Hi Gabriel,
I would like to give some short commentaries inside your text, open for
further discussion...
Gabriel Wolf wrote:
Dear Sursounders,
the initial post of "A proposal for an Ambisonics based 3D audio codec
..." by Stefan Schreiber put all the things together so clearly and
nicely. MPEG-H 3D Audio might be the beginning of commercial
breakthrough in 3D audio. Ambisonics has been there since the 70's. We
all know its outstanding advantages. The public doesn't know.
Thanks for the warm words... Obviously we do agree.
....
Auro-3D and IOSONO don't have to be *replaced* -- Ambisonics has to be
*added*.
Nothing will be or should be < replaced >, but here a short reality check:
I am pretty sure that IOSONO won't be part of the MPEG standard, nor
does this have to be.
Iosono is already a defined and partially propietary technology. It
works since some time as a true 3D audio rendering technology. After
doing a little bit research I would say that IOSONO (specifications not
avaiable to me...) can't use "original" WFS over whole the upper
hemisphere, because otherwise they should need something like 90.000
loudspeakers. This doesn't seem to be the case, so what are they
actually doing for the "vertical" rendering part? I am eager to learn
something about, cos they don't tell this on their webpage... ;-)
And this is the difference between a 3D format < to be defined >, and
IOSONO. Iosono can be used to render 3D soundscapes, a "better 5.1
representation for every listener/spectator" in the cinema, etc.
But IOSONO doesn't seem to define a 3D audio format for general use.
(Correct me if I am wrong about...)
Auro-3D: My impression is that Auro-3D < will be included > as part of
the MPEG-H standard. (At least it seems they have applied to become part
of the standard.)
Ambisonic's problem was, that there had to be an expensive physical
decoder. This changed. Nowadays every 3D audio format needs to be
decoded. Even Dolby realized, that decoding arbitrary speaker feeds is
indispensable to next generation 3D audio.
There still has to be a physical decoder in the cinema, a surround
decoder (receiver) at home, or an AMB --> binaural decoder for headphone
use - this one maybe including head-tracking technology and HT feedback
to the decoder.
This is a chance we only have once: Ambisonics being included in a
world wide standard - for free!
But it will not happen magically by itself. This list has so much
manpower. If we are all pulling together, we can do it. It is possible.
I already try my best to make this happen, and certainly there are more
people.
IMO, Ambisonics is really in such a mature state that this really <
should > be included into MPEG-H. They will include some binaural
techniques, for sure.
(I have given some feedback to MPEG via another person and company
during last summer, stressing that surround sound might enter the mass
market if we/they find a convincing way in representing surround sound /
3D audio via headphones. Especially if we look into the real-world
habits of people, looking to the crazy amount of mobile devices around...
Secondly, I stated that Ambisonics is about the oldest form of 3D audio.
It is a speaker-independent technology, which is an advantage in an
increasingly heterogeneous environment of loudspeaker layouts.
Thirdly, Ambisonics might be the most efficient way to represent
surround sound/3D audio. Even if this is hard to prove in a scientific
way, there are definitively some facts which give my opinion some base.
You can't represent 2D surround in less than 3 channels, and 3D audio in
less than 4 channels. Ambisonics does this; I leave the strict proof of
"maximum efficiency" for Prof. Greene, who now has some serious work to
do... :-)
If you're feeling addressed >> get in touch with me off-list
(ecr...@gmail.com)
Please consider:
1. How much time are you willing to invest?
2. What is your expertise? What do you do best?
3. What is your motivation?
From your posting, it is actually not completely clear what you want to
achieve:
- To include Ambisonics/.AMB/HOA into MPEG-H?
The ISO/MPEG would not care for any of these questions. Safe bet.
- Or to start some "open project", which would probably have to include
some reference to open hardware?
This is a possibility, and I have thought about this.
A place where open standardization could be done would be here:
http://xiph.org/
Note that Vorbis and Flac are audio codecs, Opus is actually an official
Internet audio codec. (The other one is ITU G.711, the old PCM
"telephone codec".)
It is possibly best to see what will happen in Genève, at least for my
part...
Best regards,
Stefan
P.S.: The IETF itself might have some interest in surround sound/3D
audio transmission.
Better Internet radio?! Maybe not yet, but stay tuned... O:-)
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
