12.1 & Audiences

[Martin Naef]

Michael Zacherl wrote:
>>> sweet!  I'd like to listen to this!!  :-)
>> Come to Glasgow then...
> is there a deadline?  ;-)

You'll never know - though it's unlikely that I move away too quickly. 
If you want to plan a quick detour around Synth-DIY, that'd be safe. ;-)

> how big is this place where you're listening?

Not big - imagine a good-sized living room. I can't remember the 
dimensions for sure, but I think it something like 8x4.5m max, although 
the room is split in half by a back-projection screen. We lost another 
.5m in each dimension due to the acoustic treatment on the walls. 
There's really only space for two people to listen to the sound properly.

>>> The 12 speakers, how are they set up/situated?
>> Imagine 3 levels: Ground, ear height, and ceiling. Each with 4 
>> speakers. Ground and ceiling have front, left, right, back, whereas
>> the ear- level is rotated 45 degree, giving front-left,
>> front-right, back-left, back-right. All speakers are set up at an
>> equal distance to the sweet-spot, although the playback system does
>> compensate for some differences. Then there is the sub.
> 
> what speakers are used and what's the rms output of the
> speakers/amps?

Yamaha MSP 5. Can't remember the wattage, but with 12 of them, you can 
make it quite uncomfortably loud without pushing the speakers too far... 
The sub is a Tannoy Sub 15. The system is calibrated so we can play back 
recordings at exactly the same level they were recorded - and we have 
recordings from train stations and very busy underpasses...

>> I recorded the concert on an Edirol R4-Pro (four channels).
> 
> so you needed 3 of them synchronised via SMPTE? (I had to look up
> what that is)

Ahh, no, just one - I think I need to explain what Ambisonic is (also: 
google for it). A first order Ambisonic recording is just four channels 
WXYZ, which represents the *soundfield* at a given position: W is the 
pressure (as if you used an omni mic), and XYZ is the 3D pressure 
gradient - imagine 3 figure-of-eight mics. There are special microphones 
(google: Soundfield - we use the ST-350) that output exactly those 4 
channels.

BTW: If you understand M-S stereo, you also understand Ambisonic - it's 
just two additional channels to represent the other dimensions too.

But how do you play back such a recording? You just use a bunch of 
speakers to somehow recreate the original sound field at the sweet spot. 
The more speakers you have, the better you can represent the field - 
hence our 12 speaker system.

But you can also just use the information to "simulate" and derive a 
range of coincident microphone placements: Stereo, or any kind of 
surround. So it's quite handy for multi-purpose recordings: You just 
record the sound field, and then later on you can choose the target medium.

>> We play it back through Max/MSP, so there's a PC driving the
>> playback - so the recording sits on a harddisk.
> 
> what interface(s) do you use for playback? what about the sub
> channel? Hows that extracted?

The interface is an RME one with a 16 channel ADAT-to-analog converter. 
Nothing fancy. The sub receives the W channel (low-pass filtered).

Bye
Martin