On 7/5/20 1:31 AM, Michael Jarosch wrote:
Hi!
Hey,
If I may add a few comments in addition to what Fons already wrote...
Which border do I have to reach in my project that I should consider
switching to higher (second, third or more) order ambisonics? Is it the
number of output channels? Or is it: Always use the highest order you
can afford?
:-)
If Ambisonics is a good fit for you (read what Fons wrote) that would be
my initial reaction... After you have your piece rendered in, say, 5th
order, and if your rendering environment cannot do that (not enough
speakers) then you just truncate to the order you can render (it should
be the same except for the spatial resolution you obtain - caveat: the
"feel" - sorry, very unscientific - of the piece might not be exactly
the same precisely because the spatial resolution will be different).
These days I would not generate content at less that 3rd order
periphonic (16 channels). In my experience and in the context of concert
reproduction and if you have enough speakers, it makes a big difference
in localization accuracy and (specially) size of usable sweet spot.
For example, a dome of 24 speakers (12 + 8 + 4) should give (pretty)
good reproduction of 3rd order, but only (as Fons pointed out) if the
spatial arrangement of the speakers is evenly spaced around the audience.
In my experience f you have more speakers then 5th order sounds even
better, more accurate and bigger sweet spot (we have a 56.8 space at
CCRMA). I don't have personal experience with orders above 5th/6th, but
apparently the advantages of going above 7th order are tiny or
non-existent[*].
...
Having said that it would seem that it is always the case that higher
order is better.
Not necessarily... :-)
While higher orders will give you pinpoint accuracy in sound placement
in space if you are doing synthetic panning of mono sources, lower
orders will give you more envelopment and diffuse sources, which might
be important as well (those should be possible to achieve in higher
orders but I don't think currently available software does that).
I have been using mixed order for a while, two layers of 5th and 2nd
order - I started experimenting with that because I am capturing 2nd
order with microphones I built so that is a natural fit (spatially
processing the output of the microphone). The 5th order layer adds very
accurate positioning of sounds, and the combination sounds great - if I
may say.
Another quick test I did not long ago is 1st vs 3rd order rendering in
binaural and for me 3rd sounds better (for my purposes). Caveat: I only
tested using the Resonance binaural renderer. We are using 3rd order for
our online concerts (the "Quarantine Sessions" series at CCRMA) where
all performers are "panned" using Ambisonics.
Hope this helps...
-- Fernando
[*] this has more concrete information about HOA:
https://www.natashabarrett.org/Barrett_IRCAM_Report_v2.pdf
[**] it is, as always, all up to the artistry and technical knowledge of
the composer, I have heard 1st order pieces that are simply stunning -
go no further than the Epiphanie Sequence pieces by Joe Anderson...
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@xxxxxxxxxxxxxxxxxxxx
https://lists.linuxaudio.org/listinfo/linux-audio-user
