Be careful, merely neglecting ambisonics in general for pointsources
includes higher order ambisonics, which are *well* suited for localized
sources as of the third order and up (although I have not heard any
ambisonic decoding beyond the third) in spherical 3D sound.
Also, unlike what you're saying, multidimensional audio should not
distract people from the screen. If that is so, then the sound engineer
is immaturely using the technology! The key is a tight interplay between
the audio and the image. A well balanced 3D mix will keep the viewer
fixed on what is happening in front (as long as the image is interesting
enough in relation, of course). Moreover, Lucasfilm, if my memory serves
me right, extensively used rear channels for cinemas to produce laser
sounds among other attention catching sounds coming from behind. Never
heard that this ever bothered anyone watching the Star Wars series.
Concerning the problems, first of all, first order ambisonics suffer in
spatial accuracy. Another problem I see are the technical difficulties
higher order ambisonics face. One important factor is the decoding for
irregular layouts, which is still not very well solved for higher
orders. Using a regular decoding for an irregular layout could work, but
also defeats all the effort and beauty out into the mathematics involved
- and you have no quality guarantee! If you think about how many people
take care to set up their 5.1 in the exact positions as specified by the
standard, you can easily imagine how hard it will be to convince your
wife to place 20+ speakers in exactly equidistant positions in the
living room. Also, cinemas are not spherical and would require to place
speakers in custom positions.
Also, there are no higher order microphones - aside from the eigenmike,
which does not come with a higher order ambisonic solution out of the
box (i.e. if you cannot write a higher order encoder yourself in matlab)
as far as I know.
All in all, while first order ambisonics are not sufficiently accurate
in their spatial resolution, higher order ambisonics are still
relatively unknown outside the research community and still require much
work to mature. Too many open parameters prevent a standardization for
higher order ambisonics and everybody is implementing their own encoder.
The industry is moving towards object oriented encoding of 3D
soundtracks anyway. This is perhaps the least elegant, but the most
accurate, as every sound is stored in isolation of the others, with
exact meta information of its spatial position. Theoretically, you could
take this soundtrack and render it over any system you, be it
ambisonics, higher order ambisonics, vbap or wave field synthesis among
possibly others...
On 5/15/13 4:27 PM, Augustine Leudar wrote:
I think the thing is people think that ambisonics is some incredible magic
spatialisation technique that surpasses all others - yet is so complicated
that nobody can understand it except for a few mega nerds and
mathematicians that speak in mysterious riddles whenever you ask them
anything and therefore this feat of incredible genius is doomed to
commercial failure. This opinion it seems to me is help by people who
havent had a lot of hands on experience actually using it in a proffesional
context (ie actually producing film, theatre, sound design etc) - in fact
thats what I originally thought about . A lot of people seem to equate
ambisonics with "surround sound with height " as well (hands up I was
guilty of that too once) . Basically if I was going to design a full 360
degree soundtrack for a film (which will probably not happen by the way -
the last thing film producers want is people turning away from the screen
and looking over their shoulder because a dog barked behind them - roll on
holospheres !) - the last thing I would use ambisonics for is point sources
- ideally I would use an individual speaker in the right position as the
point source, and failing that some sort of amplitude panning . I would use
it for panning sometimes - and for some weird phase effects and a few other
things , ideally you will mix ambisonics with other spatialisation
techniques and then render your multichannel audio tracks.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://mail.music.vt.edu/mailman/private/sursound/attachments/20130515/eca0cdeb/attachment.html>
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
