June 2001
Review Apr 2001 Meeting
|
|
June 2001 Review Apr 2001 Meeting |
Floyd Toole: Loudspeakers, Rooms & Their Interactions
Presented by Floyd Toole, Corporate Vice President -Engineering, Harman International
Tuesday April 24th was a day highly anticipated by many of the Toronto Section. 60 members and guests attended the presentation in the comfortable confines of Ryerson's Eaton Lecture Theatre and were not disappointed as Mr. Toole 's talk provided pearls of wisdom on a long string, stimulating lively questions and discussion.
Perhaps stating the overlooked obvious, Mr. Toole said that the aim of all this science was nothing more (or less) than the preservation and presentation of art. Starting with a brief overview of room acoustics as related to reproduction he stated that the room environment dominates the imaging qualities of reproduction by reflections but totally controls bass quality by its modal response. He then quickly moved to an explanation of an audio chicken and egg scenario, the Circle of Confusion, where recordings made on coloured loudspeakers and microphones are evaluated on coloured loudspeakers. The only way to break this cycle is by rigorous measurement and a solid understanding of psychoacoustics.
This bootstrapping process is built on the twin pillars of objective measurements and analysis of subjective measurements. While most aspects of objective loudspeaker and room measurements are agreed upon, it has been Mr. Toole who has developed and refined many of the methods for bridging the subjective listening test with the objective world of science by means of strict methodology. The result is the Holy Grail of science - repeatability. Accurate & comprehensive measurements combined with very carefully controlled listening tests have a synergistic effect because over time, the relationship between various factors becomes apparent when viewed from both angles.
In listening tests, it is essential that the listener is more than well trained and possessed of excellent hearing, they must also be talented. Some people are just better at listening in the same way that some are outstanding musicians or athletes and these talented listeners are sought out by screening. A good listener can determine within 3-5% on a standard scale with a standard deviation of 5-7%! Perhaps as accurate as an analog mutimeter! To maintain this accuracy, the listening environment is critical. All tests are done blind with differing loudspeaker systems level-matched and moved to exactly the same room position. Strangely, when you can see what you're listening to, you can't hear it! The sound qualities are diluted completely by looks, cost, brand loyalty and reviewer's opinions.
Mr. Toole then reached into his endless bag of power-point slides for more nuggets, taking his direction from audience questions. What was presented to us was a small part of a 6-hour presentation so there was some editing of content. For instance: Speakers are evaluated with 72 sets of measurements to determine their overall response that must be smooth off-axis as well as on-axis. A broad, low-Q resonance of a few decibels is more audible than a greater narrow high-Q resonance.
Phase shift is not audible. Harmonic distortion is not as audible as non-linear distortion. Power compression is also quite audible. When placing a sub, start with a corner. Mr. Toole has a spreadsheet available through e-mail to aid in placement and says that satisfactory results can be obtained by placement and using a one-band parametric equalizer.
We would like to thank Mr. Toole for sharing his time and knowledge with us and, late as it was, only wished it could have lasted longer than it did!
by Paul Reibling, Secretary/Treasurer Toronto Section
Return to the Front Page | Preview of this Meeting