Musician and Vocalist Artist Interviews

Mark Levinson: For the Love of Music - Part 2

Earlier this year, I had the privilege of interviewing Mark Levinson, one of the most influential men in the entire high-end audio industry. We spent hours discussing a wide range of topics—so many, in fact, that they could not be squeezed into a single article.

If you haven't yet read part 1, I encourage you to do so before moving on to part 2. Here, Levinson shares his thoughts on analog versus digital audio, the effect of cables on sound quality, 2-channel versus multichannel music, movie soundtracks, and the problems facing consumers shopping for audio gear.

Analog vs. Digital Audio
Naturally, I was interested in Levinson's take on analog versus digital audio. "For me, the first thing is the content. If the content is only available on SACD or CD or LP or download or wherever, that's what I'm going to listen to, because if I don't, I won't hear it at all. If it's a recording I want to listen to, I don't care what it's on, I want to listen to that particular piece of music.

"Every recording medium has its limitations. I've always liked analog tape, but it's very expensive and messy. The recorders are big and heavy, and you have to calibrate them with mechanical and electrical adjustments. It's very time consuming and requires a lot of test equipment. Plus, the tape stock varies and can be hard to get. It goes on and on."

What about digital? "There are two kinds of digital audio. There's PCM (pulse-code modulation), which is used on CDs and DVD-Audio, and DSD (direct-stream digital), which is the basis of SACD (Super Audio Compact Disc). In terms of sound quality, PCM has always been a mixed bag; there have always been people who don't like it. Dr. John Diamond went further than anyone in understanding the problems of PCM, but the world didn't want to hear what he had to say." For more on Diamond's findings, see www.drjohndiamond.com/digital.

"On the other hand, DSD was very well received from the outset. The problem was fraud, which occurred when Sony wanted to popularize SACD and took PCM recordings and converted them to DSD and expected people to get excited about it. But in fact, they were giving people the same thing they already had on CD, but in a different format.

"What Sony really wanted to do is stop piracy. If they had said that, it would have been fine. But instead, they tried to tell people how great DSD was, and then put out CD material as SACD. No wonder people weren't impressed! They were being sold the same thing they already had.

"In some cases, the audio was recorded in DSD but converted to PCM for editing. In fact, that was common because Sony didn't make any tools like EQ or compression for DSD. So engineers recorded in DSD, converted to PCM for processing, then back to DSD for the final SACD. This wiped out the advantage of DSD; it was a travesty.

"When DSD is done properly, it's very transparent. But it requires very good converters. It's full of promise, but it has been largely abandoned. A few record companies still put out SACDs, mostly because it's almost impossible to copy them, which was Sony's original intent.

"In the end, Sony abandoned the medium and did not make any tools for engineers to use. So if you wanted to do any post-processing, you had to convert to PCM. By comparison, you can get a set of plug-ins with 300 different algorithms to use for mastering PCM. For DSD, there's nothing like that.

"Using John Diamond's methodology, we tested PCM systems up to the full resolution of DSD. In that case, we recorded in DSD, then sent the data through a PCM processor called DSD Wide with 32-bit floating-point resolution, then converted it back to DSD. And it failed.

We proved that it isn't bits or bandwidth or resolution. It's something else. There's something physiological going on.

"The industry is in no great rush to do any testing for this sort of thing, because they want to make money by investing the least amount they can get away with and get the maximum return in the shortest time. The principles of return-on-investment dictate that any scrutiny be kept to a minimum; as long as they can sell it, they do it. It's putting the customer last."

Despite the controversy, Levinson is focused on optimizing PCM because that is the global format of all music and audio. "To me, the most important thing is that musicians need income, and the way things are going, they're not going to get it from analog LPs or SACDs. We need to make PCM work, so we can continue to have musicians on the planet. My goal is to make PCM sound and feel more like analog or DSD, and I've made some progress in that direction. In that process, PCM becomes acceptable to me as a medium."

Levinson will soon release a new audio-software product for the Mac called Master Class that is, in part, a digital version of his legendary Audio Palette. "The Audio Palette was for the privileged few; Master Class does everything the Audio Palette did and much more, yet is within reach of a wider audience."

The Great Cable Debate
Another contentious topic is the effect of cables on the sound of an audio system. As Levinson recounts, "Let's say you put a sine wave into a preamp, and you have a cable between the preamp and power amp. At the output of the power amp, you measure the frequency response. If you change the cable between them, the measured frequency response will change slightly, which shows you that a lot of what we hear is the result of small differences in frequency response. I'm not saying that all cables are like that, but more than anyone would guess, this is what's happening.

"At Daniel Hertz, we built a box with a stereo input and stereo output. It also has a stereo loop and a switch. With the switch in one position, the signal goes straight through from input to output, and in the other position, the signal goes through the loop. The idea is to connect a cable to the loop and see if you can hear a difference in the sound due to inserting that cable."

