Tuesday, February 17, 2009

An interview with James A. Moorer, pt.2

by Matteo Milani, February 2009

(Continued from Page 1)

MM: Did Lucas himself choose the name SoundDroid?

JM: No. It came from the name of the group, which was "The Droid Works". Peter Langston came up with the name "Droid Works".

MM: Is it true that SoundDroid was the second generation digital audio workstation after the ASP? Or the ASP was called later SoundDroid?

JM: The ASP was later renamed SoundDroid. We were working on a second generation when the work was stopped.

The Droid Works closed before the production-ready version of the device was completed, before any products were delivered. Still, the audio research that went into it lived on. In the last months of 1986, Lucasfilm tried to find an industry leader to purchase the technologies had been created there, with no success.
[excerpt from Droidmaker

MM: What kind of sound processing did you implement? Filtering, eq, dynamics?

JM: All the kinds that were known at that time - filtering, dynamics, spatialization, reverberation, pitch shifting, denoising, and many more.

MM: Did your pioneering work on the Phase Vocoder lead to NoNoise for removing hiss, noise, clicks and pops from recordings?

JM: The Phase Vocoder work did lead directly to the algorithms for removing clicks and pops. The broadband noise reduction came out of some work I did with John Strawn on the "Karen Silkwood Tapes" back at Stanford. These were the techniques used for the Amadeus restoration projects.

Milos Forman was [...] mixing “Amadeus” at Fantasy Films (Saul Zaentz). [...] There was a scene in the movie (the asylum scene) where the performances of Salieri and Mozart were excellent but the audio recording was not. Forman's sound engineers, Mark Berger and Tom Scott, knew that Lucas's teams were pioneering new technology for exactly this kind of problem. They brought the tapes over to Andy Moorer and asked if his ASP could help.[...] Andy immediately digitized the tapes into the ASP. Then he excised small samples of pure noise and had the computer analyze them - getting, in essence, a fingerprint of the background noise. Then the pure noise was subtracted from the dialog tracks, leaving the human voices intact.
"About half of the tapes they brought me we were able to clean up," said Moorer. Amadeus was the first feature film to utilize their new noise reduction technology.
[excerpt from

MM: The visual waveform had been demonstrated originally in your program called "S" (for Sound) at Stanford in 1970. How was your contribution to the user interface design? Where did you find your inspiration of waveforms, from optical recordings?

JM: ALL modern digital audio workstations use the SAME waveform display that I created in the program "S". All modern digital audio workstations are descended directly from "S". I have no idea where the idea for the display on "S" came from. It just seemed good at the time.

MM: What was the audio recording format for the SoundDroid?

JM: We used 16 or 24-bit samples at 48 kHz.

MM: How big was the storage of the SoundDroid?

JM: We used 300 MByte "Storage Module Drives" on the original ASP. One drive would hold about 50 minutes of monaural sound at 48K and 16-bit samples. You could attach up to 4 drives to one ASP. With SoundDroid, we got some newer drives that could hold 850 MBytes, or about 2 hours of mono. Four of those drives adds up to about 1 hour of eight-track audio.

MM: How was your connection with Tomlinson Holman? Did you collaborate at that time?

JM: I enjoyed working with Tom Holman, although we had different responsibilities. He was all analog and didn't really work much with digital audio.

MM: How was your relationship with Ben Burtt? Was he the only figure who helped you sweetening the software?

JM: Ben helped me understand what sound designers did and how they do it.

"Ben Burtt said the biggest bottleneck was the spatializing of sounds; it was his biggest time sink," recalled Andy [Moorer]. Ben needed fine control over the room size around a sound, a magic knob that would adjust the spatial environment so he could immediately hear how a sound could be varied.
His traditional method was slow and tedious, with tape decks and amplifiers. And creating flybys! All those objects - space ships, motorcycles, airplanes - that had to be made to sound like they were moving fast by an imaginary microphone. He needed a Doppler shift that could be added to a sound.
Andy provided a solution. "We put a box in his little mixing room in the basement of C building, connected through the wall to the sound pit, where the ASP resided." He wrote Ben a basic set of programs that would allow the box to input a sound, fly it around at a desired speed, and output it.
[excerpt from

MM: The Doppler machine you invented for Ben was stand-alone or was it a remote controller doing a specific task?

JM: It was a remote terminal on the ASP. We would load a specific program into the machine for him to use.

[Prev | 1 | 2 | 3 | Next]

No comments:

Post a Comment