Hi Ray,
On 05/22/2013 01:24 AM, revery wrote:
Hello jörn,
Thinking about what you say here, is this working by having pure M
from the front and S from 90 degrees to the side, effectively
'mixing' the M S signals in the air as they reach the ears/brain?
(Maybe I'm thinking about this too much, my brain is hurting.)
Maybe :)
Think about it this way: MS is a subset of Ambisonics, effectively
missing the front-back and up-down information. So we can use it as an
Ambi mic:
The Mid signal gets panned where I need it. The Side signal is then used
to give it a bit of width. For a frontal source, it will be fed to the Y
channel only. Note there is no pressure component W from this signal.
The only slight complication is that your side signal is not coincident
with the main microphone, so you have to watch out for your overall image.
If so,
is there significant distortion/corruption of the effect from the two
ears receiving different variations of the M and S signals?
There is no coloration other than what's inherent in Ambisonics.
I realise
that ear crosstalk effect is an issue with standard two speaker
stereo as well, but the consequences with this kind of signal
presentation seem to me to be quite different. As part of this, if
the head turns say 45 degrees to the left, the ear difference would
seem to be at a maximum, with the left ear receiving a significant
amount of the opposite lobe of the figure 8 with little cancellation
effect from the M in front. Perhaps this is all part of the plan….?
Like I said, there is no magic mixing in the air.
Ear crosstalk is not an issue in Ambisonics - we try to recreate a sound
field, and the head is in there like it would be in the original field
at the concert. So "ear crosstalk" is very much part of the experience.
What you will hear is pretty much a widened version of the M signal.
It's not strictly orthodox, but it works, as long as you don't overdo it
and you get your delays right.
For a less confusing way of mixing MS spots into Ambisonics, you can
render them to Left-Spot and Right-Spot with a conventional MS matrix,
and then pan those individually. Because they are coincident, there will
be no comb filter artefacts, no matter how close you pan them.
The reason I treat the S signal separately is because I usually work in
third order, so the M mic is pretty sharp, and the S mic is then fed to
first order only.
The problem with this kind of mixed-order hackery is that the sound
might shift when you truncate orders on playback (if no 3rd order system
is available), but I already have this problem: my main mic is a
first-order tetrahedron anyways. So I just watch out for it while mixing
and frequently cross-check at lower orders to arrive at a useful middle
ground.
Best,
Jörn
--
Jörn Nettingsmeier
Lortzingstr. 11, 45128 Essen, Tel. +49 177 7937487
Meister für Veranstaltungstechnik (Bühne/Studio)
Tonmeister VDT
http://stackingdwarves.net
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
