[Csnd] update on RT 5.1 surround?
| Date | 2014-04-18 17:15 |
| From | David Worrall |
| Subject | [Csnd] update on RT 5.1 surround? |
| Hi All, I need to render csound output in real-time to a external 5.1 device (Creative Sound Blaster X-Fi Surround 5.1 Pro USB Sound Card:) I've never used 5.1 before so I was wondering what other sounder's experience is. For example, is there any update on I'd really appreciate hearing of other's experience. David ______________________________________ Prof. Dr. David Worrall Emerging Audio Research (EAR) Audio Department International Audio Laboratories Erlangen Fraunhofer-Institut für Integrierte Schaltungen IIS Am Wolfsmantel 33 91058 Erlangen E-Mail: david.worrall@iis.fraunhofer.de Internet: www.iis.fraunhofer.de --- Adjunct Senior Research Fellow School of Music, Australian National University |
| Date | 2014-04-18 18:03 |
| From | Michael Gogins |
| Subject | Re: [Csnd] update on RT 5.1 surround? |
I am not speaking from extensive experience here, because I am just getting started with spatialization myself. But I think you have two options, vector-based audio panning (VBAP) and Ambisonics. They each have their own advantages and disadvantages. The most impressive demonstration of spatialization that I have heard that used Csound was Jan Jacob Hoffman's piece at the Linux Audio Conference in Hannover. He has created his own rather extensive system that builds upon and extends Ambisonics. It can be used in Steven Yi's composition environment blue. To use VBAP, you would simply figure out where the 5.1 speakers are in your room, then directly pan your sounds using these positions. To use Ambisonics, you would encode your sounds with panning to Ambisonics B-format, then you would decode the Ambisoncs signal to something approximating your speaker rig or, if you are moderately sophisticated mathematically, you can actually pan to your measured 5.1 rig.
If you don't want to do your own decoding math, you can use the 5 speaker Ambisoncs decoder with a 5.1 speaker rig, but then I think you need to let the 5.1 center and subwoofer speakers remain silent.
Generally, spatialization considers mainly angle and elevation, not distance. Hofmann's code can be borrowed to implement distance cues: Doppler effects, attenuation of loudness and high frequencies by distance, and other more sophisticated cues.
H ----------------------------------------------------- Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Fri, Apr 18, 2014 at 12:15 PM, David Worrall <david.worrall@iis.fraunhofer.de> wrote:
|
| Date | 2014-04-18 19:02 |
| From | David Worrall |
| Subject | Re: [Csnd] update on RT 5.1 surround? |
| Hi Michael, Thanks for your extensive reply. I appreciate it. I don't have a problem w. the math - or the psychoacoustics; I have previous experience with 20.4 ambisonics - but in those situations we had our own (hand built hardware) decoder that was controlled independently of the synthesis engines. What I am ignorant of how to address the 5.1 USB card from csound i.e. what signals to put on each channel. If I understand you correctly, you seem to be saying that I could just send it resultants from one of the vbap opcodes ( i.e. the card doesn't do anything special to the inputs - just passes them on to the individual outputs (+ perhaps the 5.1 card does that lowpass filtering of all channels to the sub. automatically…) Or I could encode the 5 channels using the ambisonic opcode … If this is the case, perhaps I could use two 5.1 boxes (each on a separate USD port on the computer, and have a 10.2 system. Or is that floored logic? D. On 18.04.2014, at 19:03, Michael Gogins <michael.gogins@gmail.com> wrote:
______________________________________ Prof. Dr. David Worrall Emerging Audio Research (EAR) Audio Department International Audio Laboratories Erlangen Fraunhofer-Institut für Integrierte Schaltungen IIS Am Wolfsmantel 33 91058 Erlangen Telefon +49 (0) 91 31 / 7 76-62 44 Fax +49 (0) 91 31 / 7 76-20 99 E-Mail: david.worrall@iis.fraunhofer.de Internet: www.iis.fraunhofer.de --- Adjunct Senior Research Fellow School of Music, Australian National University |
| Date | 2014-04-18 19:13 |
| From | Michael Gogins |
| Subject | Re: [Csnd] update on RT 5.1 surround? |
The Ambisecond opcodes have a limited set of out of the box speaker rigs:
1. Stereo - L(90), R(-90); this is an M+S style stereo decode. 2. Quad - FL(45), BL(135), BR(-135), FR(-45). This is a first-order `in-phase' decode.
3. 5.0 - L(30),R(-30),C(0),BL(110),BR(-110). Note that many people do not actually use the angles above for their speaker arrays and a good decode for DVD etc can be achieved using the Quad configuration to feed L, R, BL and BR (leaving C silent).
4. Octagon - FFL(22.5),FLL(67.5),BLL(112.5),BBL(157.5),BBR(-157.5),BRR(-112.5),FRR(-67.5),FFR(-22.5). This is a first-, second- or third-order `in-phase' decode, depending on the number of input channels.
5. Cube - FLD(45,-35.26),FLU(45,35.26),BLD(135,-35.26),BLU(135,35.26),BRD(-135,-35.26),BRU(-135,35.26),FRD(-45,-35.26),FRU(-45,35.26). This is a first-order `in-phase' decode.
I was wrong before, you would use quad to send to the side speakers of 5.1. I think for 5.1 output (not talking about spatialization here, just output, and I haven't tried this either) you should set nchnls to 6 and send zeros to the 5.1 center channels, and quad to the 5.1 side channels. Or, you can write soundfiles and load them into a DAW and use the DAW to send to the speakers or do additional panning.
Hope this helps, Mike -----------------------------------------------------
Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Fri, Apr 18, 2014 at 2:02 PM, David Worrall <david.worrall@iis.fraunhofer.de> wrote:
|