Hi-Fi and Less Fi

I’ve written here before about some of the extremely dubious products that are marketed as being capable of improving the sound reproduction of a stereo system, including fancy wire, and even volume control knobs.  The New York Times recently had an article that illustrates what you might call the flip side of this phenomenon.

Whereas the people who buy these questionable, and typically very expensive, gadgets, are looking for better sound, even by means that violate the laws of physics, many music fans today are apparently satisfied with relatively mediocre sound reproduction.  They apparently favor the convenience of the iPod and other ubiquitous MP3 music players.  But, as I pointed out in that previous post, in order to keep their size down, MP3 files are compressed with a lossy compression algorithm, which severely reduces high frequency response, down by about 30 dB at 15kHz.  (That previous post has graphs comparing the original frequency distribution of a sample of music with the frequency distribution from an MP3.)   As the Times puts it:

In many ways, the quality of what people hear — how well the playback reflects the original sound— has taken a step back. To many expert ears, compressed music files produce a crackly, tinnier and thinner sound than music on CDs.

As I’ve mentioned, I have spent a fair amount of time as a semi-serious amateur musician, and  I know what a live performance sounds like.  Even with my middle-aged ears, I can easily hear that my old-timer’s stereo system produces much better, more realistic sound than an MP3 player.

What I think is a bit sad is that I suspect some younger listeners may come to prefer the (unrealistic) MP3 sound, mainly because they don’t know any better.

Jonathan Berger, a professor of music at Stanford, said he had conducted an informal study among his students and found that, over the roughly seven years of the study, an increasing number of them preferred the sound of files with less data over the high-fidelity recordings.

If all someone has ever heard is low-fidelity sound reproduction, he or she may come to believe that’s what music is supposed to sound like.  I know from my own experience that serious musicians work hard at getting a good sound; they spend huge sums on quality instruments (have you priced a Stradivarius lately?).   It would be a shame to lose that appreciation.

Hi-Fi Hokum

I have been meaning to write this post for a few days now, since I saw a product advertisement (discussed below) that really left me shaking my head at the apparent gullibility of some consumers.  Then my friend Phil sent me a link to an article at the Skeptic Web site, dealing with more cases of the same thing: audio products that are sold on the basis of claims that frequently defy not only science but any sort of sense.   This is not a new phenomenon.  I’ve been interested in sound reproduction systems since I was in high school.   I enjoy listening and also have been a fairly serious amateur musician.  I’ve played the trumpet in orchestra and jazz band, and have done a fair amount of singing, so I do know what live music sounds like.

That actually brings me to the first important point I want to make.  I am interested in the quality of sound reproduction: how accurately does the equipment in question replicate the sound as originally recorded?  Some people express a preference for the sound produced by particular types of components, saying that they sound “warmer” or “smoother”, for example.  If that quality was not in the original, it is a distortion.  It is perfectly OK to like the sound, but one is then talking more about something like a musical instrument, rather than a system for sound reproduction.

There is an enormous amount of misinformation pumped out on this topic, despite the fact that the key principles involved have been well understood for decades.

In my 35 years as a professional audio engineer and musician, I’ve seen some of the most outrageous pseudoscience sold to consumers, and even to other audio pros who should know better. Not unlike claims for alternative medicine, nonsense is shrouded in scientific-sounding jargon to confuse the uneducated, or a sales pitch will cite science that is legitimate but irrelevant. The result is endless arguments among audiophiles over basic scientific principles that have been fully understood for fifty years or more.

As the author of the article, Ethan Winer, points out, there are just four basic things that affect the quality of sound reproduction:

• Noise (e.g., hiss on analog magnetic tapes)
• Frequency Response (the notional range of human hearing is 20 Hz –  20 kHz)
• Distortion (essentially, non-linearities in the reproduced signal)
• Time-based errors (more common with electro-mechanical devices, like turntables)

Room acoustics also have an enormous effect on the sound that reaches your ears, but these of course are independent of the equipment used.

One of the oldest products of very dubious value is special, allegedly super-duper wire for connecting loudspeakers to a power amplifier.  Here we are talking about transmitting a relatively low voltage, low frequency (compared to RF signals, for example), and high current signal over a distance that is typically a few feet.  There is no reason to believe that fancy, expensive speaker cables can do this job any better than the stranded lamp cord that you can buy at the hardware store for a few cents a foot.  Even sillier (and this was a new one to me) is the idea that the sound can be improved by using a special AC power cord between the wall socket and the amplifier.  As Mr. Winer correctly points out, the techniques for keeping AC line noise and 60 Hz hum out of the electronics are effective and very well known.  Even if they weren’t, the wiring inside the walls of the user’s house is not going to be replaced.

