DANKOSKY: I think our caller Bill, though, from Indiana maybe gets it, one of the points here between vinyl and CD. Bill, go ahead.
BILL: Yes. I wanted to agree with the – at least I believe the point that the first lady was talking about, that although the sound of CDs and MP3s and that sort of thing is certainly better, there’s a whole ritualistic quality of taking a vinyl record out and placing it on the turntable, using a little device to clean the dust off of it, setting the needle on, watching a little stroboscope to make sure it’s turning at exactly 33-and-a-third revolutions a minute that I miss with the new technology that you can just pop in or turn on a button or whatever, and all of a sudden you’re listening to it. And I just – the incorporation of your other senses and everything is just something that I really miss with the new technology.
DANKOSKY: Yeah, that smell of taking the record out of a jacket the very first time. I know what you’re talking about, Bill.
BILL: Oh, absolutely.
(SOUNDBITE OF LAUGHTER)
METCALFE: John, you know, there’s a couple of things that come to mind with that. One is that I completely agree, and I’m not preaching that one format is really better than the other, because I would never tell an artist, no, you shouldn’t paint on canvas because I think paper looks better. So there’s that extent. There’s also – we have lost the ritual, and we’ve lost the experience of sitting down and focusing on just a recording. What we’ve gained, in a way – and I’m not saying one’s necessarily that much better than the other – is most people walk around with a huge collection of music on their belt or in their jacket pocket.
And I think we probably consume more music now than we ever did. So there’s a little bit of a tradeoff there. We can still make room for the experience if that’s our interest.
DANKOSKY: Sean Olive, how about this perception that may change the way we feel about how one piece of music sounds over another?
OLIVE: Right. Well, there’s a number of factors involved in our perception of sound quality, and a lot of them have nothing to do with the sound itself. So in the research that we do, we’ve looked at the – and we call these things nuisance variables or biases. So one of the ones that we deal with is psychological nuisance variables, which have to do with your knowledge or expectation of what you’re hearing. Thomas Edison knew this a hundred years ago. He said people will hear what you tell them to hear. So if you’re aware of – that you’re listening to a violin, you’ve been told it sounds great, if it costs more money, all of those factors play into your perception of how it sounds.
DANKOSKY: And I can only imagine that in your line of work, sometimes people will spend a whole lot of money on a set of speakers, and they’ll think those speakers have to sound much better because they just cost me a lot of money.
OLIVE: Right. And often it’s a case where the price has very little to do with the sound quality.
DANKOSKY: Is that really true that you can’t really buy your speakers on a price point?
OLIVE: No. I mean, we have many examples where we do double blind tests, and some of the competitors are very expensive, but they don’t sound very good because I think it’s a matter of, you know, the more expensive the speaker is the fewer you sell. So your R and D budget is much smaller, and you just don’t have the money to spend on good measurements, for example.
DANKOSKY: Let’s go to Brian in Ann Arbor, Michigan. Hi, Brian.
BRIAN: Hi there. I appreciate the discussion on the CD versus LP, but wanted to add into the mix CD versus MP3 and AAC, et cetera, because we’re losing a lot of information when we go to that next level. And it’s kind of sad that that’s the way we consume music now.
DANKOSKY: And that’s exactly what we want to get to. So, Scott Metcalfe, what happens here? We’re talking MP3s and different types of MP3s. Maybe you can explain the differences between these various formats.
METCALFE: Sure. Well, the – I was referring to before with the portability of music now. That’s the upshot of MP3s and AACs. And even in my work, when I’m working with an artist who’s not nearby, it’s very convenient to put a demonstration of a mix or an edit, into a compressed format like that that I can put on to an FTP server, and they can download and audition. But there is definitely a quality difference. Just today, I put an actual CD in the FedEx to a client that I’m working with so he can hear exactly what it sounds like, not like what, you know, what we are working with online.
There is loss – there’s a loss of depth of field in a smaller format. Just to say, though, MP3 and AAC excludes a lot of details. You can have very low-resolution MP3s that offer you really small file sizes but a pretty big hit in sound quality. And you can get bigger MP3s or what we would call the kind of the next generation of MP3, which is the AAC, which most people know as the format of iTunes. The equivalent AAC to MP3, to my ears, does sound better. And when I get into the higher ranges like 190, 320 kilobits per second, those are pretty respectable sounding.
