>> VDP and DBAP is based on the same idea, but DBAP as presented in the ICMC 
>> 2009 paper
>> 
>> http://www.trondlossius.no/system/fileattachments/30/original/icmc2009-dbap.pdf
>> 
>> has a number of additional features. The important difference of DBAP and 
>> ViMiC as compared to ambisonics and VBAP is that there are no restrictions 
>> on the positioning of loudspeakers or listener. Loudspeakers are not 
>> restricted to a ring/sphere surrounding the listener, but could e.g. be laid 
>> out as a regular or irregular grid in the space. This is what makes it 
>> useful for installations in one or more spaces, such as art galleries and 
>> museums, where rings and spheres of speakers might be impractical and the 
>> audience is free and expected to move about.
> 
> I really don't understand  "...ambisonics and VBAP is that there are no 
> restrictions on the positioning of loudspeakers or listener.", at least as 
> regards to VBAP. We are currently - and have been for over two years now - 
> using VBAP in The Morning Line sculpture 
> (http://www.worldarchitecturenews.com/index.php?fuseaction=wanappln.projectview&upload_id=14039)
>  and the reason we used it rather than Ambisonics was because it could cope 
> with extremely irregular arrays of speakers. Comments?


Thanks for the link, that seems to be a very interesting project!

It's difficult to figure out the loudspeaker setup in this sonic pavilion from 
the image, but reading the AES paper:

http://www.aes.org/e-lib/browse.cfm?elib=15370

the 41 speakers seems to be subdivided into 6 "rooms". Each of these rooms seem 
to have the speakers pointing towards the sweet spot of the room (appendix A). 
Did these rooms coincide architectonically with the positions that audience 
were likely to be standing at in order to experience the work?

As the azimuth and elevation of each speaker relative to the sweet spot of the 
room seems to be irregular, I think it makes a lot of sense to use VBAP rather 
than ambisonics.

If you want a source to move all around the sculpture, how do make it move from 
one room to the next? Do you cross-fade from one sphere onto the next as it 
moves? Or was "room 7", the one used for spatialization into the structure as a 
whole using a subset of the speakers at the margins of the sculpture to create 
yet another sphere?

With DBAP you wouldn't have to subdivide the array in order to enforce a 
"sphere surrounding sweet spot" way of thinking on to the sculpture. You could 
just describe the position of all of the speakers, and then freely move the 
sources around the sculpture. The theory of DBAP ideally assumes loudspeakers 
to be radiating the same way in all direction, and I personally know very few 
speakers that are close to doing so, so the sounding results would be 
influenced by the direction of the loudspeakers. This would need to be 
accounted for in the compositional process, but could spatially and 
sculpturally be used in meaningful ways, leave impressions of sound sources 
being directed towards or away from you depending on your and the speakers 
positions.

DBAP was in fact first developed for an installation with a loudspeaker setup 
somewhat similar to the one of The Morning Line, although being 2D instead of 
3D. I was working on a sound installation for Galleri KiT, the gallery space of 
the art academy in Trondheim, Norway. This gallery consists of a number of 
semi-connected rooms. I wanted to have a total of 16 loudspeakers distributed 
along the walls of the various rooms, and then have sound sources drifting 
around the space, from one room to the next.

I first tried setting up a set of VBAP rooms, with a mechanism for moving from 
one room/VBAP system to the next as sources drifted. However I did not manage 
to find a way of doing so that felt smooth and convincing. Sound seemed to be 
glued to the walls of the rooms, sneaking from the wall of one room to the next 
in a way that I felt to be abrupt and unconvincing for what I was after.

DBAP was developed late the night before the opening, and helped creating a 
much smoother drift. As speakers were positioned along the walls, the direction 
of the speakers was not experienced as problematic, it rather added to the 
feeling that the sound had now moved on to the next room.

I hope this helps clarifying the difference.

Another way of explaining would be the following:

Imagine that you have a regular 3D matrix/grid of 4x4x4 speakers. The grid 
could be subdivided into 3x3x3 rooms, and ambisonic decoding for a cube of 8 
speakers used for each of the rooms. You would then need to find a way of 
switching or crossfading from one cube to the next as the source moves about. 
Alternatively, using DBAP, you would deal with all 64 loudspeakers in one go. 
DBAP would simply feed the highest amount of gain to the speakers closest to 
the virtual source, and next to none to the ones that are far away, all while 
keeping the total intensity of all the loudspeakers at a consistent level.


Best,
Trond
_______________________________________________
Sursound mailing list
[email protected]
https://mail.music.vt.edu/mailman/listinfo/sursound

Reply via email to