On Thu, 20 Apr 2006, chris clepper wrote:
On 4/20/06, Hans-Christoph Steiner hans@eds.org wrote:
The preference is a bit odd. The option is "defeat real-time scheduling", so for basic audio work, leave it off.
It's not even 'real-time' because no time slicing is negotiated with the kernel. The code just jacks up the pthread priority for Pd. It's a great way to lock yourself out of the GUI when using GEM.
It shows even more with GridFlow, because the example patches have their [metro]s too fast for what the cpu can do, especially because of the OSX-specific problem that GridFlow has.
Anyone with a MacIntel can verify whether the slowdown is due to OSX or to PPC ?
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada