Aspects NOT Captured by Digital Sampling
By Richard J. Weisenberger, forwarded by Karl Petersen
[ Editor's Note: The message which follows appears to be a forward
[ from the Pipe Organ mailing list...
From: "RICHARD J. WEISENBERGER" <RJWEIS00@UKCC.UKY.EDU>
Through: piporg-l@smtibm.albany.com
Subject: Tonal Aspects NOT Captured by Digital Sampling
Although I maintain that the general public could be fooled by someone installing the speakers of a state of the art digital electronic organ behind a facade of dummy display pipes, there are certain aspects that even digital sampling can't capture. If the electronic manufacturers wanted to, they could even sample the sounds of trackers or electropneumatic valves in action, to further heighten the realism, but even this would not totally capture every aspect of pipe produced sound. More people need to be made aware of WHAT to listen for rather than continue to aimlessly argue moot points.
The most often overlooked aspect is that pipes and speakers radiate sound in a totally different manner, and this single aspect would be very difficult (if not more expensive than actual pipes) to reproduce. It's the SPATIAL aspects of sound that really set pipes apart from electronics and give pipes their "live" quality.
Unless you use large speaker arrays, constant directivity horns, or small (and inefficient) dome tweeters, all speakers become increasingly directional with an increase in frequency. Frequencies below 200 Hz are for all intents and purposes non directional for virtually any speaker (since the driver is small compared to the wavelength).
Pipes, on the other hand, maintain much of their directional properties independent of frequency, as they are always a radiator of a given fraction of a wavelength. Stopped flue pipes are, for all intents and purposes, non directional radiators at any frequency. The reason for this is that the majority of the radiation takes place at the mouth, which is very small compared to the wavelength (the entire pipe being less than 1/4 wavelength).
Open flue pipes behave as two radiating sources in phase, separated by 1/2 wavelength. This is true whether the pipe is 32 ft. long or only 3/4"! Thus, a 32 ft. pipe at 16 Hz is a directional source, just as are the top octave pipes of a 2 ft. flute! This is the reason that the sound appears stronger at a distance from certain large pipes than up close.
The closer we get to the pipe's axis (directly above or below) the closer we are to the fundamental's null point. The sound below a large open pipe becomes thin and mainly composed of the few weak harmonics that are present. As we go farther away, the fundamental radiation fron the mouth and top become more and more in phase with each other and the sound becomes full-bodied. This is the best (acoustical) argument for mounting open flue pipes vertically, rather than horizontally as en chamade reeds.
Reed resonators radiate from the open bell end, which is very small compared to the fundamental's wavelength, but large compared to the upper harmonics' wavelengths. For this reason, en chamade trumpets appear very bright directly on axis in front of them.
Since speakers appear as fractional wavelength sources at low frequencies and as multi wavelength sources at high frequencies, there is no way they can duplicate the spatial aspects of actual pipes without having a speaker "system" for each note!
The "electronic pipes" several electronic organ manufacturers tried a couple decades ago was an attempt to recreate this aspect. As we all know, they fell short of actually doing it.
It's time we educate the public as to the REAL differences between these two instruments, as digital electronics is narrowing the gap in most other aspects. The spatial aspect is one significant gap I cannot foresee being narrowed within the foreseeable future.
>>
[ Editor's Note:
[
[ I feel compelled to stick my two cents worth in here. This article may
[ explain some of the experience of being in a hall with a large pipe organ --
[ to me its awe inspiring. However, I have some CD's which capture very
[ well the experience of listening to one of these organs -- if you are
[ not allowed to turn or move your head. It would appear from the author's
[ analysis that even the "fixed vantage point" listener will observe different
[ hall acoustic effects for each rank -- in fact for each individual pipe.
[ I'd be really happy if I could "compute" a large theatre organ, or for
[ that matter even a small band organ, from samples, models, or whatever,
[ accurately enough that the experience, as observed with a set of good
[ headphones from a "fixed vantage point" was close to the real thing.
[
[ Jody |
(Message sent Thu 16 Nov 1995, 23:36:41 GMT, from time zone GMT-0700.) |
|
|
Mechanical
Music Digest™ Archives