And I’ll use those, you know, as audition levels. The types of things I hear in the MP3s that I don’t hear in the original, excuse me, in particular things like drums. A snare drum tends to sound more like noise to me. The rattle of the snares loses its definition. The stereophony of the sound gets a little bit collapsed. You know, if I’m listening in the car and there’s a lot of noise there, I’m not terribly concerned about it. I go into my home to listen.
I listen in the recording studio, and it’s blatantly obvious to me. Or occasionally, I’ll hear somebody playing, you know, through a PA system at a party or, you know, a reception or something from an MP3, and it’s almost painful for me to listen to. And I think if we did a quick comparison there, people would know what I’m listening to.
DANKOSKY: Well, we have some comparisons that you sent along to us, and we want to listen to those now. We have an original recording of a rock band that you made. And maybe you can just tell us about – when you say an original, how is this made? What are we listening to, this first track?
METCALFE: Yeah. This is a band called Bronze Radio Return from Hartford, Connecticut, and we – Chris Henderson, who’s the lead singer of the group and songwriter, was a student of mine back at The Harrt School in Connecticut. And I had brought him down to – into a class that I was teaching down at Yale in the sound design, Yale School of Drama sound design department. And we brought the band and we recorded the class project. And so that’s the material I decided to use. And I tested this on my own just to see what we would get. And immediately, I picked up on the cymbals and the snare drum having kind of this distortion noise quality to them that…
DANKOSKY: Yeah.
METCALFE: …I didn’t pick up in the original.
DANKOSKY: Well, let’s listen to a little bit of this original recording first here.
(SOUNDBITE OF SONG, “OVER YOUR HEAD”)
CHRIS HENDERSON: (Singing) …this place because the water is over your head. It’s over your head.
DANKOSKY: Now, we’re going to play a little bit of an 8-bit recording, and this is what you would consider to be almost like an old-style MP3. And we – I think most people should be able to hear this even on their car stereos, the difference here.
(SOUNDBITE OF SONG, “OVER YOUR HEAD”)
HENDERSON: (Singing) …this place because the water is over your head.
DANKOSKY: So what are we losing in there, Scott Metcalfe?
METCALFE: Sure. Well, first of all, that wasn’t really a format that people downloaded. The reason I put that in there is that it illustrates what the difference is or what is being lost with bit rate and sample rate. Sample rate – the higher the sample rate, the higher audio bandwidth you’re able to capture. And most people would say that, on average, human beings hear up to about 20,000 cycles, wave cycles per second. So a CD is going to capture sound – not capture sound, but reproduce sound of 44,100 samples per second.
And we have a little formula that says if you divide that in half, whatever our sample rate is, we divide that in half, and that’s what we’re going to be able to capture. So the recording you just played was a 12 kilohertz sample rate. So therefore, we are only hearing up to six kilohertz. So we are missing a lot of overtones, a lot of content in there. And the 8-bit just means there’s – essentially, there’s quite a bit more noise. There is a lot of information that was lost there.
DANKOSKY: We’re going to hear this original recording that you made. Once again, this higher fidelity recording. Let’s listen for a moment here.
(SOUNDBITE OF SONG, “OVER YOUR HEAD”)
HENDERSON: (Singing) …this place because the water is over your head.
DANKOSKY: And now, we’re going to go to an MP3, a more traditional MP3 recording. The type that you might download now.
(SOUNDBITE OF SONG, “OVER YOUR HEAD”)
HENDERSON: (Singing) ….this place because the water is over your head.
DANKOSKY: Now, I say Scott, I can’t hear really any different through my headphones and probably, at home you can’t hear any difference. But you sent us an interesting comparison here, and the next thing that we’re going to hear is what’s missing. What’s essentially missing between the original and the MP3. Let’s listen to this.
(SOUNDBITE OF SONG, “OVER YOUR HEAD”)
HENDERSON: (Singing) …this place because the water is over your head.
DANKOSKY: So explain this, Scott. The data we’re hearing there is what? It’s just stuff that’s out of the original recording to make this MP3, to compress it if you will?
