On Thu, Apr 23, 2009 at 10:21 AM, Andrew Pascoe <aylutar@...> wrote:
Block sizes should be powers of two.
El 23/04/2009, a las 11:27, Scott Worthington escribió:
If I run your code and run s.options.dump, the blockSize is as you've set it. It seems that if you set SCs block size is larger than the hardware's there is no sound.
Yes, powers of two less than or equal the hardwareBufferSize (thanks James!)
I don't find this restrictions in the documentation.
I might go out on a whim and ask if you're trying to set the block size like a k-rate in Csound? I only ask because of the new block size is 1/10 of a typical sampling rate.
Yes, I'm doing a language/environment comparison and have Csound code with that kr (the idea is to achieve the same result signal, or so).
El 23/04/2009, a las 11:44, James Harkins escribió:
What would be the benefit of a 0.1sec control rate? That's going to be
really bad for kr UGen shapes (which will turn into 0.1sec line
segments instead of nice curves), timing resolution on the server,
probably more stuff I can't think of now.
Perhaps only a methodological difference betwen both, or the resources are different.
What do you think?
Thanks you very much.
