggeration of his claims. (He seems to be claiming that some sort of theoretical 'maximum' quantity is lower for MP3s than for CDs, which isn't _exactly_ the same thing as saying "I have proven that CD > MP3."
posted by blenderfish at 4:05 PM on September 22

Even ignoring compression, one other flaw in the 'fidelity' metric would seem to be that a 44Khz 16-bit recording, a 1-bit 705Khz recording, and a 64-bit 11Khz recording all have the same 'bits per second' (and therefore his 'maximum fidelity' would be the same,) yet the last two would sound dreadful.

The 705Khz 1-bit would be poor because you couldn't hear any frequencies above around 22Khz, and the ones you could hear would have a horrible dynamic range.

The 11Khz 64-bit would be poor because it wouldn't be able to (usefully) reproduce frequencies above 5.5Khz.
posted by blenderfish at 4:46 PM on September 22

Wine skeptics seem unaware that there are several axes on which to asses wine. You don't need a dog's nose to tell if a wine is balanced: not too sweet, not too tannic, not too acid, not too burn-ey. That's what separates many $3 bottles from many $10+ bottles. The well-known other axis, the subtle notes that we label "leather", "hint of berries", "neiges d'antan", whatever, that's dog's-nose territory, and subject to much malarkey, I think.

But even a novice wine drinker can make the easy distinctions; you admit that yourself. It is the more subtle notes that are supposed to mark an expert and there is little evidence that the vast majority of people can distinguish anything meaningful on that axis. This is true of water, it is true of wine, and it is true of audiophile crap. Even among those that can make the very subtle distinctions, it isn't clear how the correlation works. If 95% of tasters prefer a "worse" wine and only 5% prefer a "better" wine, are the words "better" and "worse" meaningful in this context?
posted by Justinian at 4:54 PM on September 22

Depends on whether or not you think "better" and "worse" are meaningful in context to, say, McDonald's.
posted by adipocere at 4:55 PM on September 22

I'm not sure that holds true: Would a McDonald's hamburger beat an In-n-Out burger in a blind taste test for the great majority of people? Would Chicken McNuggets beat a well prepare oven roasted blah blah whatever in a blind taste test? I tend to believe they would not. Bad wines, on the other hand, tend to beat good wines in blind taste tests.

Note also that New Coke greatly outperformed Coke Classic in blind taste tests. So I'm not being some sort of anti-wine-snobbery-snob. This phenomenon is widespread.
posted by Justinian at 4:58 PM on September 22

If 95% of tasters prefer a "worse" wine and only 5% prefer a "better" wine, are the words "better" and "worse" meaningful in this context?

95% of people probably prefer Dan Brown over Thomas Pynchon, too.
posted by empath at 5:17 PM on September 22

I think some people do have a more refined palate, and to them it's worth spending the money and time on finding wine that appeals to them. I don't think it necessarily means that the wine they prefer is objectively 'better', it just means that they prefer it.

I think that applies to pretty much anything that people can become enthusiasts of -- cars, sound equipment, music, literature. For the vast majority of people who can't tell the difference, it doesn't matter. For the small percentage of people who take the time to train themselves to be able to spot quality, it does.

Though I happen to think most 'audiophiles' are probably kidding themselves, i'm willing to believe that some people really do know what they're talking about. Some sound equipment must sound better than other sound equipment, and some people must be able to tell the difference. I just don't think that everyone who claims that they can tell the difference really can.
posted by empath at 5:25 PM on September 22

And to get all philosophical for a second:

If the point of listening to a song is just the abstract idea of the melody, vocals and words, then the reproduction quality doesn't really matter very much, because it's all happening in your head anyway, as long as you can pick out the melody, understand the words and hum along. Some people actually do just enjoy music on this level and are just as happy listening to a tune on a shitty AM radio as on a great sound system.

If music is the actual particular performance of the musical idea, than the sound quality actually makes a great deal of difference, and the production quality makes a difference. In this case, what's important is the psychoacoustic space -- your internal model of what's creating the sound and where it's coming from -- and the higher fidelity and more realism the better. It's important that you have correct stereo spread, that the dynamics aren't flattened out, that all frequencies are reproduced well, etc, or else some of the intended affect of the performance/production is lost.

I do think there's a point of diminishing returns here, but I can understand how people can be annoyed if you know how a song is SUPPOSED to sound and something is missing. It depends on how much you get into the music and how well you're able to place everything into an imagined space while you're listening.
posted by empath at 5:36 PM on September 22 [2 favorites]

Umder his methodology, I can't see why an MP3 shouldn't have the sane equivalent bit rate number as a CD. It's still storing 16 bits of dynamic range and 44.1KHz of samples, albeit stored in a lossy way.

