On 03/26/2016 05:22 PM, Albert Leusink wrote:
Jörn Nettingsmeier <nettings@...> writes:
i don't see why you would want to do that. the effect will be quite
strange... why would any part of the sound mix stay constant wrt head
position?
the effect would be a bit like rotating the music bed in the cinema
every time the camera pans - funny, but certainly irritating.
It would be for off-camera audio (voice over, music etc.) that don't have
any relation to the camera position/rotation.
To quote:
"What are you doing?", asked Minsky.
"I am training a randomly wired neural net to play Tic-tac-toe",
Sussman replied.
"Why is the net wired randomly?", asked Minsky.
"I do not want it to have any preconceptions of how to play",
Sussman said.
Minsky then shut his eyes.
"Why do you close your eyes?" Sussman asked his teacher.
"So that the room will be empty."
At that moment, Sussman was enlightened.
Imagine a video with an on-camera actor (dialog), a voice over and a music
track. You would want the on-camera dialog to match the video position (so
counter-rotate) , but the VO and music track will not rotate.
I still don't think that makes sense. Certainly it will be detrimental
to the proper externalisation of the VR content.
If you want to create two disjunct acoustic spaces, do it with _space_,
i.e. (lack of) reverb. A close-miked voice-over will be perceptually
clearly separated from any surround location recording at all times.
If you really want to freak out your users, you could route VO and music
to W only - the result will be a source inside your head. But my guess
is your brain will just flip you a birdie at that point and hitherto
refuse to externalise anything you throw at it, not just the voice-over.
if you absolutely have to do it, the only way is to deliver two streams,
one head-tracked and counter-rotated, the other not. which means you'd
have to have control over the listener's player software.
That's what I was afraid of...so I would need 6 channels instead of 4.
If and only if you find it actually benefits your production.
the only way to get two rotationally invariant signals into the stream
is a cardioid pointing up and another one pointing down. if your player
ignores head tilt, the result is like summing to mono and mixing into W.
if it supports head tilt, the result is likely even worse :-
Would that be the same as rotating the encoded stereo stream (set to 0º
spread) by 90º vertically?
That was not meant to be an actual solution, I was merely stating that
with this technique, you could insert two signals into your B-format
stream that would be rotationally invariant and could be extracted in
the player after the head tracking stage. But they would each spill into
your horizontal signal at -6dB, because there is only really room for
one extra signal... But if you can do DSP after headtracking, you could
also do it right and use two extra channels, iff it actually made sense
to do it from a perceptual point of view. So this whole paragraph is
hypo³thetical :-D
--
Jörn Nettingsmeier
Lortzingstr. 11, 45128 Essen, Tel. +49 177 7937487
Meister für Veranstaltungstechnik (Bühne/Studio)
Tonmeister VDT
http://stackingdwarves.net
_______________________________________________
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.
