[vorbis-dev] True surround sound for Ogg -- a proposal

[David Carter]

On Fri, Jul 27, 2001 at 07:53:23PM -0700, Michael Paine wrote:
> Has this sblive/ambisonic idea gone anywhere?  I would love if it did..

Not that I've heard of.  The obvious prerequisite would be for the ALSA sound
drivers to include support for four-channel output.  I don't use ALSA myself
(I still use the vanilla SBLive driver in the kernel), so it's very possible
that this has been added.  I suppose I should try ALSA sometime and see how
it is...  I'm assuming here that the DACs for the rear channels are driven by
the same clock as the front channels, as that is the most cost effective
solution (you only need one oscillator), so we're most likely fine there.
We just need driver support.  If only there was a cheap card with synched
4-channel inputs that had linux drivers -- however, the demand is such that
there probably isn't.  Also, existing recording software (I prefer gramofile,
which has great buffering which eliminates the need for a process priority
<0) would need to be modified to capture 4 channels instead of two.  (Some
app may already exist for this, but I haven't had the need, so I haven't
looked.)

It would be a good idea to determine how many of these things already exist,
and how many would need to be found/created/modified.  As for the sound HW,
users will probably just have to bite the bullet and buy one of the existing
cards.  The software side is a lot more open.

BTW, I have an Ambisonic test disc that's encoded in DTS format, but at this
point I don't have a way to get it in uncompressed format.  (If I had a
sound card with four channel input, I could digitize the output of my
surround receiver's preamp outputs, but I don't.)  It's quite impressive,
considering the B-format source was decoded in software to a 4-speaker
square.  I can close my eyes and point to where the musicians are... :)
(The original source was a sound field mic, recorded in B-format.)  I may
be able to track down the guy that recorded the disc and see if he would be
willing to send me uncompressed source files.  Those would be much easier to
work with.

On the software side, surround support has not been added yet, beyond the
generic multichannel encoding that oggenc supports (if you feed it a 4-channel
WAV, it'll make a 4-channel ogg file, but AFAIK there's no support for mapping
the channels to Ambisonic signals).  Someone should still e-mail Richard
Furse about possibly relicensing his software Ambisonic decoder under GPL-
compatible terms (or looser, if he was amenable) so we wouldn't have to write
one from scratch.  I think he also has software to mix discrete channels into
an Ambisonic stream, positioning them arbitrarily.  That would also be useful.
His software page is at http://www.muse.demon.co.uk/mn_index.html.

I believe that there is some sort of field defined in the stream format for
surround metadata (e.g. channel info, etc.), but someone else will have to
answer that for sure.

Do you have any skills, equipment or anything else that would be helpful in
adding any of this?  I'd love for it to get off the ground again.  I'm CCing
the list -- perhaps this will spark some discussion again. :)

        David


-- 
David Carter ** dcarter at sigfs.org ** dcarter at visi.com
PGP Key 581CBE61: E07EE199C767C752 8A8B1A9F015BF2EA
Key available by finger or www.keyserver.net

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-dev-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.