h1

Audio fidelity is overrated

April 8, 2010

For most of us, more audio fidelity isn’t better. It’s a bar. And above that bar, you’re fine.

My car is old-school enough that it doesn’t have an MP3 input, which means I listen to everything on CD. That means I’m regularly swapping between purchased CDs (LPCM audio) and burned MP3s – it’s not quite an A-B comparison, but it’s close. Could I tell the difference if I was sitting quietly in a soundproofed room? Maybe. Can I tell the difference over engine and traffic noise on my factory-installed car stereo? Not a chance.

In general, advances in reproduction of music have been about making it more accessible, not about making it sound better. From live musicians to player pianos, from record players to iPods, most consumer-oriented music technology has led to music being available to a wider range of people, in a wider range of environments. And Paul Lamere made a related point in a discussion earlier this week: “The audio fidelity you can buy for $100 today is a lot higher than what $100 would have gotten you thirty years ago.”

Note that I can think of two counterexamples to this general progression of increased access and lower quality. One was deliberate: the move from AM to FM radio, which sounds much better but generally has a  shorter range. The second counterexample is really more of a side effect: the move from vinyl to CD, which was clearly mandated by the convenience of the shiny little discs. I know that many people argue that vinyl sounds better than CDs. I’ve never done the comparison myself (although I’m sure that many people reading this post have), but I can readily believe that a pristine LP on an expensive system would sound better than a CD. But in the real world? I have CDs that I’ve been listening to regularly for well over a decade, and they sound as good as they did when I first ripped off the cellophane. I doubt that would be true for a record. Digital fidelity is not to be sneezed at.

So this idea of being prescriptive about audio fidelity—”Oh noes! You’re storing all your music at MP3s! You’ll regret it!”—doesn’t seem to be in line with what people actually do with music, which is to readily trade fidelity for accessibility. Kryder’s Law being what it is, it’s increasingly possible to store lossless versions of music on your hard drive—but how many people will take advantage of more hard drive space to simply store more songs? I love music, but I have no illusions about being an audiophile. Based on the overwhelming evidence, I’m not alone.

The open question remains: how low can you go? A friend of mine finds satellite radio intolerable because of the high degree of audio compression low bitrate [see EDIT, below] (many people are oblivious, including me, although I do find it intolerable that there are 200 channels and barely anything worth listening to). As streaming becomes an increasingly viable alternative to downloading, is something similar going to happen with audio on mobile devices?

Feel free to flame me for being a audio Philistine in the comments. Or just share what you think.

EDIT: Mike corrected me in the comments, below: satellite radios use a lower bitrate, not a higher degree of compression.

Image: MP3 vs CD quality (PCM) by Flickr user filicudi, used here under its Creative Commons license.

11 comments

  1. I’ve always pretended fidelity matters to me. And for the most part it does. I think people care about fidelity to the extent that they notice it. But I think the big factor is that the difference in “kinda bad” to “really bad” fidelity is a lot easier to notice to the average person than “kinda good” to “really good.” Once music starts sounding good, most people won’t notice when you increase the fidelity, either by a little or by a lot.

    But as much as I occasionally act like an audiophile, there are two things seriously preventing me from ever becoming one.

    1: Budget. My “sound system” is high enough for my standards but I know I’m not getting the best possible sound out of anything (especially tapes and vinyl). If I had the money, I would definitely drop it on some high end speakers and a sweet as turntable but that day may never come…

    2: I’m 80% deaf in my right ear (and my left ear ain’t perfect either). I think this kinda rules out any chance I had at becoming a fidelity snob.

    In regards to streaming and “how low can you go,” I think society has made the change slowly enough that most people aren’t going to care. The fidelity has decreased slightly with each new audio format (and audio output devices), so that something like streaming music through a phone will hardly be a noticeable difference to most people. And if they do notice, I feel like they’ll have been conditioned by their iPod earbuds to not really care.


  2. I absolutely agree with you about the diminishing returns for increased audio quality (both in terms of what you can hear and what it costs). And yeah, that we don’t notice the difference between ‘kinda good’ and ‘really good’ is what I mean when I say there’s a bar, and you just need to be above it.

    I also worry that we are gradually lowering our standards for fidelity. But cassette tapes still sound terrible to me, and everyone can instantly recognize AM radio as sounding like crap (and remember, back in the day everyone thought they sounded fine). So I’m not sure I buy the argument that our concern for fidelity is diminished.


  3. Your title is completely accurate. Audio fidelity in this day and age is overrated. Am I happy about this? Not at all.

    The producer/engineer side of me of read this post and I had a hard time letting that statement go and not disagree with it.

    But I own an iPod and an iPhone. And I listen to music both on a regular basis happily. I have a library of music on my desktop that is ripped from CD to mp3, AAC 256 to be exact. (I can’t get into FLAC) So I can’t disagree with the point of what conditions we listen to music in. I agree you will not notice the difference between an mp3 or wav file in your day to day listening environments.

    Very quick side note. I LOVE the sound of AM radio. It is has coloration that is somewhat relatable to analog tape. Moving on.

    Audio fidelity is NOT overrated in my eyes. Vinyl is becoming cool and hip. I love this. I buy vinyl for many reasons. One is that I have an original mono pressing of Pet Sounds that is 40+ years old. IT STILL SOUNDS AMAZING after all of those years. CD’s will never last this way. I have burnt CD’s and even with my anal care have found ways to have them skip. That is like fingernails on a chalkboard to me.