And it he had Minidisc on there it'd be 20 bits * 44.1KHz - and therefore better than CD - because ATRAC has 20 bits worth of dynamic range available.
posted by cillit bang at 5:40 PM on September 22

I use crappy ear buds. I hate them.

But what else can I wear on the subway without looking ridiculous?
posted by rokusan at 5:43 PM on September 22

Can't resist contributing to the wine derail. I probably know less about wine than you can possibly imagine, but it's more a case of knowing what I don't know. I've sold wine, drunk wine, conducted wine tastings, and taught wine classes to waitstaff. And I pretty much don't know a goddamn thing about the stuff. BUT, and this is where I differ from Justinian, I think it's less that wine experts are able to distinguish greater complexity (though of course that must be true to some degree) and more that people who haven't had experience with wine are unable to DESCRIBE that complexity. Wine terminology, and the pseudo-scientific mystical hoodoo that sadly accompanies much wine service in the States, is obscure and off-putting to most folks, who are perfectly capable of describing wine accurately if they learn the terminology and learn to associate the taste on their tongue with the term. I gave a wine class to a group of servers once and part of the class was a blind taste of the cheap plonk I was gonna serve as a house pour. This is not a "nice" restaurant. It's a locally-owned erstwhile-Applebee's. I had three different reds at about the same wholesale price and I figured I'd let the servers decide which one they liked-- they're the ones selling the stuff, right? I polled them and asked them to write down their impressions of each wine. One waitress wrote,"I liked this one the best. It tasted purple." And it did. It DID taste purple. It was a fruity, way-too-sweet, soft cheap caberent, and when I put it on the menu, it was a hit. Terrible stuff. But that waitress could enthusiastically tell every table of diners sitting there wolfing down mozzarella sticks with ranch dressing that she had an awesome purple winE she just knew they would love, and people lapped it up.
posted by BitterOldPunk at 5:50 PM on September 22 [2 favorites]

> I'd love to see a study done on the psychoacoustics. I've often wondered, are the
> frequency ranges we "miss" the same for everyone, or does that vary by person,
> and, if so, how much?

That work was actually done in the 1930s and 40s, by S. S. Stevens at Harvard among others. It varies by person and (unsuprisingly) by age--your upper-range sensitivity tapers off a bit as you get older, not by the same amount for everybody. See the Handbook of Experimental Psychology, sensation and perception section. Amazon wants twelve hundred bucks for it so there probably aren't too many customers for it who aren't departments of experimental psychology or purchasing librarians. Wiley has made the newest edition available online. (Sort-of-available. You click through several drill-down browsing pages before finding out it's not free but at least you can buy single articles instead of the whole monster at one go)
posted by jfuller at 6:13 PM on September 22 [1 favorite]

> But what else can I wear on the subway without looking ridiculous?
> posted by rokusan at 8:43 PM on September 22 [+] [!]

To paraphrase T. E. Lawrence, the trick to being a geek is not to care that you look ridiculous.
posted by jfuller at 6:24 PM on September 22

rokusan: Etymotic hf2. $120 bucks gets you awesome results without looking like a total douchebag (also lowers likelihood of your stuff being stolen, as it's incredibly nondescript, unlike white ipod "STEAL ME"). In-ear noise-canceling is the way to go, much better bang for your buck and results than moron audiophiles plunking down $10,000 for bigass speakers.
posted by amuseDetachment at 6:25 PM on September 22

From a subjectivist standpoint, isn't it illogical to say that people think the worse one sounds better? Whatever people think is better, is better, in their world.

posted by HotPants at 12:20 PM on September 22

Doesn't that last statement pretty much fly in the face of any notion of qualitative analysis? What if I encounter exhibit A early on and decide I like it, but later encounter many other similar exhibits and in hindsight decide that my favoring of exhibit A was simply due to lack of exposure? More often than not our opinions are based as much on ignorance as they are on facts... but if you're saying that adding to those facts and reevaluating your opinion afterward leads to a merely different opinion - one neither better nor worse than the previous one - I have to call bullshit on that one.

We can harp on straw man audiophiles all we want but until someone goes out and proves that people objectively get more enjoyment listening to music out of tin can ear buds, regardless of how diverse a sonic palette they've previously been exposed to (ie. never), then I think most of us are just indulging in some good old fashioned, Harrison Bergeron-esque bring-em-down-to-our-level snarking.
posted by squeakyfromme at 9:13 PM on September 22

My copy of 'Siamese Dream' by Smashing Pumpkins got something like 400 plays in my crappy car stereo, and eventually the wobble and warmth of the worn-out tape became part of my experience of the album. Similarly, my LPs had a familiar pattern of crackles and pops that were there every time I played them, and when I hear the same music without it now, it feels wrong, even though it's 'objectively' better sounding without the noise.

