fons, aaron, thanks for your replies.
On 03/18/2011 01:51 AM, Fons Adriaensen wrote:
To behave as a line source rather than a point, they have to be larger
than the wavelength, this imposes a lower frequency limit. On the other
side, once the wavelenght is comparable to the distance between the
drivers they will start to appear as a collection of discrete sources,
and this limits the range at the high end.
Combining these two limits results in a rather narrow frequency range
for any practical array.
well, with a 3.5m array (which is quite a standard length), you're down
to 100 hz, which is Pretty Darn Good.
as for the upper limit, with some intelligent waveguiding and other
magic fairy dust, you can get the individual elements to emit something
close to a plane wavefront already. which means they will combine in a
more effective way. hence, hf level loss due to dispersion over distance
is less of an issue in practice than the theory (assuming
omnidirectional tweeters) would imply. in fact, for many systems the
opposite is true: the hf beam has an annoying tendency to constrict,
resulting in even more focused coverage on-axis than desired.
And anyway the extent of the near field is
proportional to the array size - once you are far enough they will
again appear as a point source.
true, but there is still great potential. line arrays can be made to
work ok'ish over more than a hundred meters. of course, there is a more
or less severe impact on timbre, but: concertgoers will perceive the
gradual degradation as natural. it's not a near-field listening
situation, we don't want to create an illusion of extreme proximity to
the sound.
a rocknroll concert heard at a distance of 100m will sound like a
rocknroll concert heard at a distance of 100m, with perfect fidelity :-D
so for practical purposes, line arrays do change the rules for large
ambisonic rigs:
* you don't necessarily have to resort to a "controlled opposites",
strict cardioid decode, in order to avoid the collapsing of the image
into the opposite speaker(s) for listeners close to those speakers,
because the "good speakers" retain more level close to the "bad
speakers". this will improve localisation in the inner listening area
considerably.
* with more uniform coverage over distance, we can create larger
listening areas where speakers still contribute evenly to the sound
field in terms of energy.
* we can reach listening area sizes where all perceptual assumptions
that hold for usual ambisonic reproduction break down completely,
because while the speakers contribute evenly in terms of energy, they no
longer do so with respect to time.
It's not exactly true that rV/rE 'focus on the sweet spot', they are
metrics that can be applied anywhere, and that apparently map well
to perception (respectively for low and mid/high frequencies).
what i meant is: we tacitly assume that the contributions of all the
speakers will blend within a given perceptual window.
phase sensitivity corresponds to the LF band, and more or less to the
effect called summing localisation, with a time difference window of <
1ms (depending on frequency).
rE sensitivity covers the rest, but only if the contributions of the
individual speaker stacks arrive within the "haas window" < 30ms, i.e.
before you begin to discern them as separate auditory events.
large arrays will have areas where these assumptions no longer hold.
so there are 2 area constraints:
* the area of perfect reproduction (aka sweet spot), which is quite
small, a function of the frequency and order, and which allows us to
optimize for phase and energy gradient separately.
* a larger area of usable reconstruction, where rV is no longer
meaningful due to the phase errors introduced by run-time differences,
but where plausible rE recombination still happens. still a function of
the ambisonic order.
this is where current experience with ambisonics ends (unless yours
doesn't, in which case i want to hear from you!).
large systems introduce a third area, where runtime differences are so
large that you cannot expect meaningful recombination of the individual
speaker stacks to one coherent image, at least not according to the same
perceptual mechanisms we assume for small systems.
this is what i'm currently thinking about: where is the boundary at
which ambisonicitiness breaks down for good, regardless of order?
In case you have a 'pre-echo' not corresponding to the source
direction, you could still apply the rE metric to each of the
'pre-echo' and the 'main' part separately - I guess this would
be the first step in analysing such a situation.
very good point. so i could probably start by analysing the pre-echo and
the "correct" cue separately, treat those as two loudspeaker sources,
and apply some known theory about directional masking (i.e. how much
louder does a second sound event need to be in order to mask the
directional cue from a correlated earlier sound event).
the problem i anticipate is that those two theoretical sources will be
blurred already, maybe to the point where they each fragment into
separate auditory events.
it remains to be seen whether that is good or bad for the auditory
illusion we're after, and if there's some meaningful listening enjoyment
to be had in this outer area.
funnily enough, at the sizes i have in mind, we have to think about
relativistic groove: there is no concept of simultaneity any more, and
if the dj decides to put the bass drum due north and the shaker due
south, the rhythm will "swing" in various ways depending on where you
are. so we are certainly beyond hi-fi reproduction that will work for
any type of content.
another aspect where large systems are instructive: the perceived size
of virtual sources and how to control them. others have shared their
listening experience of john leonard's five-metre geese before - you can
create airplane-sized geese with line arrays. but more interesting is
the question how to shrink them again, to plausible proportions.
fact is, we have very little control about the perceived size of
sources, and they behave in just the wrong way: the larger the array
diameter, the larger the sources appear, when it should be just the
other way around. easy to explain, but often hard to tolerate.
How much 'pre-echo' can be tolerated in function of its relative
timing, level and direction will probably depend very much on the
nature of the source material. Anything with distinct envelope
transients will probably reveal it more than more steady signals.
As far as I know there is little definitive material about this,
except for some simple cases.
there is, but for the case of two loudspeakers only afaik.
--
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
