>Not sure if this would help, but the Carbon Event Manager is supposed to
>provide more reliable idle timing than the classic WaitNextEvent stuff.
>I'm not sure if the Carbon Event Manager (an event-based paradigm) would
>play well with Squeak (a polling paradigm?)
>
>-Eric

I have considered that, but each time I read the Carbon Event manual 
my eyes cross. However maybe I'll reconsider this week or next 
depending on my work load. Of course this becomes very carbon 
specific and I've been avoiding taking that leap.

One of the interesting things here is what to do when Squeak calls 
primitiveRelinquishProcessor which is to Relinquish the Processor for 
N microSeconds. Using 1/60 sec impacts morphic stepping. For Carbon I 
attempt a 1/60 wait every 100 calls which then lowers Squeak CPU 
usage from 100%. I guess I rummage about in the BSD toolkit.
-- 
--
===========================================================================
John M. McIntosh <johnmci at smalltalkconsulting.com> 1-800-477-2659
Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
===========================================================================