Letters, Dec. 9, 1946

(4 of 4)

Such statements are no more in accord with the facts than is TIME'S statement that pickups ranging to 7,000 cycles ". . . fail to catch the high overtones of a flute." Of all musical instruments, the flute is distinguished by its complete lack of overtones on its higher notes, and the extremely thin harmonic structure of its low tones. The highest overtone produced by a flute is about 2,500 cycles.

The response curve of the Cobra is the result of more than three years' experimentation and hundreds of listening tests. Our human guinea pigs ranged from factory workers to outstanding musicians. We decided on the curve we finally adopted when the nonmusicians voted "Swell" and some of the musicians heard in old records nuances and even instruments that they had never known were there. . . .

We could easily build the Cobra to almost any response curve desired. Some of our first experiments had cartridges peaked at nearly 10,000 cycles. They did not give the fidelity of reproduction, did not make each instrument speak with its own natural voice, as did the curve finally adopted. This curve, measured on response to pure frequencies, peaks at a point slightly below 5,000 cycles, falls away, and then continues out flat to well beyond 10,000 cycles. . . .

There is much yet to be learned about both recording and reproduction of music. Undue emphasis on "high" numerical fidelity has tended to obscure the importance of true fidelity throughout the more important, musically, middle frequency ranges where occur nearly all of the tones and overtones in the music we hear played. . . .

G. E. GUSTAFSON Vice President in charge of Engineering Zenith Radio Corp. Chicago