    I don’t have the perfect listening environment for vinyl but I do for digital music. I have studio monitors that I run tests on monthly to make sure they are perfectly in phase and in balance. The reason I went to AAC 256 was after hours of A/B ing the difference between this compression and the lossless format of wav. I notice almost no difference. But with vinyl it’s more of an experience than a question of audiophile rants.

    I now realize I came to no point here. But audio compression used improperly SUCKS. Post about it if so inclined: http://dysonsound.com/2010/01/audio-compression-is-a-killer/

    Thanks for making me ramble and make no point at all debcha :)


  4. You’re always welcome to come and share your interesting rambles, dirkler.

    A couple of points:

    Burned CDs aren’t the same as purchased CDs. Off the top of my head, purchased CDs are pressed, with actual pits in aluminum. Burned CDs use lasers to make dark spots in resin (hence the name). That material is, almost by definition, not going to be archival. But the actual MP3 can last indefinitely, incorruptible.

    And don’t get me wrong. I don’t have a problem with vinyl. I love the vinyl experience. Or rather, I love the idea of the vinyl experience. I recently came across a track I knew I wanted dedicated listening time for, and it took me four days to find 20 minutes to sit still with headphones on. Frankly, I’d much rather use my limited dedicated listening time to go out to live shows – still the ultimate high fidelity experience.


    • Yes to everything but your last sentence.
      Live shows are amazing. They can make or break an artist for most people. (including me)

      But NO WAY are they the ultimate high fidelity experience. Live sound engineers do the best they can but face ultimate doom when mixing live shows. It is a nightmare of a job. The only time you ever notice a live sound engineer is when he/she does a BAD job. If you can hear the vocals, it’s considered a success. That is such a low bar to hold against these engineers but 95% of the crowd will only notice the lack of vocals or potential bits of feedback.

      As far as the fidelity topic goes. We have to leave live shows out of this. Those are a full body experience. Nothing will replace seeing a band live. No matter how piss poor the drunk guy doing sound is that night.


  5. Aargh – you’re absolutely right, and yes, I can absolutely hear the sound mix (for better or for worse). But as far as dedicated listening experiences go, I’ll still plump for live music as often as I can.

    I think that my favourite audio experience ever was lying on the floor underneath a grand piano while a concert pianist played. My second favourite was dancing all night to techno on the superb sound system at Watergate, in Berlin. While the sound in both cases was utterly amazing, it’s impossible to separate the audio quality from the experience itself.


  6. I think your sentiment is totally right. I don’t care at all unless it sounds terrible to me. That said, two examples of terrible to me are Myspace streaming rate and satellite radio. I literally can’t even begin to comprehend how people stand either one, particularly satellite (40 kbps or so), but most people don’t seem to notice. I guess I challenge the idea that people don’t notice. I think there may be a tendency to like music less when heard via low-bitrate streams like satellite radio and Myspace. It’s sad for artists in a way. So much time is spent to make music sound good and to arrange songs to convey specific ideas and then they end up 40 kbps over satellite radio (if you’re lucky enough to get airplay). It inarguably detracts from the intended experience. That said, we all push for our songs to get played whenever and wherever possible.

    I won’t name the band, but I listened to a well-reputed band via Myspace once and thought they sounded terrible. I would have immediately dismissed them, but I knew the producer/engineer combo they used would never let anything out the door that was so awful sounding. So, right then, I switched to their Reverbnation stream and it was like hearing a totally different song. I am very song-focused and listen to a lot of very lo-fi stuff, but in this case, the quality loss was sufficient to alter my categorical reaction. (I am perhaps unfairly prone to like/dislike categorical reactions with music these days.) This resulted in my immediate removal of the Myspace player from all my pages in favor of replacing it with a Reverbnation embed.

    That is my long-winded way of saying that maybe you don’t notice it sounding “bad,” but that you do have a reduced likelihood of liking something that has been sonically damaged. I don’t think high-bitrate MP3 vs. CD makes much of a difference on this front, but I do think that satellite radio does suffer here…or rather, the artists suffer.


    • That’s an interesting point, Mike – that you don’t notice the audio quality per se, you just notice that you don’t like the song (I shudder to think of what it’d take to put that to the test!)


  7. There’s a nifty article/survey to determine whether listeners could determine the difference between MP3s with different data encoding rates: Few listeners can distinguish between “average” and “best” MP3 samples:

    http://scienceblogs.com/cognitivedaily/2007/11/few_listeners_can_distinguish.php

    The results being that most people cannot tell the difference between 128 and 256 bit encoded mp3s.


    • Thanks for the link to the study, Paul! We did a similar poll here at z=z on the difference between 128 and 320 kbps (inspired by your post, in fact). It looks like it’s running at about 2:1 correctly identifying the higher bitrate but, of course, it’s not remotely scientific (I would expect some pretty serious selection bias). Mike Epstein and I have talked about investigating it ‘for real’ – I think it’s one of many projects both of us would like to do that’s taking a back seat to the rest of our research.


  8. […] posts to get you started: Lady Gaga vs other artists: a graph; Audio fidelity is overrated; Why do we love the songs we love?; What will music fans pay […]



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

%d bloggers like this: