Expert testimony
Jürgen Ackermann is a 37-year-old psychologist living in Frankfurt, Germany. He has long been interested in music and its reproduction, building amplifiers and speakers for himself as well as for some friends. His current home system includes a home-brew tube preamp, a home-brew single-ended triode power amp (the power in question being all of 2W from a single 2A3 per channel), and modified Klipschorns. This system is seriously loud when required, those sound bursts from Flim and the BB's Tricycle coming across as positively threatening---yet it whispers with a clarity and conviction most minimonitors fall short of. His amp is remarkable in that there is none of the hum that is generally unavoidable with direct-heated triodes. He has designed an indirect heating that relies on very precise balancing of voltages, and has made it work beautifully.
As part of his doctoral thesis, Ackermann researched the experience of music reproduction in the home. He conducted an experiment, setting up three systems in a room of the Frankfurt Hochschule für Musik und Darstellende Kunst (Music and Performing Arts University). The first system consisted of an analog record player, ca $4800, and a tube pre- and triode power-amp combination worth ca $4500 (hereinafter called the analog system). The second system substituted a respected CD player, ca $2400, which has been well reviewed worldwide, including in the pages of Stereophile, but retained the tube amps. The third system kept the CD player but was powered by a transistor pre-/power combination worth ca $11,000 (hereinafter called the digital system).
The components had been selected as being reasonably representative of their kind. The loudspeakers were held constant and had been selected for their ability to sound equally good driven by tubes or transistors. If anything, the system favored the expensive transistor combo, which had been selected because it was one of the best-selling combinations in its price range, and also because comparative listening tests against some other transistor amps had revealed this combo to sound particularly good in the test configuration. All three systems were played at exactly the same loudness level.
Ackermann found 53 people from all walks of life willing to participate in his experiment: hi-fi enthusiasts, musicians, and "normal" people with no special relation to music or its reproduction. The selection of participants was not truly stochastic, but the sample was large enough to give meaningful results.
Participants were seated in a room before a pair of loudspeakers. The part of the room behind the speakers was partitioned off with dense cloth so that the participants could not know what went on behind this curtain. Indeed, they had no idea what was going on or what, if anything, was changed between trials, except that they were going to be interviewed on their reactions to several pieces of music. Ackermann made the system changeovers without once interacting with the participants.
The participants were received and instructed by a student who was paid for her time. This student, who had no knowledge of things hi-fi, was instructed to sit behind the participants so she could not influence the participants even subconsciously. The student first gave the participants a questionnaire that asked for their musical likes and dislikes. A second questionnaire asked how the participants normally listened to music, and a third questionnaire tried to establish the emotional base level at which each participant entered into the experiment. These and all other questionnaires were standard forms developed for musico-sociological research, and had been pretested to be meaningful and easily understandable by the participants.
Then the participants were played a standardized set of three musical pieces. These were tracks from Larry Conklin's Dolphin Grace (light jazz), Sally Barker's This Rhythm Is Mine (pop), and Italian Violin Musik, 1600-1750 on Edition Open Window (baroque classical music). The tracks had been selected after a preparatory experiment showed that they gave meaningful results. None was offensive to the participants; strong individual likes or dislikes could not influence the experiment's outcome.
After the first run-through, participants were given three more questionnaires: one asking for their emotional balance (the same questionnaire as before the music began), one asking how the participants had experienced the musical tracks, and one asking for their opinions on these tracks.
Then the participants were played the same tracks on a different system, and again had to fill out the three questionnaires; and so on with the third system. The sequence of the three systems was randomized so that familiarity effects, or fatigue, could not influence the overall outcome.
After the third trial, the participants were asked to fill out, besides the three standard questionnaires, a final questionnaire asking whether they had a music system at home, what it consisted of, and how expensive the components were.
Finally, the participants were asked by the student which of the three still-unidentified systems they would buy. The student also took notes of the participants' behavior during the tests: Did they react to the music by moving their feet? Did they sit through the presentation, or did they talk or stand up while the music was playing? and so on.
