[Vorbis-dev] 5.1 surround channel coupling

[Richard Lee]

With some homework, I now add pseudo Vorbis guru to my Ambisonic & DSP pseudo guru hats.

Looking at old Vorbis posts ...

>What's missing is tuning of the reference vorbis encoder to produce better quality/bitrate for 5.1 surround mixes by taking advantage of redundancy between the channels. Without that we're not technically competitive with AC3.

Yes.  Good multichannel "lossless coupling" is essential.  It is this which gives DD a 2:1 advantage over DTS and allows MLP to meet the DVD-A bitrate spec.

>Monty has also cited the lack of (not lossily compressed) 5.1 sources to test with as an obstacle here.

There is a large library of impressive uncompressed high quality Ambisonic recordings at 

www.ambisonicbootlegs.net

>However, most of the ambisonic literature is based on the the original 
quadrophonic implementations (where were also patented) and this isn't 
enough channels to accurately represent the planar source arrangment 
most content is mixed for.

No.  Ambisonics is definely NOT based on quad.  That's why it works (even with 4.0) when quad doesn't.  In fact Ambi 4.0 works better than 5.1

By separating the definition of the soundfield (B format where 3 channels is sufficient for horizontal and 4 for a sphere) and the speaker feeds (depends on your wife), Ambisonics optimises both.

The Speaker Decoder is based on where your speakers are and on ALL existing theories of Auditory Localisation (except for Pinnae & HF ITD models).

And they sound good too.  This was an important objective.

Higher Order Ambisonics allows even more pinpoint localisation but as Sebastian Olter has pointed out, simple 3 or 4 channel Ambisonics is already a lot better than the present naive 5.1 one speaker one channel systems.

But the really useful trick is that we can translate 5.1 to 3 channel B-format and if played back on eg an Ambi 4.0 system, might even sound better than the original.

>But Ambisonics already works =] I have some 130-140 kb/s movie sound
tracks "recorded" ambisonically and encoded with q=0 and they sound
quite good (much better than faac 5.1 at 250-300 kb/s), sometimes it's
hard to distinguish between them and the original soundtrack. The
"recording" takes place in time domain so it can be done even in oggenc.



-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.441 / Virus Database: 268.18.4/705 - Release Date: 27/02/07 15:24