In February, a music professor at Stanford, Jonathan Berger, revealed that he has found evidence that younger listeners have come to prefer lo-fi versions of rock songs to hi-fi ones. For six years, Berger played different versions of the same rock songs to his students and asked them to say which ones they liked best. Each year, more students said that they liked what they heard from MP3s better than what came from CDs. To a new generation of iPod listeners, rock music is supposed to sound lo-fi.
Here is more. The whole series — notable new ideas picked out by the NYT — starts here and as usual it is worth perusing the entire list.
It is a mistake to think that “hi-fi” should be preferable to lossy mp3. Andy Warhol prints and Lichtenstein paintings do not suffer because of lack of resolved detail. Some music sound much better compressed and with rounded off top end. Besides anybody that thinks that CD’s are inherently hi-fi are deaf to harshness of some of the recordings.
But anon, a perfectly fine CD player costs, what, $40? Amplifiers and speakers are the same whether you listen to a CD or mp3.
It’s hard to imagine anyone who can afford an mp3 player but cannot afford to listen to CDs.
MP3 encoding throws away the parts of the sound which the brain doesn’t perceive. So perhaps an MP3 encoded song isn’t ‘lo-fi’ but is simply less cluttered?
Certainly much wider choice is excellent, but surely its not being “arrogant” to have reservations about the significant differences between, say, a live performance of a sonata, an LP recording, a CD, and an MP3 version. The point being that the live version is close (on the whole) to the composer’s intent. The drift due to various forms of electronic compression produces something else. We may come to like it, but it is still something else. On the other hand, this line of reasoning probably doesn’t apply to Lady Gaga (who I admire greatly..)
I prefer toting around my midget string quartet. Now that’s LO fi !
Rock music in particular has had mixed feelings about sound quality. Many songs from the first two decades of rock were first exposed over AM radio, and made a particular sonic impression. Some producers mixed to obtain a desired “sound” in mono and for the radio.
When older songs and later albums originally recorded for LP’s were reissued on CDs, engineers and producers found the result was different. In many cases better, but in many cases worse, but usually different from how previous listeners had experienced the recording.
Now that many older albums have been reissued multiple times, from different generations of tapes or with modern enhancement techniques from original tapes, it is possible to own three or four versions of the same album, each with different sonic qualities. Some are more agreeable than others, but the “best definition” versions are not necessarily artistically superior to those with lesser fidelity. Or to the resulting MP3 versions.
I can hear the same effect in differing releases of classic bluegrass recordings, where crisper fidelity of the modern technically enhanced recordings derived from original tapes suffer in the ensemble aspects of the music. Individual players are more readily discerned, but that is not always desirable, just as is the case with some rock music.
I could not find the source in NT materials with in-depth description of experiment setups, so I can only hope that it was a proper blind test with mp3 constructed from the same CD.
When the recording is crappily played/mastered, you can indeed prefer low-rate mp3 to the CD, as much of the crappiness in high frequences will be cut off. I, for one, cannot listen to much of contemporary pop on a high-quality system just because I can often hear compression/exciter artifacts, unnatural sounds in high drums and strings, and it spoils the fun. MP3 is sometimes better because it cuts off messed up high frequences
This result is obvious to anyone who has spent significant time listening to vintage Swing music or its modern interpretations.
To put on my grammar purist hat for a minute, please stop using “peruse” to mean its direct opposite. We have to fight the good fight before this incorrect usage creeps up to the first definition listed.
Some college students prefer PBR. Sure it is beer, but it’s no Russian River Redemption.
Basti, you are confused. mp3 compression has *no* effect on dynamic range whatsoever, and the dynamic range of an mp3 is identical to the dynamic range of an uncompressed CD quality wave. I will provide proof if you desire. The loss of dynamic range in modern pop music is done in the recording, mixing, and mastering process, a bizarre mixture of aesthetic preference and technological misuse. But it’s done well before before mp3 or even CDs are considered.
A CD is as high-fi as any audio format needs to be.”
I do not think this is accurate. In order to make a sound “fit” CD quality – that is, a bandwidth up to 22.5 kHz – frequencies above 22.5 must be filtered out to avoid aliasing. (This is called an anti-aliasing filter.) Since the limits of CD quality audio are so close to the audible range you will either have to use a steep filter and introduce phase-shift and other artifacts that will occur in the audible band, or you will have to use a less steep filter that will attenuate much of the high-end. The benefits of DVD audio should be apparent – the sampling rate is twice that of CD quality audio, so the anti-aliasing filter can be less steep AND farther from the audible range.
The result is no different than similar tests decades ago. It has been decades since I was very interested in audio reproduction fidelity, so I can’t recall the details; vaguely, I think Edison ran tests to demonstrate his cylinder gramophone was preferable to listeners than the original.
Edgar Villchur sought to convince people of the higher fidelity of AR speakers based on his inventions; he set up listening test studios in places like Grand Central Station, and came up with ways to move from live to sound reproduced by his speakers to demonstrate their fidelity. I believe competitors setup tests to show people didn’t like the sound of live.
His more famous student was a guy named Kloss; together they founded AR, circa 1953, which spawned a few years later when Kloss co-founded KLH, then Advent,…
What audio test engineers discovered is the need to train your listeners to the way things actually sound, though even that can lead you astray.
I think the question to ask is “how many of those polled actually listen to live music?”
When I was on a college campus in the 60s I listened to a lot of live music, from the folkies playing on the lawn, the formal recitals, the guest appearances of professional artists, music students practicing, people just entertaining themselves or relaxing, rock bands on the weekends, other styles performed with amplification (say a Joan Baez).
Its been more than two years since I’ve heard live music, a jazz band demonstrating styles and techniques for an intro to music appreciation. I can’t think of a place in the area with live music of any type, while decades ago live music was required for many tavern-ish popular dining places, even in smaller cities and towns.
Do you listen to live music????
If not, how do you know if CDs are more accurate than MP3s?
Two comments:
1. There is substantial misinformation in these comments regarding the technicalities of MP3 compression. Compression in data is quite different from ‘compression’ in sound design.
2. I think it was with the electric guitar that we first broadly realized the joy of distorting sounds. As a musician, I’m very fond of effects that downsample and introduce aliasing.
In the not-too-distant future, portable players will be able to hold as many lossless (CD quality) files as most people would care to have.
MP3 players are already able to store tens of thousands of 256 kbp files. Granted, most people these days don’t want hard drive models, but a 32gig can hold thousands.
Comments on this entry are closed.