An advertisement for another very questionable product first renewed my interest in these issues.  It is an integrated amplifier that, according to the advertisement,

… teams up a vacuum tube preamp stage with a state-of-the-art digital-to-analog converter (DAC) and high-current, solid-state power amplifier to deliver rich, satisfying sound. The tube preamp lends smoothness and warmth to your music’s high and mid-range frequencies, while solid-state amplification delivers tight, punchy bass. Sophisticated digital-to-analog conversion means better sound for your digital music library.

Although the supposed superior qualities of vacuum tube amplifiers have been a staple of audio magazine discussions for at least 40 years, I am aware of no evidence from a controlled experiment that, assuming the amplifier is not over-driven to the point of distortion, anyone is able to hear the difference between tube and solid-state amplifiers of equal specs.  And the idea that different devices have some sort of affinity for particular audio frequency ranges is just moonshine.

The best thing about this product, though, is that it is sold as a way to play back music in MP3 format with better sound.   Even if the claims made for “smoothness”  and “warmth” had some validity, the reality is that MP3 files are compressed using a lossy compression technique; among other things, this involves a dramatic cut-off (> 30 dB down) of frequency response above about 15 kHz.   For example, here is a frequency graph of a snippet of musical content in the original:

And here is the corresponding frequency graph for the snippet recorded in MP3 format:

It should be reasonably apparent that having better high frequency response in the amplifier is not going to buy you much.  Oh, and I forgot to mention, the price of this amplifier is $ 1,219.99.

But even this is not the most egregious example.  The Skeptic article mentions a replacement knob for your volume control, which is claimed to have near-miraculous properties:

“The new knobs are custom made with beech wood and bronze … How can this make a difference??? Well, hearing is believing as we always say. The sound becomes much more open and free flowing with a nice improvement in resolution. Dynamics are better and overall naturalness is improved.”

And this amazing knob can be yours for only $485!  I am at somewhat of a loss to explain how anyone can be taken in by this nonsense.  I suppose, if we needed it, it is one more piece of evidence that P.T. Barnum’s Law of Applied Economics is still very much in force.

Twitterdämmerung

About a month ago, I posted a note about a Royal Opera House project to create an opera libretto on the basis of public submissions through Twitter, the micro-blogging service.  The Washington Post today has a report on the “Tweet Success” of the opera’s public performance.

Composed by more than 900 people, the world’s first Twitter opera, as organizers are calling it, made its debut over the weekend at the prestigious Royal Opera House in central London.

It will come as no surprise to readers that have some knowledge of opera that this event was awaited with decidedly mixed feelings:

Opera aficionados had been holding their collective breath — or nose — for this premiere since early last month.  …

For some critics, the sheer idea of uttering the words “Twitter” and “opera” in the same sentence seemed ghastly, as if the high arts were embarking on a garish fling with bubblegum-smacking Miley Cyrus fans.

However, the results, once the mass of Tweets had been edited down to a reasonable length and dubbed “Twitterdämmerung: The Twitter Opera”, were apparently not that bad nor that bizarre – at least by opera standards.

The opera was “actually watchable, listenable and rather funny,” wrote the Daily Telegraph‘s opera critic, Igor Toronyi-Lalic. One couldn’t escape the fact that it was a gimmick, he said, “but as cheap gimmicks go, this was a good ‘un.”

The Twitter Opera was only 20 minutes long, mercifully, nowhere near the length of a Wagnerian epic.  (One can scarcely imagine someone whose normal world view is comprised of 140-character messages sitting through the Ring cycle.)  It was performed in a small room, not in the main opera theater; among other things, this allowed the provision of laptops around the sides of the room for any of the audience who experienced acute Twitter withdrawal symptoms.  Apparently the four performances attracted a total audience of about 1,000.

Although it’s easy to make fun of this project (and probably some of that is deserved), I think it’s kind of encouraging that the folks at the Royal Opera House are willing to take a few risks to promote their art.  Although I grew up in a fairly musical environment, and always liked classical music, I initially was pretty uninterested in opera, feeling that it was almost like music embalmed.  I was finally persuaded by a friend to attend a performance of Mozart’s Die Zauberflöte, and I was hooked.  Since then, I’ve dragged a few other friends along, and have made more converts.

The idea of putting the story together from a mass of individual contributions may seem a bit strange; but after all, Wagner’s Ring cycle (Der Ring des Nibelungen) is a story taken from old Norse and German epics.  In any case, the idea of plot is a rather loose one in operas generally.

So I wish the Royal Opera folks well with their experiments, and I’m looking forward to see what their next wacky idea is.