What were the results? "The effect of different cables was sometimes audible and usually measureable, but the differences were generally small. Plus, the effect of the cable depends on the situation. You're not listening to the cable alone, you're listening to the interaction of the output impedance at one end, the input impedance at the other end, and the cable in between. You're listening to the interaction of impedance, inductance, and so on." Of course, in Levinson's cable tests, all the electronics and cabling (other than the cable being tested) were the same, and the relationship between output and input impedances were optimized for high-quality audio, preventing these elements from undermining the results.

In addition to a preamp, power amp, and speakers, Daniel Hertz also makes interconnect and speaker cables—or, to be more precise, one interconnect and one speaker cable. "Our cables are based on three factors. First, silver is the best known conductor of electricity; copper is number 2, and gold is number 7. Second, Litz construction offers the lowest inductance, and third, Teflon dielectric offers the lowest capacitance. So that's what we use. And we only need one cable, not 10 or 20 different models. Why? Because it's transparent, end of story. No one so far has been able to hear any difference in our interconnect cables connected to that loop/switch box.


Daniel Hertz Premium 500 speaker cables include 500 strands of individually coated silver wire in a Litz configuration, yet they are no thicker than a pencil.

"In my opinion, the high-end audio-cable business is mostly a farce, and in some respects it's based on deception and fraud. The definition of fraud is willfully misleading the public for financial gain, and that's exactly what I believe many of these cable companies are doing. They know it's not real, but they do it anyway. It's unethical, and I deplore it."

2-Channel vs. Multichannel Music
With his deep knowledge and experience in music recording, I wanted to find out how Levinson feels about multichannel recordings as opposed to traditional stereo. "There's merit to all formats—mono, 2-channel, 5-channel. The listening experience depends on how well the format is utilized.

"Most surround-music recordings are gimmicks. When we listen to music, the sound is usually coming from in front of us. In general, music in surround has not been effectively utilized, but the potential is there. If you want to have the music come from all around, that's fine, but I would use multichannel music recording differently.

"Bell Labs published some papers in the 1930s, and Paul Klipsch gave me a set. These papers described the advantages of different multichannel formats beyond mono. They ended up recommending three channels—left, center, and right. But because of the technical limitations at the time, they settled on left, left+right in the middle, and right. They never really advocated 2-channel stereo. Paul Klipsch also advocated exactly that.

"It makes sense. You want music in the center, so how can we get that without a speaker in the center? Stereo works on the phantom-center principle, but the ideal would be to have three channels. It just hasn't been practical until now. How do you make an LP with three channels?

"In the 1950s, some Ampex recorders were equipped with 3-channel heads, and they used three microphones. They mixed it to mono or stereo, but they actually had the foresight to think that someday there might be a 3-channel reproduction system. But if you made a 3-channel recording, say with three microphones across the front, how would you play it back? What would you do with the surround channels? At this point, it's an experimental idea. I'm just not sure how it would be made compatible with the world.

"In any case, the key is not the format, it's the content. What's the content, and how do you use multichannel to enhance it? In most cases, the music is in front of you, so the best you can do is put reverb in the surround channels.

"The industry has become obsessed with complexity and over-produced results without enough thought given to the content itself. It's the number of channels or bandwidth or dynamic range or something else except the quality of the talent, the quality of the music. The attention is on parameters, not on soul. I know of mono recordings from the 1950s that are infinitely more enjoyable than many modern recordings, not because of the format, but because of the music.

"Among the many recordings I've mastered, the one that people have fallen in love with the most is a 1950s recording of Charlie Parker playing 'Out of Nowhere.' The original recording quality was what you would call primitive; no audio store would use it in a demo. I remastered it, which produced a file that has people mesmerized. It feels like Charlie Parker is in the room with you, it sounds like a contemporary recording.

"When I was demonstrating the Daniel Hertz system at CES, I met a guy who produces a jazz radio program in Chicago. He told me that the single most moving thing I played him was that track of Charlie Parker. That alone was worth the trip to CES for him. That's what I'm trying to say—the purpose of the equipment is to experience the content."

Audio for Movies
These days, of course, movie soundtracks are multichannel by nature. But when Cello got involved in home theater, Levinson recognized some problems with that paradigm. "We realized that, in many ways, surround sound was flawed. It was designed for movie theaters, in which you can't have a left and right channel; people far from one channel in a large theater can't hear it. Movie theaters have to be mono.

"But then they added incidental sounds in the surround channels. If you shut off the surround channels, you won't miss anything. It might be more interesting to hear the incidental sounds, but they do not contain anything necessary to understanding the movie. All the dialog, music, sound effects, everything of importance is in the center channel.

"When you get into a home, most people have a television. You can't put a speaker behind a television like you can in a theater with a perforated screen. So the one channel that's carrying all the dialog, music, sound effects—99 percent of the information—is the smallest speaker in the system. Those nice big speakers on the left and right of the TV don't do anything.

"This is a very tough problem. It doesn't bother the big companies too much because they sell by marketing, they get everyone to buy, so they're making money. This is an example of the true illness of the industry right now, which is that the customer comes last. The first thing is profit for manufacturers and dealers—whatever makes money, they do it.

"If you really want to know the truth, try listening to two really good speakers and electronics, then take that money and divide it up 5.1 ways and see what your money will buy you. If you like the 5.1, buy that. If you like the 2-channel, buy that. I'm not advocating one or the other. What I'm saying is that the way the industry is set up now, you're buying six channels of amplification and six speakers, and 99 percent of what you are hearing in a movie is coming from one speaker, the center. Listen for yourself, see what you think. A lot of people find that their money is better spent on two good channels than six mediocre channels.

"Even if you have a great center channel, where are you going to put it? You can't put it in front of your TV, and you can't put it behind your TV. It has to go underneath or above it. Another alternative is to use speakers on the right and left of the screen and run them in mono. Then, it will image right in the middle of the screen. But that's not the industry standard."

Levinson also notes that movie soundtracks are not accurate representations of real-life sounds. "In real life, if a helicopter lands in front of you, you know it. With many sounds that people experience when they're watching a movie—like a gunshot, door slam, or helicopter—there's no way a normal audio system could even approximate that sound.

"On the other hand, if someone says, 'I'd like to see how close we can come,' we can do that. But it requires a whole lot of science and a whole lot of money.

"There are really three areas you want to think about in terms of audio. One is tone and timbre. Does a voice sound like a real voice? Does a guitar sound like a real guitar? Is the sound natural? The more natural it is, the more you get into the story. The less natural it is, the more removed you are.

"The second one is bass. Good bass is expensive; the deeper the bass, the more expensive the system is going to be. But it's not only low frequency response, it's also the quality of the bass. It can't be tubby or boomy; it must be clean.

"The third thing is dynamic range. For example, let's say someone fires a gun six feet away. That is going to be a very loud sound. There's no way you can reproduce it on normal audio equipment. Even the most dynamic systems are challenged by something like that.

"Film soundtracks are not accurate. They are designed so they won't distort when reproduced on even very inexpensive systems. For example, a television with built-in speakers the size of a quarter can't distort or people won't like it. So they have to homogenize the tracks so they won't distort when played on extremely simple systems.

"This means the dynamic range has been compressed and the frequency range has been limited. They have to do that, otherwise they'd be creating a lot of customer dissatisfaction. While you might want to reproduce a gun shot or helicopter, that isn't going to happen. Not only are most audio systems incapable of reproducing these sounds accurately, they can't even be captured completely.

"That being said, what's the best you can do under the circumstances? There is enough information that, if you take care in designing the audio system, you can have amazing results. But the problem is, there are different levels of quality. People generally understand that a TV's picture looks great or not; it's a pretty easy concept, because you can see it. But with audio, most people are not as connected to the technology, so it's a little more difficult to understand the importance of the audio system."

Shopping Advice
Levinson recognizes that consumers are at a distinct disadvantage when it comes to shopping for audio gear. "One of the biggest problems for consumers is, where do you get your information? On what do you base your buying decision? How do you form your opinions? The best way is direct experience, but that isn't always practical or even possible. So you can read reviews and such, but most people don't have time for that.

"So most people go to a store and talk to a salesperson. There are some great people in retail, and if you go to the right place and talk to the right person, you can get some great help. But for every great one, there are 1000 people who are not educated, so they are not in a position to help you.

"Worse, they are given directives by the management to sell one product or another. Or manufacturers give them a spiff. So they have these incentives, which means the consumer is in a difficult position. The industry is controlled by big money. These multibillion-dollar companies pay for the ads and finance the inventory. They control what happens, and the consumer must survive that.

"This is one reason why there is still a place for the specialty store. In the early days of audio, these were family-run businesses, and they loved what they did. It was friendly. Sadly, most of these family stores went out of business because of the big chains.

"The Internet can be very helpful, because you can read what consumers say about their experiences with a product. Amazon has used this concept to great effect. It's what the Internet was designed to do in the first place: share information.

"There is also a place for knowledgeable, trustworthy, articulate reviewers, and the Internet provides an avenue to get that information to consumers. In that respect, it's a great solution to the problem we're talking about."

Coda


As you've no doubt discerned by now, Levinson's first priority is to reproduce music recordings as accurately and faithfully as possible with equipment that is well-engineered, robust, and easy to use. He is justifiably proud that products he designed decades ago were manufactured for many years with virtually no change, and they're still going strong, commanding high prices on the used market—if you can find anyone willing to part with them at all.

What does the future hold? If the past is any indication, Mark Levinson will continue his quest to create the best-sounding, most bulletproof audio products for true music lovers everywhere. And I have no doubt that we haven't heard the last from this man of singular passion and vision.