The agenda and the report from the previous meeting were approved without amendment.
AES5id Review of AES5-1998: AES recommended practice for professional digital audio -- Preferred sampling frequencies for applications employing pulse-code modulation
Prince noted that this document required reaffirmation. There were no objections.
AES2 Revision of AES2-1984 (r1997): AES Recommended Practice -- Specification of loudspeaker components used in professional audio and sound reinforcement
Due to time constraints, AES2 was proposed for reaffirmation without objection.
The SC-04-03-A task group will continue with its revision which was reported as near completion.. There was a desire to remove Xmax as currently defined but process prohibits easy removal. B Olson to be added to SC-04-03-A task group. D Button reviewed a power test signal proposal to SC-04-03-A task group. The signal will be more completely described later in the SC-04-03A program.
AES19-R. Review of AES19-1992 (r1998): AES-ALMA Standard Test Method for Audio Engineering -- Measurement of the lowest resonance frequency of loudspeaker cones
There have been no objections in meetings or on the reflector to removing this document from AES jurisdiction. ALMA has accepted responsibility.
AES-X72. Acoustic Center of Loudspeakers
Munich meeting discussions indicated a misunderstanding of group's findings and report. Prince reviewed intent of their report. W. Ahnert agreed that acoustic center is an undefinable concept in practice and that he would like to not use the term at all. D Gunness agreed that the term shouldn't be used since it implies that such a point exists.
Ahnert outlined the point of rotation and reference point concept. This was felt to be useful. Prince will revise the draft to include Point of Rotation and Point of Reference.
AES-X103 Large Signal Parameters of Low-Frequency Loudspeaker Drivers
W Klippel will provide a report to the working group. After feedback period (1-2 years) on various large signal parameters now in practice, there will be a recommendation on appropriate parameters to include in standards.
AES-X129. Loudspeaker Distortion Perception and Measurement
J Stewart provided bibliography of ASA reports. Stewart feels that we don't have a clear understanding of all the work that exists inside or outside the audio world. Prince reviewed purpose of project. Prince called attention to R Cabot references. Cabot felt that these references had bearing in general psychoacoustics. He noted E Benjamins work on perception of distortion and jitter and F Toole�s review paper including non-linear distortion. Voishvillo felt that in his 1985 paper Toole was observing the unimportance of non-linear distortion. Prince noted that in conversations, both Toole and S Olive felt that the investigation may be valuable.
Prince felt that both thresholds and certain levels of perception and acceptability were desirable. E Geddes discussed the relationship between higher orders and perception. Prince noted that he had proposed that the AES Technical Council on Psychoacoustics hold a session on non-linlear distortion. A task group was formed with Temme and Stewart as leaders and members to include Prince, Button, Geddes, Cabot, Berry, S Bech, D Keele.
Prince asked that task group define its initial scope as literature review and pre-test. Stewart noted the negative response he received from many people he spoke with...that the task was insurmountable and that after an enormous amount of work, the results wouldn't be useful. Prince noted that it's the task groups responsibility to see what is achievable. Prince noted that the end goal is to find meaningful distortion data. If the group can only provide a report on the limitations of certain distortion measurements, then that should be its goal. Gunness doubted it would be possible to define a clear acceptability threshold for distortion but it would be useful for the group to find measurements that correlate with perception even if a threshold is not defined. Voishvillo noted that it would be useful if we could define better test signals.
Geddes noted that B Moore commented that there is no current literature available that adequately correlates distortion measures and perception. Geddes felt that any signal based method would be unlikely to produce results, but proposed an alternative method.
Voishvillo noted two different things, identification of non-linearity and black box symptom measurement. Geddes disagreed.
Klippel noted V Schmidt had the task assigned for PhD of finding thresholds and correlation for measures and perception. The frustrating thing was that the target was not clear. Should one provide numbers or dependencies? The goal would probably fail with numbers because the system is too complex and the signals are too varied. She was not able to document all the dependencies but they are worthwhile thinking about. Organ music is very sensitive where rock music is not.
Voishvillo thinks that for low frequency drivers, it could be achievable but for full range systems, the system becomes very complex.
D Blore suggested starting by measuring several speakers and listening to them. Button responded that it is difficult to separate linear from non-linear distortion. Since every loudspeaker starts to sound bad at some point, one can take a single loudspeaker and record music playback at progressively louder levels. After normalizing the playback level of those recordings, subject them to listener testing. Then see which one of all the measurement techniques correlates to the listener data.
Button volunteered to measure speakers at various levels, normalize and run quick statistical group to provide first analysis.
[Secretariat note: Harris has subsequently accepted the position of vice chair.]
The next meeting will be held in conjunction with the AES 114th Convention in Amsterdam, the Netherlands.