I agree that one of the fundamental appeals of music is that it provides nostalgic touchstones (touch tones?) to our past, but I'd also counter that your "experiential" argument for how we enjoy the music we do reduces it to not much more than that. Not that I'm saying it's an invalid approach but it does seem to ignore alternative, equally valid values that others may get out of that same music.

That said, I do think there's cause for concern in the music industry but I agree with others in that the so called "loudness war" is the main issue to worry about. If the entire industry decides that recording all of their artists as if they were a Burger King commercial is the way to stay competitive in a declining market, that may not adversely affect those who never listen to music on anything but an iPod but it doesn't leave those who prefer a more nuanced listening experience any choice in the matter.

I would also argue that if your barometer for whether a song is good or not is whether it sounds good on a cheap pair of 6x9's then you are merely arguing for particular genres or styles and not really accounting for music as a whole. Certainly garage rock and 50s pop sound great on cheap sound systems - technical limitations of the time precluded them from being mixed for much of anything else - and there's a lot of that loud, brassy top 40 radio that sounds fine on a car stereo, not to mention a huge swath of modern indie rock that is typically recorded on a shoestring budget to begin with... but then there's a lot of electronic and hip hop music that definitely benefits from better speakers, and I think it would be outright ludicrous to insist that all classical or jazz that doesn't sound just as good blaring out of a clock radio speaker is not legit in some way.
posted by squeakyfromme at 9:52 PM on September 22

OK, objective proof:

While any number of devices used to play music might be better than an MP3 player and a $30 pair of computer speakers, the MP3 player and the $30 pair of speakers has infinitely better sound quality than the 2-8° refrigerator, big ass centrifuge and the auto-sampler that was specially designed to make a god-awful whine whenever it does it's little thing.

Try as I might, I can not fit 50 lbs of vinyl, a turntable, tube amplifier and decent speakers into my little man purse. Bumping into something and breaking one of the tubes on your amp tends to reduce one's enjoyment - as does picking the little shards of glass out of your leg. I can carry an MP3 player with me and liberally scatter sets of computer speakers throughout my life.

Final analysis - # of hours I spend in a fortress of solitude-esque audio citadel (0) + feasibility of taking high end audio equipment with me where I go (0) + pleasure derived from no music when I'm not in a fortress of solitude-esque audio citadel (0). Total enjoyment derived from listening to MP3's on my crappy little setup > 0. Q.E.D. and stuff.
posted by Kid Charlemagne at 9:53 PM on September 22

Yeah, but Kid Charlemagne, who is arguing that there aren't any practical reasons why people may listen to most or all of their music on MP3 players? It's really the opposite that seems to be the case if anything: the MP3/iPod advocates are attempting to force not just the audiophiles but anyone that has better than cut rate speakers into a defensive position... like there's no difference anyway so give up the charade.

I mean, if you're listening to music at home there's no reason why you can't get a decent pair of speakers with a subwoofer for less than $50, but if you're more of a mobile MP3 jockey, hey, do your thing... I just don't understand what it is that has the latter group up in arms when others want to advocate for better sounding songs. Are you afraid your free MP3s are going to be too expensive if Radiohead spends a million bucks on their record as opposed to taking it to the garage and hitting record on the 80s-era boom box?
posted by squeakyfromme at 10:10 PM on September 22

"Well, now I feel justified lugging my turntable, speakers and a portable generator on a cart behind me as I run. Still can't figure out how to stop the records from skipping though...

"I don't know what any of those words mean."

I can help with some of those:

A Cart is a trade lingo for a Fidelipac cartridge used in radio.
A Speaker is the presiding officer in a legislative assembly. I guess speakers would be a pair of such individuals.
Records are individual lines in a database.
Skipping is where one rhythmically jumps over a swung rope.
Lugging is when you run your truck engine at too low a speed to maintain smooth even rotation.
A turntable is what crowded rail yards used to turn steam locomotives.
Running is some sort of self induced torture commonly performed in the big blue room.
A generator is a device that turns mechanical motion into electricity.

Even knowing those thing I can't really parse the sentence. Something about if being ok to slowly turn locomotives if there is more than one speaker on board being tortured while recording the whole thing for radio however that will still cause the database to jump over the rope. I guess the generator is attached to the tuntable to provide power for the cart machine. Loony I know.
posted by Mitheral at 10:33 PM on September 22 [7 favorites]

Quality come always after convenience. Getting non-DRM mp3s are far more convenient than any CD or Vinyl. Lossless (I'm a FLAC fan) is a better choice if you need a good back-up.
posted by zouhair at 4:32 PM on September 23

"I have a chart! Listen to me!"
posted by chairface at 8:02 PM on September 23