The tests were not exhaustive, in the sense that further questions might have shed even more light on the subjects' response to the three systems. But, as each test took about two hours, it was felt that this was the maximum time that people without any interest in the outcome of the experiment would be willing to be subjected to the rigors of being under very close scrutiny (13 multi-page questionnaires to fill out---what a chore).
Care was taken to keep exterior factors constant. The listening room was not darkened, because it was felt that listening in a dark room would be too far outside everyday experience for most participants. It is well known that lighting conditions have an effect on people's mood (or why do you turn out the lights when you want to share a little intimacy with your partner?). To keep lighting conditions constant, the experiments were restricted to a time slot between around 10am and 2pm, which meant that only two or, at a pinch, three persons a day could be interviewed. The time of day at which each interview was conducted was noted; it will be interesting to see if there is a correlation between time of day and the results.
Giving the complete results of Jürgen Ackermann's experiment would be way beyond the scope of an article such as this one; besides, Ackermann has not yet completed his statistical analysis. But there are already some results that seem interesting enough to warrant a preliminary report.
Let's start with the emotional states of the participants. The participants began with a base tension level of 3.26; with the digital system this dropped to 2.35, and with the analog system to 1.75. Nervousness was raised from a base level of 1.8 to 2.2 by the digital system, but fell to 1.1 with the analog system. The need for relaxation fell from a base level of 2.6 to 1.9 with the analog system, but rose to 2.9 with the digital system. The ability to concentrate remained constant with the analog system at 4.3, but fell to 3.6 with the digital system. Relaxedness stayed constant with the digital system at 4.0, but rose to 4.6 with the analog system. This shows that the analog system worked toward a feeling of serenity in the participant, whereas the digital system heightened tension and stress.
Equally interesting was the response to the question of whether the participants liked the music they were played. With the analog system, 43 out of the 53 participants said they liked the Larry Conklin piece, 46 the baroque music, and 38 the Sally Barker piece. The music was heard as interesting, emotionally appealing, and engaging. Via the digital system, the levels fell to 31, 33, and 33, respectively. The same music was now more often experienced as boring. Food for thought.
The questionnaire asking for the listeners' experience of the music gave just as interesting results. Thirty participants sang along with the music under their breaths when it was played via the analog system, and only 19 with the digital system. Forty-seven participants said they had let themselves be carried along by the analog system, 19 with the digital system. When questioned whether the music had influenced their movements (tapping their feet, etc.), the numbers were 30 and 25. Forty-six participants had been inspired to think about the music by the analog system, 34 by the digital system. Forty-seven participants said the music had improved their sense of well-being via the analog system, 31 via the digital.
Conversely, no participant said that the analog system had impaired their sense of well-being, but 16 participants said so of the digital system! This must be one of the most astonishing, and irritating, results of Ackermann's experiment. How can it be that we spend a lot of money on something that makes us feel worse?!
The results of the "intermediate" CD/tube system were consistently between those of the digital and analog systems.
At the end of the test, the participants were asked which of the systems they would buy. Those listeners who had some experience of things hi-fi preferred the digital system, which they thought sounded better. Those participants without such experience preferred the analog system's sound. The conclusion Ackermann drew from this is that the sound of modern hi-fi is the result of a learning process. When told that a certain sound is what they should aim for, often enough people will accept this concept of sound as their internal reference.
Another inference that may be drawn from this question is that there was no correlation between what the participants experienced as good sound and which system made them feel good. In other words, the perceived quality of sound had no influence on whether the participants liked the music and its emotional impact on the listeners. One participant, a musician, even responded that he could hear absolutely no difference in sound between the presentations, yet his emotional response was very different on the three trials, and showed complete conformity with the rest of the participants.
Some of you will have noticed that there was one long-term test subject: the student who accompanied the participants during their time in the listening room. The poor girl had to listen to the above-mentioned pieces 159 times! At the end of the experiment, she asked Ackermann what the systems were. She said she couldn't stand the sound of one of the systems anymore, feeling physically attacked by its sound. By now, it won't surprise you that the system in question turned out to be the CD/solid-state one.
