On 14 October 2012 01:34, Sampo Syreeni <de...@iki.fi> wrote:
> On 2012-10-05, Richard Furse wrote:
>
>
> Tell me... In games most of the individual sound sources, apart from general
> ambience, seem to be well placed monophonic ones, fed from a single channel.
> So in essence, they are "encoded" at infinite order. Has anybody done any
> work on how to overlay such sources optimally against a lower order, perhaps
> recorded, background?
This is the (old) problem of properly representing (and reproducing)
the spatial extent of sound sources. If the source is actually rather
small - say, a small bird or, as we are in games mode, the snick of a
safety catch - or so distant as to be effectively small, there is no
problem in simply applying the standard encoding equations for the
order in use. For large or close sources, the problem is significantly
different. In games audio, the sound of, something with a significant
angular extent - say a vehicle - is often represented with a
collection of several monophonic sound files, each making up part of
the overall sounding object (in this case, these would probably
include at least the exhaust, engine, and tyre noise). which can all
be panned separately into the sound image, as dictated by the
geometrical relationship between the player and the vehicle. This
geometrically determined separate panning of parts of the sound image
may even extend to early reflections, if enough computing power is
left over from the graphics.
>
> There has been some talk about mixed order playback in the past, and it's
> always ended up with somebody saying that different orders don't really mesh
> too well.
They don't when you are mixing together B formats of different orders,
but this is not really what you would be doing here, as the order of
the directional sampling of individual sources is set by the encoding
used to pan the source into the soundfield, not the fact that the
monaural is effectively infinite order if played over a single
speaker.
> So, how well *can* they mesh, given that the stuff games put out
> are grossly higher sampled in direction than any realistic playback rig? Any
> ideas of how to efficiently spatially sample them back to the rig geometry,
> and regularize the decoding problem?
All I can say is long experience (albeit it in the context of
electroacoustic music, rather tha games) has shown that the
"suspension of disbelief" effect is very powerful and can quite easily
enable a listener to accept a mono sound playing back from a single
speaker provided that it is a believable sound, in context.
All the best
Dave
--
As of 1st October 2012, I have retired from the University, so this
disclaimer is redundant....
These are my own views and may or may not be shared by my employer
Dave Malham
Ex-Music Research Centre
Department of Music
The University of York
Heslington
York YO10 5DD
UK
'Ambisonics - Component Imaging for Audio'
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