METCALFE: Yeah. This is a fun little trick that we do when we want to really hear the difference between two pieces of audio depending on what kind of process we’re applying to it. And what I did was took the excerpt, the original excerpt, and I inverted the polarity, which is a technical term for anything that was the positive side of the waveform is now on the bottom, is negative and vice versa.
So when you sum them together, if they’re the same exact thing, you hear absolutely nothing. If you hear anything at all then you’re hearing what’s different between those two recordings. So this is just a cheap and simple way of hearing what was lost. And, you know, of course, my job requires me to listen really critically, but – and I wasn’t sure how much I was going to get over the headphones here over the ISDN line. But I could clearly hear a lack of low frequency in the MP3 and just the real thinness to the rest of the texture. I’m not sure how much is that lost over the FM broadcast.
DANKOSKY: Well, I think an awful lot is lost. I’ll just say quickly that I’m John Dankosky, and this is SCIENCE FRIDAY from NPR. Now, Sean Olive, younger people listen mostly to MP3s. Do you think that this affects their preference in audio speakers? Do they have an appreciation for higher fidelity sound anymore?
OLIVE: Well, that’s an interesting question because in the last couple of years there’s been a lot of press saying that kid, you know, we’re approaching the death of fidelity because of the younger generations have been raised on MP3 in low-quality earbuds. And that they now, you know, may prefer that over high-quality sound.
So about a year ago, we did a small study where we brought in high school students and more recently college students and we asked them to listen to MP3 at a low bit rate, 128 kilobits per second, versus CD. And, of course, it was a blind test, so they weren’t aware of which was which. And what we found out was that in most cases, they prefer the CD over the MP3. And then we random through some speaker tests, where they heard four different loudspeakers that varied from one that was very accurate and neutral to one that was quite the opposite, and quite surprisingly they prefer the most accurate neutral loudspeaker.
DANKOSKY: Now, has your industry had to change it all with the idea though that because the music industry is making MP3s as the commodity, do you have to do different things in order to make a lower resolution piece of music sound good through a set of speakers?
OLIVE: No. We basically designed our speakers so that they will sound good if you play good sound and music through them. We don’t try to compensate for – in the loudspeaker system for deficiencies in the MP3 or vinyl. We are working out some algorithms that will try to improve low-quality MP3 so it sounds closer CD quality.
DANKOSKY: I want to get to Zack(ph) in San Francisco. Hey, Zack.
ZACK: Hi. You’re actually speaking to my question right now. As you’re talking about formats on how to use the CDs and MP3, it seems to me that a lot of the newer music coming out is more manipulated and actually made to be heard on earbuds or actually on cell phones. It’s more distorted or twangier(ph) or treblier(ph) , if you will.
DANKOSKY: Scott, what do you say? Is that true, Scott Metcalfe?
METCALFE: I’m not sure how much how that happens, but it is another place where, you know, years ago the litmus test was always, let’s go play this in the car and see what is sounds like. If it sounds good there, it’s going to sound good everywhere. I do know a lot of people – and I’ve done this myself – who will listen on earbuds since that’s going to be the most common way that the music is going to be listened to. You know, it brings up an interesting point though is that as people gets faster connections at home and storage on phones and computers, it becomes comparatively much less expensive.
More record labels – right now, it’s kind of the boutique labels, but we’re seeing more record labels offering very high resolution downloads for purchase. And, you know, iTunes has gone up to give the option of downloading. I think it’s iTunes Plus, where you get a much higher quality file than you did. So it’ll be kind of interesting to see what – if the market sort of stirs us. The market stirred us towards the portability of music. It’ll be interesting to see if the demand starts to come back.
DANKOSKY: If we can hit, we can start to head the other direction. Well, Scott Metcalfe, director of recording arts and sciences at The Peabody Institute of Johns Hopkins. Thank you so much for joining us.
METCALFE: Thank you.
DANKOSKY: And thanks also to Sean Olive, director of acoustic research at Harman International. Thank you.
OLIVE: Thank you.
DANKOSKY: Thanks to Scott for the clips we played. We heard “Over Your Head” by Chris Henderson of the Hartford band, Bronze Radio Return, in New York. I’m John Dankosky with SCIENCE FRIDAY.
Originally posted on NPR.COM
