When you close your eyes, you already know that Willie Nelson and his band are not in your living room. How can you be sure of anything?
The article linked below measured acoustic sound power using a spherical array of 32 microphones in an anechoic chamber, with professional musicians playing a large selection of orchestral instruments.
Their definition of "sound power", stated a bit simplistically, was the average sound pressure across those 32 microphones, normalized to 1 meter distance from an instrument centered as well as possible within the sphere. They asked each performer to play two samples of each semitone in the range of the instrument, with one sample as loudly as the performer could produce with a musically useful tone, and the other as softly as musically possible. They called it Lw, with units in dB.
The point is that acoustic power and loudness are the same thing, as long as the loudness is integrated non-directionally. This is the issue with loudspeakers--conventional designs can't produce the same sphere of sound in an anechoic chamber. But in an acoustically reflective space, like our listening rooms, the reflections will fill up the space outside the beam. If a recording has a recorded room ambience that is similar to what our brains expect in the room the eyes are seeing, the illusion of live sound can absolutely be sustained, if the listener is capable of the willing suspension of disbelief while sitting there looking at some loudspeakers.
What won't be possible is getting the same directional cues that live performers might produce in any given space, but that doesn't mean the directional cues that are presented won't be sufficiently realistic. It's not an easy experiment to conduct. Maybe my willing suspension of disbelief differs from your own.
There were some basic trends the authors observed--wind instruments get louder as they get higher while string instruments not so much, some instruments were capable of much greater dynamic range than others (which they defined technically as the difference in sound power, in dB, between the soft and loud notes, averaged across musically relevant pitches).
I note for personal interest that the tuba measured at 122 dB for the loudest and 71 dB for the softest, for 51 dB of dynamic range. It was the loudest of the instruments they measured
I doubt that range has ever been issued on a recording, and no sound system on earth can make up for that lack. The clarinet, with the greatest dynamic range of the instruments they measured, ranged from 53 to 110 dB. For classical instruments (which predate the tuba), they also measured them normalized to Beethoven symphonies, and the range was much less, but still over 30 dB for clarinets.Edit: I just dug into the supporting materials, and discovered that the tuba used was a B&S 3100 F tuba. I happen to have one sitting in my living room. These are orchestral F tubas, large for F tubas, but dwarfed by any orchestra-sized contrabass tuba in B-flat or C. (3100 F tuba on the left, Hirsbrunner HBS193 B-flat orchestral 'kaiser" contrabass tuba on the right. The Hirsbrunner is louder, but not by much. It is deeper in tone, assuredly.)
(End of edit.)
I also note that standing with one's head right in the bell of a tuba, while a professional performer is playing as loudly as possible, will be...uncomfortable. But tubas in orchestras point up, not directly at the audience, and no audience member is going to be closer than 30 or 40 feet. Tubas in marching bands point forward, but they mostly play outdoors from a substantial distance.
And for my own personal calibration, they measured an Lw of 108 dB for timpani, and I have measured exactly that 108 dB SPL (unweighted) sitting a meter in front of the timpani when playing loudly in an acoustically damped rehearsal space. (I was collecting data to support a request for a plexiglass shield.)
I don't know to what extent we are actually disagreeing, but it gave me an excuse to dig up this paper
Rick "who has that copy of Shotgun Willie on vinyl, bought back in the day" Denney
Last edited:
