Michael Chapman wrote:
Sebastià V. Amengual wrote:
That was one of the main cons. With the four distances that I used in my
measurements (10.44mm, 23.49mm, 48.34mm, 98.35mm)
if we wanted to cover the whole frequency range, we needed at least 3
microphones with different spacing, and then the
signals could be combined depending on the frequency. But as Fons
pointed out, a gain mismatch has quite a big influence
on the directivity, and self-noise is also increased. In our case we
used GRAS matched mic's, so this problem was mainly
solved, but I assume with cheap MEMs microphones the game changes...
Ignoring, for the moment, commerce : how good do the '360 degree' video
people want the audio to be?
Obviously 'good enough', but what is that?
To my mind you cannot make anything approaching accurate full-sphere video.
To make full circle horizontal images would require a specialised camera*.
Yes, you can 'glue together' a number of individual images, but the
perspective is actually all wrong.
A true panorama has _no_ perspective, as there is no unique viewing position.
(One can make true panoramas 'at home' with a video camera (and software).)
With VR one wants to be able to create views from any viewing position.
There is not enough data in a panorama to create perspective from any
position, let alone from all.
The perspective of a true panorama (i.e. VR) camera seems to be "you"
(watcher) at the center, focus at infinity. There will be some
positional errors later (which will depend on the video stitching
software too), but you can't say there is no perspective at all.
(Otherwise a video camera would also have no perspective.)
(The < unique viewing position > of a panorama camera is decided at
viewing time.)
Different focus and "perfection" would require some LF (light field)
camera. Is this your point?
So it is all a case of what is 'good enough'.
We made two videos:
one
-a simple360 degree pan,
the other
-scrolling along a true panorama (with no horizontal perspective) from the
same place.
These were then mounted side-by-side as a wide screen video to demonstrate
the differences.
The result was disappointing. There are differences, but they do not 'jump
out -at (at least) me!
The Crossrail video I referred to earlier apparently used six individual
cameras to create a full sphere video.
That would seem a bit clunky, but reports are that the result is impressive.
So is our 'perspective' on audio vastly different from that of our
audiovisual colleagues?
Yes, because natural hearing happens in "360º" - with certain
directional preferences and limitations.
But we actually agree that there are differences of < perspective > in
video and audio...
Best,
Stefan
('Gaming' is I accept a different matter.)
Is audio as clunky as their video 'good enough'?
Michael
*I seem to recall in the Nineteenth Century there was a tower on the edge
of Clifton Downs in Bristol that offered a 360 degree panorama projected
on the wall of a chamberin the tower ... bit that is a very old memory.
For an artist's view there are some interesting insights in Martin Gayford
(2011) "A Bigger Message. Conversations with Daid Hockney".
(At page 58 "I have said that perhaps the big mistakes of the West were
the introduction of the external vanishing point and the internal
combustion engine". Quoting other material would be a bit meaningless
without reproduction of the artworks that are reproduced in the book.)
_______________________________________________
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.
_______________________________________________
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.
