Hi Michael, Ed, and all! The fader is already in the works for the new release. Currently ChucK is mono, for all intents and purposes, though it would be easy to "turn on" stereo/multichannel, since the underlying system is already built to support that. However, there aren't quite enough language facilities in 1.1 to support channels flexibly (i.e. ugens as members of ugens to have a clear notion of inlets and outlet, which requires the new type systems we are implementing). As for HRTF: there is a generalized Filter ugen but we would need arrays to sensibly create high order FIR filters suitable for convolving HRIR (head related impulse response). Arrays will be avaible in 1.2. (The above time-domain FIR filtering can be quite (to very) slow, depending on the length of the impulse response. As mentioned earlier, BruteFIR - a FFT-based convolver may be an eventual and much faster alternative. One tradeoff is output delay proportional to the buffer size for the FFT, which for many things is not a problem.) Also, I will send a brief progress report on chuck-1.2 soon. Best, Ge! On Feb 17, 2005, at 10:09 PM, M. Edward (Ed) Borasky wrote:
Michael Heuer wrote:
A feature request, might it be possible to add a simple fader (l/r channel) to ChucK? Or what would be really useful would be a 3D spatialization generator, something like
but I haven't been able to locate an open source implementation yet.
Is there a well-defined DSP algorithm for this? I know CSound, for example, has code for the "head response transfer function". This computes what you need to put in the left headphone and the right headphone to simulate localization. If that's what you're looking for, I think it's "just a convolution" :). Can Chuck do convolutions?
_______________________________________________ chuck mailing list chuck@lists.cs.princeton.edu https://lists.cs.princeton.edu/mailman/listinfo/chuck