20 Jan
2009
20 Jan
'09
1:32 p.m.
Fons, you've described my crude experiment with this. Left, Centre, Right,
and a complete IR for 'global' as the tail. I'll have to experiment some
more, and possibly truncate the 3 short irs further, but i'm getting the
idea in relation to a more realistic representation, however crude it may
be.
A question here gentlemen.
With an WXYZ setup, what do those represent? Left/Right/Front/Rear?
So any signal in would be a composite of position based on the strength of
gain between 2 or more points?
Is it a 'box' in effect, so any point represented would be a positional
calculation based on all 4?
Or do i have this wrong?
I'm off to try out the YorkMinster conf file, and see if that gives me any
listening clues.
Alex.
On Tue, Jan 20, 2009 at 9:03 PM, Fons Adriaensen <fons(a)kokkinizita.net>wrote;wrote:
On Tue, Jan 20, 2009 at 06:34:27PM +0100, Jörn
Nettingsmeier wrote:
to clarify things, there are no "short"
or "long" IRs. what fons means
is: the early reflections are the most characteristic aspect of a room,
and they affect localisation the most.
I don't think I can really agree with the first part of this
statement, but the second part is certainly true.
therefore, if you want to have
ultra-realistic reverb, use an IR that was measured with the speaker
where your intended instrument is and a soundfield microphone where the
listening spot is. of course, in practice this is not done.
so instead you use one reverb IR instead. it can be short (tail
truncated) to save CPU, because the tail is decorrelated (blurred) and
does not provide localisation cues, hence it would be wasteful to render
it in b-format.
What I suggested is:
* use a short IR, containing the early reflections, for each
source position, or a at least a small set of them for
different areas (e.g. left, center, right),
* manipulate the delay of this in function of source distance
* use the same long IR, containing the reverb tail, for all
sources.
Of course, the real reverb tail is different for each source,
but you can't hear the difference. Only its statistical
properties (such as the reverb time, etc) matter.
But I wouldn't say it is wasteful to render the reverb tail
in B-format, on the contrary, doing that makes it very realistic.
if you can make it to lac2009, let's talk
this through over pizza
(that's how i learned my first steps in ambisonics from fons, and it
works suprisingly well).
And pizza is easy to get here.
as i said, i would recommend against attempting
such a split reverb
method, because very likely things will go haywire at some point.
better bribe fons :)
I have some jconv configs containing separate short IRs for
different source positions, and a single reverb tail. They
need some reworking (*), but when you are in Parma we can
play with them.
(*) For example, removing the direct sound can be a bit
more complicated than just cutting it off. In most cases
the LF response of the direct sound continues during the
early reflection period (at very low level, you don't see
it in the IR). Just cutting away e.g. the first few ms can
leave enough LF energy from the direct sound to result in
a bass-heavy IR.
Ciao,
--
FA
Laboratorio di Acustica ed Elettroacustica
Parma, Italia
O tu, che porte, correndo si ?
E guerra e morte !