by David Moran with contributions from industry colleagues and from the Davis family.
Mark Davis, who died in April at age 78, was an audio engineer of exceptional breadth of achievement.
Born in Brooklyn and raised in Levittown, he was asked at age five by a girl down the block to fix her record player (“pretty tubes,” he recalled). No luck. For a 6th-grade science project he tried to build a tape recorder using an iron nail: again no luck. In high school he played his father’s huge collection of 78s by just holding a phono cartridge in groove contact and having its wires run to a crystal earphone. “Very pure reproduction.” On weekends he biked up to a Lafayette Radio in Syosset to spend “far too much money on stuff.” He did see to it that he got into a long electronics course in high school: 10th, 11th, and 12th grades. And he became a math star.
Davis went to MIT on full scholarship for bachelor’s, master’s, and doctoral (1980) degrees in EE / psychoacoustics (his thesis was “Computer Simulation of Static Localization Cues”). At the start, Davis wanted to learn how to design circuits, audio and other, but MIT preferred that students learn circuit-design principles, basic EE, nothing audio. Eventually he studied with Barry Blesser and Amar Bose, took a part-time sales job at Tech HiFi, and came to write a four-page critique of the Klipsch hifi of his master’s degree supervisor, Campbell Searle, focusing on driver levels and crossover choices. Searle confirmed to Davis that audio was indeed a serious, reasonable area of study.
As Davis explained in an extensive interview (with David Ranada for the Audio Critic issue 19; spring 1993), Searle “had been a straight circuit theory guy. … [H]e reached a point where it was evident to him that circuitry per se was going to become very cookbook, something only chip designers worried about, where you manipulate the stuff using computer circuit-design programs. Since circuit design was clearly drying out as a major area of research, he decided he needed to shift his academic focus and proceeded to go into psychoacoustics, where the intent was to apply the sort of signals-and-systems theory that had evolved in the electronic disciplines to human hearing — to treat hearing as a signal-processing system and then try to characterize it as best as possible to try to establish causative relationships between what sound went in your ears and what the resulting perception was. … You are not caring primarily about the biological functions or anything like that. You want to know how the whole thing acts as a system. And although tracing things down to basic physiology is a useful endeavor and a check to make sure that your model of the system is accurate, it’s the models that evolve that are really important. So for my PhD I got seriously into psychoacoustics, and there was a very strong combination of applying electrical engineering. I was using computer-based DSP to try to create signals that would impart the illusion of a sound coming from a predetermined direction. I was knee-deep in the tying together of EE, signal processing, and perception. … One of the things that was done was that we made these gooey models of all of our ears and made recordings with microphones in a dummy head with these ears fitted to the outside of the head. We found that you could make recordings that did impart a greater amount of spatial orientation than if you just used a plain-vanilla Kemar head. [But] when you tried to play one person’s recordings to another person, [we found] that it didn’t work generally very well. … I knew signal theory going into the thesis and I knew something about hearing, but I didn’t know a whole lot about statistical analysis. A great deal of time had to be spent on just coming up to speed on that.”
As a doctoral student starting in the early 1970s, then, Davis followed his strong empirical, evidentiary inclinations into giving eye- and ear-opening lecture-demonstrations on the audibility / inaudibility of various audio issues for the newly founded Boston Audio Society and for the Boston AES section. Courtesy of MIT, he would bring a complete test lab, comprising scopes, a real-time spectral analyzer fed by calibrated mikes, a pink-noise generator, and a third-octave filter set. To good-sized crowds of audio enthusiasts and local industry employees, Davis clearly demonstrated the importance (or lack) of such parameters as frequency response, phase response, phase shifts, group delay, and sundry distortions including the ear’s own; the distribution of power vs frequency in various kinds of music; in-room speaker responses and the determinative importance of smooth third-octave performance; and how third-octave EQ can make different speakers sound close to alike but for their radiation patterns. BAS members at the time included such luminaries as Roy Allison and Advent’s Tom Holman and Andy Petite (now Kotsatos; he called Davis’s tutorials “the best presentations I’ve ever seen on audio”), plus a half-dozen budding audio writers and editors who over the next 25 years took that empirical spirit to leading CE magazine and similar positions. (The Boston Audio Society Speaker v1n12 and v2n5 and n6 have writeups of these initial lec-dems, with many following; BAS founder Alvin Foster points out that “Mark was very influential in guiding the ‘scientific’ approach to audio mysteries.”)
