Camera noise idea..

[tom*]

This is just a thought, but has anyone had the idea of using the principle that one wave will cancel out another if it is it?s exact opposite? E.g. \/\/\/\/ + /\/\/\/\ = -------

A couple of practical examples of this principle that I have come across (there will be other obvious ones probably) is in cathedrals where it is fairly common to improve the acoustics by reducing the echo from the large space by relaying the sound from a choir to speakers in the high ceiling with an inverted wave, thus neutralising the wave that is reflected from the ceiling and heard as echo. I also heard of a prototype luxury car that cut road noise in a similar way. Cheap karaoke machines use the assumption that the vocal track is in the centre (both stereo channels equally) and therefore combining the channels cancels out the vocal.

What I am getting at is, perhaps of the (sync) sound track could be matched with a (sync) recording of the camera whirring away from a much closer mic. The level could be mixed until the track with the predominant camera sound had the same sound level as the camera in the background of the dialogue mic albeit inverted, cancelling out the unwanted sound (in principle).

Perhaps in practice the sound characteristics change too much with Doppler, mic differences etc? Would be a sweet solution though


------------------

[Cranium]

I'm inclined to think it won't work, but it would be worth a shot. Maybe using a piezoelectric transducer ON the camera, and try it. Then again, there's the delay factor to the actual mic.
Seems like with some slipping and sliding in CoolEdit or something, you might be able to diminish the motor sound a little bit.


------------------
Mikel Z

[Matt Pacini]

Tried it.
Didn't work.

In theory, it works, but in practice, sound is incredibly complex.
The problem is, that the sound of the camera is bouncing all over the room you're in, and those reflections all end up at the microphone at different times, with different phase relationships, etc., and there is just no way to match that wave, to cancel anything out, because it's mixed in with the audio you WANT to hear.

If you record on the camera directly, then, or later, to the human ear, it all sounds like camera noise.
But to the recorder, they are all slightly different, because of the phase problems, etc.

Maybe under really, really controlled situations, it would work, but if you could control your acoustics that much, you would be able to get a quiet camera.

Matt Pacini

------------------

[Actor]

Take two identical unidirectional shotgun mikes, one noisy camera and one actor. Let A be the position of the camera gate, B the position of mike 1's element, B' the position of mike 2's element and C the position of the actor's mouth. Position the two mikes so that the distance BB' is a small as practical. Now measure the angle ABC and point mike 1 in the direction BC. Then construct angle AB'D so that the measure of ABD is equal to the measure of AB'D and so that C and D are on opposite sides of the line AB. Point mike 2 in the direction B'D. Now null out the camera noise.

What this does is rely on the directionality of the mike as the first step in eliminating the camera noise. Positioning the second mike so that it picks up camera noise from the same relative direction as the first provides the signal for cancelling out the camera noise (its phase will have to be reversed). Pointing the second mike away from the actor is necessary otherwise we will also cancel his dialog. This should also eliminate any echos of the camera noise since the echos will also arrive at both mikes from the same direction and having traveled the same distance.

Of course you will have to make sure there is no source of unwanted sound along the line B'D.


[This message has been edited by Actor (edited February 04, 2002).]

[MovieStuff]

I understand the theory behind this. Have you tried it? Also, it would seem that the actor would have to stay in one spot for this to work.

Roger

------------------
Roger Evans
MovieStuff
http://www.afterimagephoto.tv/moviestuff.html

[Matt Pacini]

<BLOCKQUOTE><font size="1" face="Courier, Arial">quote:</font><HR>Originally posted by Actor:
[B]...This should also eliminate any echos of the camera noise since the echos will also arrive at both mikes from the same direction and having traveled the same distance...

B]<HR></BLOCKQUOTE>


Like I said, in theory it all seems to work, but it just doesn't.
I've tried it all, and i have lots of studio experience trying a zillion different micing situations, and the complexity of sound in regards to phase relationships, just cannot be overstressed.

"The echo" is not just one singular event, it is an astounging array of waves of different frequencies, bouncing off all surfaces in the room, and they will NOT arrive at TWO different places at the same time.
First of all, for your scenario to work, BOTH mic diaphrams would have to be in EXACTLY the same place at the same time, which not only impossible, but would defeat the whole idea anyway.

Second of all, different frequencies are absorbed differently by different surfaces, high frequencies have less energy than low frequencies (that's why you hear mostly jsut bass coming through the wall when someone is listening to music loud in another room), and different elements of the sound are doing different things, as these different frequencies bounce all over the room.
Sound is not just one thing.
It would be like taking 20,000 ping pong balls, filling them all with different amounts of weight, so no two were the same weight, filling them with helium so they would all bunch up together in a ball in the middle of the room, then lighting an M-80 in the middle of all of them, which would send them radiating in all directions instantly.

... now you expect all of them to magically end up back at the microphone at the same time...
Not likely.

( I realize this is an exaggeration, but it does demonstrate the complexity of audio, and that is why it drives filmmakers nuts, because they expect it to behave like light, and it just doesn't!)

Matt Pacini

------------------