On 2017-01-09, Stefan Schreiber wrote:
- Backward-compatible (to stereo) forms of Ambisonics are very
probably possible.
They are not "probably possible". That's what the BHJ version of UHJ
*is*.
Yes. But you still have to put L/R + 1 or 2 ext. channels into some < stereo
MP3 or AAC file.
Yes. There's a lot of haziness around how to do that optimally, at
least. Cf. the discussions around Vorbis's multichannel capability.
There ambisonic theory is explicitly mentioned.
Our aim would be to be backward compatibility to common file and
stream stereo formats. This is not completely trivial, but can be
done. (Extension channels "buried" in file or container format. We
don't want to break existing stereo decoders...)
There's then historical ambisonic theory even for that. :D
In the UHJ hierachy we also have SHJ and THJ. Those were originally
meant as compatibility formats for augmented AM and FM stereo radio. SHJ
as a "two and a half channel" format exists because at the time sattling
a full-bandwidth third channel onto a typical radio broadcast was
technically impossible. Given the limitations of analogue electronics.
However, a half-band channel could be put on, and so SHJ was borne.
At the same time, what you'd really want in a compatibility format would
be just a third, full-bandwidth channel. That's what THJ then
represents, as the full-fledged third tier of the UHJ hierarchy. It
starts with BHJ, and then ads a suitably (complex) matrixed third
channel (T), so that the whole system inverts into pure pantophonic
B-format (a bit of a misnomer because cylindrical and spherical
harmonics can't really be so exchanged, but still close enough).
So, if you want to get complete compatibility, we have the signal format
already. The question is how to convey it within common audio formats.
That is then not much of a problem either, if you think about it. Pretty
much the only formats we have to think about are RIFF WAVE (.wav) and
MPEG2 layer 3 (.mp3). Just maybe MPEG2/4 AAC (ISO/IEC 13818-7:2006).
Wav, especially in its modern versions, supports hiding T. It also
supports a media tag to discern between just UHJ encoded stereo with a
hidden channel, from ordinary stereo with one hidden channel. We already
used it to encode B-format, in Richard Dobson's .AMB format, derived
from WAVE_FORMAT_EXTENSIBLE
(https://en.wikipedia.org/wiki/Ambisonic_data_exchange_formats).
AAC is pliable as well. Of MP3 I'm not too sure, but I think it might
be. (I'm too drunken and tired to delve into that now. If you want
certainty, remind me later. I do think I have the relevant ISO standards
laying around somewhere on my disks.)
UHJ was always a hierarchy scheme. Even so you have to find some space
for T and Q channels. It is all just about some "practical space
problems", not Makita theory itself... Just IT and standard related
stuff, not math! ;-)
Funnily, Gerzon also built up another hierachy of sorts, incompatible
with UHJ. That's the frontal stereo one for early HDTV work. I never
understood why he didn't bring it under the ambisonic compatibility
fold.
DirAC has the problem that it currebtly can be used "just as research"
tool. (I won't discuss this here. But you know this as well as I do. )
Yes. It might be patent encumbered as well. And I don't much like the
idea that the decoder's input is effectively a directed cardioid -- that
should be dealt with pretty much as the dual of an optimal decoder.
But yeah...
Maybe. But before we "just" would need some good binaural decoder for
some standard surround formats; 5.1 and FOA surely included.
If you "just want to do it", I can easily contribute the research, math
and inner loops. On the framework and integration side I'm not too good.
Until 2morrow, and many thanks for your thoughtful and constructive
posting! (I have to think about this quite a bit more....)
What the list is for, but also a bit sorry for my usual rant-length. :)
--
Sampo Syreeni, aka decoy - de...@iki.fi, http://decoy.iki.fi/front
+358-40-3255353, 025E D175 ABE5 027C 9494 EEB0 E090 8BA9 0509 85C2
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe here, edit
account or options, view archives and so on.