Davis undertook design work also, extensively so for phono preamps, the hot subject area at the time: which parts and which kinds of parts made the proper difference. Presently he concocted a line-level circuit for LF boost with steep infrasonic highpassing. His influential phono preamp was marketed for sale, and his refined bass EQ product came to be contracted by Allison Acoustics to become the Electronic Subwoofer. Davis became intensely interested as well in noise reduction and wideband compansion. Finally, experimenting with small AR speakers (reorienting the cabinet, changing individual frequency responses by angle), he delved into horizontal radiation pattern as the key factor dictating how a speaker sounds in an enclosed space, also investigating which horizontal radiation patterns might be preferable for different kinds of program.
Soon he teamed with Francis Daniel (Benchmark Acoustics) to produce a subtle ambiance-enhancing product for front or side speakers, aiming to get rid of treble harshness from beamy direct-facing drivers. As he explained to Ranada, “Upper-midrange and high-frequency sounds are being beamed very narrowly at the listener and not around the room, whereas the rest of the sounds — due to the radiation pattern of the speaker — are getting bounced around the room. The middle and high frequencies are like a pair of headlights shining in your eyes, whereas the rest of the spectrum is more like diffuse background lighting.” Davis gave a unit to one of the hosts of the Boston FM audio show ShopTalk, psychiatrist Richard Goldwater, who reported that “it sent me up to heaven surrounded by KLH9 electrostatic panels in my Cambridge apartment. I hope the speakers in actual heaven are good enough for him.”
Davis continued his didactic efforts with two important articles on audibility, “What’s Really Important in Loudspeaker Performance?” (High Fidelity June 1978) and “Audio Specifications and Human Hearing” (Stereo Review May 1982). Designing a bespoke loudspeaker for the Benchmark system had led to another listener experiment, this one targeted to discover what systemic broadband horizontal radiation pattern, beyond simple crossfiring, would keep sound centered and stable for off-center listening positions: constant directivity, that is, frequency-response-invariant. Volunteers repositioned themselves laterally in front of a pair of 2-way designs, turned to face, using a balance control in order to maintain centered stability and adjusting interaural horizontal levels to produce the oval horizontal radiation. THAT corporation engineer Gary Hebert recalls, “I was an undergrad at MIT and met Mark when he was pursuing his PhD. I was a subject for his localization experiments. A few years after I graduated, he was one of the folks that interviewed me for a job at dbx. As has been stated, Mark was a true genius. He was also always willing to take time to answer a question or listen to a young engineer, once he removed his earplugs.”
Davis approached several of the famous Boston-area loudspeaker companies about producing his “stereo everywhere,” more precisely centered mono everywhere, design. All turned him down. He approached dbx, a prominent audio engineering company not interested in loudspeakers but recently acquired by BSR. They agreed to buy the rights, and in 1981 Davis went on staff, to work on noise reduction in addition. It turned out that he was also this hardcore software developer. Colleague Bob Adams (now Analog Devices Emeritus Fellow) recalls, “His cross-expertise in psychoacoustics, circuits and signal processing made him unique, and he had a very strong drive to bring real products to market. I still remember all the speaker design software he wrote, with thousands of lines of code running on the lowly Apple 2e computer.” Within a few years, including late nights crunching 360deg impulse responses of phased array designs atop an industrial turntable to converge on the desired horizontal radiation, the dbx Soundfield One was born. It received unequaled rave reviews. (Davis explained the project in the AESJ v35n11, Nov 1987, complete with actual measured polar plots of all the models.) Program auditioning included repeated listening to Olivia Newton-John’s “Let’s Get Physical.”
