FYI...
http://www.iosono-sound.com/game-audio/
http://www.iosono-sound.com/game-audio/proximity-technology/
* Bring sound effects into the inner circle of the surround field
* Allows for highly accurate sound localization
A standard PROXIMITY kit will consist of two intelligent, wireless
PROXIMITY loudspeakers and a PROXIMITY box,
* Connects to PC or console via standard HDMI
* Available from 8 December 2014
* Playback with PROXIMITY headphones or speakers
* Integration into the existing content creation tools
At the moment I can only try to guess how this (not yet available)
technology might work: Are they creating distance perception via ICC
cues? (see "Configuration and Installation" image. The added speakers
are - very obviously - side speakers...)
Further thoughts: If you would try to add distance cues to a
"conventional" binaural system, you would certainly have to implement
head-tracking, but also at least one set of "near-field HRTFs". (HRIRs
change with distance. Our measured HRTF libraries are far-field. Opinions?!)
I speculate that PROXIMITY headphones would code the distance cues in
the same way as Iosono's (PROXIMITY) speaker system does. (So < maybe >
via ICC cues, s. above.)
As Iosono wants to be seen as a competent company, I don't believe that
they will sell some snake oil. Especially since the market for Iosono
PROXIMITY are serious gamers and not HiFi enthusiasts... :-D
Best,
Stefan
P.S.:
Bo-Erik etc.: I start to believe that the complete "standard" HRTF
approach (binaural representation of surround sound via HRTF filters)
might need a major overhaul.
What would you need to deliver some (imaginary) perfect binaural system,
which is supposed to reproduce all we can hear? (The question is valid,
at least as a thought experiment.)
You already have to measure "enough" HRIRs over the complete (3D)
sphere, as usual. You would have to consider that possibly even small
head movements vs. the torso might or will change the IR(s). (One
additional degree of freedom) And if you want to include distance into
the system (see above: < perfect > the system shall be, and nothing
less...), you will add yet another degree of freedom. All the measured
responses you already have in your library - considering head-torso
changes - will fit to just (=only) one distance point. Accordingly, you
would have to measure the complete HRTF sets at several distance points.
(And even this "exploded" library might not be enough, I would argue.
Starts to look all very depressing...)
In the end, it really could be easier to < calculate > the HRTF sets
than to < measure > (huge) HRTF sets.
P.S. 2: I know that we have left standard binaural theories and/or
practices. But I recently have had several discussions which seem < all
> to indicate that our current HRTF models are far less complete or
perfect than most people would assume.
< Maybe this also indicates that we (currently) should apply BRIRs/BRTFs
in surround/3DA --> binaural decoders, NOT (anechoic) HRTFs. Because
BRIRs seem to deliver a more robust solution. Our "professional"
colleagues from Smyth Research, IRT (BRS system) and Mpeg have recently
all taken the BRIR route, an empirical fact which can't be ignored. >
So: Why do BRIR-based binaural systems seem to work better than
HRIR-based systems, even if HRIRs and not BRIRs should (theoretically)
deliver the best solution? (A part of the answer has probably been
given. Maybe our HRTF sets don't fit to real people, and people's
real-world movements...)
_______________________________________________
Sursound mailing list
[email protected]
https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe here, edit
account or options, view archives and so on.
