|
Posts: 8192
Joined: July 4, 2002
|
Hi Bill
Hey that’s right, one doesn’t need or often even want $full bandwidth$ for “voice only”.
Also, often those old systems had to withstand and operate in very high noise levels and explosions which greatly limited what could be done reliably. I don’t recall the frequency bands off hand but the MTF measurements used in the STIpa predictions were only the voice band and weighted by amplitude to voice spectrum.
http://en.wikipedia.org/wiki/Speech_transmission_index
While not yet used for anything ‘hifi”, I am sure Modulation Transfer Function measurements can point the direction “if” preserving the information within the signal is desired (and this not a requirement for enjoyment only resolution).
If you have ARTA, you can play with them.
Unlike the STIpa measurements, these are not weighted and cover a much wider scope and modulation rate. The down side is they do not produce a “number” or quantity but the curves can be compared from on case to another, the higher rates of modulation (preservation of signal information) the better.
The reasoning is that “information” like ineligibility is carried within the dynamics of the signal and no other measurements relate directly to dynamics.
The idea is that with various different tones, say 1Khz, the tone is amplitude modulated starting at a very slow rate and ending at say 30 times a second. It is the depth of modulation, the ability of the speaker’s signal to turn on and off rapidly, that governs how much of the desired input signal is reaching the microphone.
This is repeated at a number of base frequencies.
In optics where it originated;
http://photo.net/learn/optics/mtf/
What is clear to me is that while MANY things can corrupt the MTF measurements, that when this aspect is preserved, one has a more realistic reproduction. When you shave off enough of the various warts, preserve enough of the signal at the LP, one can feed the same voice signal to both speakers, sit at the listening position and only hear a solid, real (like a speaker there) phantom image with NO trace of a right and left source.
Like I said too though this is not directly tied to enjoyment, one can have very enveloping sounds which have no intelligibility, for example a choir in a large church or have loudspeakers which paint a wall of sound but produce a mono phantom as diffuse and or with a very obvious right and left source of sound.
Concerts are enjoyable too but the large line arrays have so many different sources and path lengths that the quiet parts of the MTF modulation are filled in with late arrivals and it’s no surprise if you stand in the mix booth and listen to the system (normally in mono) sometimes, you can sort of tell there is a mono phantom image when ideally, that’s all there should be.
That’s why the stadium business has really taken off, that’s all that was available previously was large scale concert stuff..
Horns can have a big edge here, but you know that haha.
Here are a couple big horns, pop on some headphones. Like the SH-50 etc, you can walk up to these big ones, put your head in the mouth and move it around and you never hear that there is more than one source of sound floating somewhere in front of you, radiate as a single source in time and space over a broad band.
The second system is very powerful and has 6 subwoofers and is -3dB at 28Hz, at onset of limiting, can produce 106dBa slow at 800 feet.
https://www.dropbox.com/s/lqeg3cf0daqv9ti/20120726114748.mts
https://www.dropbox.com/s/oyosfc3adc6j1du/20130723135350.mts
What’s new? How have you been?
Best,
Tom
|