Along with Davis, and then afterward, protégé Michael Chamness extended the beamforming Soundfield line deploying progressively fewer drivers (and prototyping lensed horns to achieve the same result), noting, “He was my first mentor out of school and had a huge impact on my 44 years of loudspeaker work [EAW, Boston Acoustics, and Sonos post-dbx]. Thankful to have worked with him.” Dbx technical communications staffer David Moran found Davis, “as a careful wordsmith himself, to be exceedingly helpful to a lay audio writer, if self-effacing almost to a fault. We took two road trips together, one to Long Island to deliver a company bribe to an audio magazine columnist and the other to spend a weekend interviewing audio pioneer Edgar Villchur [published recently in BASS v45n4]. Mark’s quick wit, modesty of instructional tone, habit of ear-twiddling when energized, and fondness for marijuana were most endearing in a colleague going back to BAS days. He especially enjoyed it when I introduced him to a large audience of engineers as ‘a psychoacoustician — one word, not two.’ I miss him greatly already, having many more audio questions to ask. Including about his deep enthusiasm for Korean pop.”
Experimentation for unrealized dbx type III noise reduction led Davis to design the unusual, complex, psychoacoustically optimized difference-channel companding system used in US MTS TV, winning that national competition over solutions from the CBS Technology Center and Dolby Labs. Davis was known to Dolby engineers from Consumer Electronics Shows where he demonstrated his loudspeaker. In 1985 old friend Holman, then of Lucasfilm, put the company in touch with him for an offer, and he left dbx for Dolby straightaway. Dolby had just acquired its first engineering computer, a Sun SPARC workstation with Mercury Array Processor. Davis was a master of the SPARC realm in those days and could make the machine do whatever was in its range of capabilities, in particular building emulations of early digital audio solutions.
There were many talented audio engineers at Dolby, but Davis was the first PhD to join after founder Ray Dolby himself. Davis and Ray built a great working relationship from day one, as both were deeply knowledgeable in the fundamentals and in state-of-the-art knowhow in their areas of expertise and interest. Ray relied heavily on Davis’s intellect especially in the latter portions of their careers, spending hours discussing technical issues, tackling engineering challenges using multidisciplinary perspectives, concocting unconstrained creative approaches to craft innovative solutions that, even if a stretch, were ultimately realizable with the technology of the day.
Davis became highly respected also for his willingness to help others with solutions, and for his ability to distill problems to fundamentals and rapidly prototype in code. One example of the last was his work to apply his Module.PC (originally written at MIT for Holman’s equalization for the Advent 400 table radio speaker). Holman: “SPICE was the design software of choice for electrical circuits, but you had to tell it the part values in order to produce a desired response. It was not at all over Mark’s head to write Module.PC, which did the reverse: give it a response and have it wiggle the part values until you got what you wanted. The program was important for many years and remains useful today.” It helped automate production of Dolby SR, the company’s premier signal processing solution for professional analog audio recording on magnetic tape and 35mm optical motion picture film, and development of the Dolby AC-3 digital multichannel surround sound encoder/decoder, which enabled five fullband plus one LF-effects channel to be carried in a digital format, the data recorded between the sprocket holes adjacent to the analog sound track on 35mm film. Also achievement in combination: Dolby SR*D Motion Picture Film solution employed SR on optical soundtracks along with AC-3 multichannel surround sound, for a single backward-compatible 35mm optical print delivering both analog and digital multichannel sound to cinemas worldwide.
In his spare time Davis continued work on circuit-analysis, circuit-design, and DSP programs.
For his contribution to the development of Dolby Digital, the AC-3 multichannel sound coding solution used in Digital Television Broadcast, DVD, and streaming consumer applications, in the mid-1990s Davis was awarded, along with seven colleagues, a Technical Emmy. In 2007 for his career endeavors he received SMPTE’s Samuel L. Warner Award, and in 2017 the AES Silver Medal. Davis retired in 2018, having authored or coauthored over a dozen professional technical papers and holding almost three dozen patents.
Mark F. Davis is survived by his wife of 44 years, Helen, daughters Marisa and Susan, and granddaughter Myla.
