It is not very nice to have to comment on your own stuff. Instead, I
will present a paper of the BBC on the perspectives of Ambisonics for
broadcast use. (In my former posting I gave some strong recommendation
to include stereo backward-compatible forms of FOA < and > HOA into the
proposed "work in progress" standard.)
"Upping the Auntie, A broadcaster's take on Ambisonics"
http://downloads.bbc.co.uk/rd/pubs/whp/whp-pdf-files/WHP221.pdf
(This was not a "source" for the issued standard/format(s) proposal.)
Some observations from the BBC article:
If a B-Format [2]
(or high order Ambisonics) representation of the sound field were used
in place of the mono, stereo
and 5.1, this might provide compatibility with current formats and
could even help future proof
the archive.
Finally, an obvious advantage of Ambisonics is its ability to convey
height information. As will
be discussed later, this may offer an improved listening experience.
However it can also act as a
useful marketing tool (sic...), as it can be considered to be a ‘3D’
audio format.
(4.1)
There is no doubt that the ability to include height is
one of the best known examples of Ambisonics, and it gets people
interested and excited about
the technology. However, the effect that height information has on the
listening experience is not
currently well-understood. To assess the value of 3D audio, the
effects of the representation of
height need to be investigated.
(Commentary: Yes, but actually which form of height? "Some" height,
semi-spherical, full-spherical height? A test has to consider this.
There is not just one form height, consequently. Most current
loudspeaker configurations with height are semi-spherical, possibly
including < some > negative elevation but in very modest form.)
(4.2.4)
Figure 11 shows that the subjects rated stereo as inferior to both 5.0 and
Ambisonics when it came to music.
(Commentary: Even 192/24 stereo bitrates won't change this picture, to
answer some recent posting revealing hidden plans of some CEA-linked
group to re-edcuate fast food consumers to become wine
conaisseus/conoiseurs. No wonder, cos many audiophiles went back to the
booming vinyl market, in absence of other choices... :-D )
(4.2.5)
For the two drama
examples more than half of the subjects tested said they struggled to
grade the Ambisonics because
they preferred the Ambisonic representation of the sound effects but
not the dialogue. Comments
such as “the sound effects are really good but vocal is not so good,
she sounds muted, the sound
effects are lovely” were typical.
Interesting is obviously the chapter 5 ("barriers") section.
Whilst there are channel-ordering
conventions for B-format and higher orders, the increasing dominance
of file-based production tech-
niques means that standardised file formats and metadata are essential
for effective broadcasting
and archiving.
The (current) mandatory part of the format proposal combines FOA
(including the established UHJ/LRTQ stereo compatible form), and mixed
order 3h1p HOA. The application of the latter has been "field-tested" by
several games. (Successfully)
(
Interest in Ambisonics outside the academic and enthusiast communities
appears to have risen in
recent years. This is most notable in computer games publishing, where
gamers have been early
adopters of surround sound, and the nature of computer games means
that complex DSP can be
used for audio. There remains a number of barriers to the adoption of
Ambisonics in broadcast
production workflows, and the case studies presented in section 3
identify a number of them.
)
Whereas problems with the production workflow is a professional
insider-problem (and seems to be fixable, at least if you < want > to
fix things), it is more important to think from the customer's point of
view.
(Chapter 7)
There could also be an investigation of hybrid methods of delivery
which could use
some combination of Ambisonics and discrete channel systems.
I have included this. The proposed combination is stereo - Ambisonics
extensions, not Ambisonics hidden in 5.1.
An investigation into how the lossy
audio compression technologies employed in the broadcast chain would
affect Ambisonics signals
would also be beneficial.
Has been studied! But I will leave it to others to provide this link.
(I intended to participate in this discussion in a constructive way... :-) )
Best,
Stefan Schreiber
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
