Home Propeller Head Plaza

Technical and scientific discussion of amps, cables and other topics.

  Register / Login

Re: A More Realistic Presentation...

Hi Tubeguy.

Another interesting subject!
I think it may be worth while to break the question into its two halves and consider each part separately at first.
Lets say you have some decent headphones, like Sennheiser hd280’s or other.
Now one has a case where one has no mutual coupling, no interaural cross talk (un intended sound from the left speaker reaching the right ear and the opposite).

Now, what does it take to really sound real here?
Rule #1 it takes a minimum of one more references than the plane you wish to describe.
For example, two channels allow one plane to be defined, right and left in the case of stereo.
On the other hand, you hear up and down, front, back and distance with only “Two” ear channels because your ear has a different frequency response for sounds coming at different angles. You are fully unaware of these “ques” because they are all you know, you have fully included them frame of reference, you can simply “tell” that a sound is coming from behind etc. Your ears are in effect, not just two channels.

So lets look at how sound is recorded.
Essentially all commercial recordings are made by taking individual tracks which are panned into the right “position” and eq’d, processed to suit.
There is no attempt to actually “capture” a stereo image of the real thing, it is tastefully “realistic” at its best.
The reason is there is no “Standard” way of capturing such an image, even a “Head” microphone is not the same.
If you see a band playing with a PA, you can see one of the problems in spades.
Stand behind the mix board during sound check and watch the VU meters. Each time the drummer taps a drum, you see ALL the VU indicators bounce.
Lets say your mixing down a live concert with 48 mic channels down to two channels.
Now rather than one arrival at your head like a normal drum would be, you have about 48 arrivals, each separated in time according to the mic to drum distance.
Now think of all the players at once playing and add in the sound from the floor monitors that goes back into the mics.
You don’t hear most of this as discrete events, rather, it is part of what makes a concert sound like a concert.
Try the fireworks recording I made with a microphone thingy which samples from one spot in space.
Use good headphones, close your eyes, you can hear the crackly fireworks high in front of you.
You can hear they are “high” because your ears know what that “sounds like”.

Remember your ears are not just two channels?
Well now lets look at the speaker end.
Out doors or in an anechoic chamber one finds that as you move a speaker farther away from you, it gets quieter and slowly a first order roll off appears at the high end, at 100 feet this is a few dB depending on humidity etc.

On the other hand with the processing power in the ear brain combo, it is normally “easy” to judge how far away a speaker is or point exactly at it when blindfolded. You ears are able to locate the origin of re-radiations, detect spatial discontinuities in the radiation and one range being radiated at another time and the resulting interference.
When you remove driver to driver interference, remove interference lobes, reduce time dispersion, then ones ability to locate the speaker are greatly reduced. All one can say is more like it’s “that direction”.
Driving a stereo pair with a mono signal produces a very concrete center image.

Like the article points out, In a room, one encounters unavoidable room interactions. Some, like the channel to channel comb filtering at lower frequencies are things you don’t consciously hear because your ears also hear the phase rotation which accompanies the amplitude notch. Likewise room gain is a room construction and size dependant variable which when present is a blessing not a curse (assuming you have an EQ to flatten the gain).

Serious (so far as preserving the recoded image, real or realistic) room effects happed when the radiation from the speaker bounces off the floor, ceiling, walls and then reaches the listeners ears, mixing with the direct sound.
This also causes cancellation notches but as with mutual interference, one doesn’t hear these as a discrete thing. What you hear is a mixture of direct sound which (to the extent the speakers preserve time) which contains the information from the recording, mixed with the reflected signals which arrive scattered in time and magnitude.
This reflected sound “sounds like” the original except much less distinct, when mixed with the actual signal, acts like “noise” so far as your ability to derive the original information.
With voice intelligibility or stereo image the ratio of direct to reflected sound (all other things being equal) is related to “goodness”.
A large panel speaker can have good directivity, reducing room effects but can also have self interference in the radiation pattern.
A Ribbon can have little self inference side to side, have significant directivity in the vertical plane reducing room reflections but often has limited output.

Personally for the output needed for “real” sound in larger spaces, CD Horns seem the best bet but they are the hardest to integrate into a system because of self interference at crossover and time issues driver to driver.

I was able to re-build an Altec A-7 a couple years ago (new drivers and crossover and minor cabinet mods) that was able to reproduce a square wave (and had about an octave more and flater response than the original in each direction) when done.
I am curious to see how your RCA’s come out, they have real potential.
Best,

Tom Danley



This post is made possible by the generous support of people like you and our sponsors:
  VH Audio  


Follow Ups Full Thread
Follow Ups


You can not post to an archived thread.