In message <20060118021133.GB12442@pell.home.ka8zrt.com>, Douglas Wade Needham 
writes:
>This machine will see no idle due to a program which runs niced to 20
>and will eat any computation cycles with almost no idle.  This
>program, which is similar to setiathome (but processing data for a
>different radio telescope), was also running before, with no changes
>in configuration, resource usage per data unit, etc.  And both of the
>kernels had APM enabled as well.  Unfortunately, I have not
>instrumented it to the point where I can tell whether it has slowed
>down in processing...that is one of the things I had been meaning to
>do, but had not yet had a chance to complete.
>

Another issue: is it the total system that's slower or the compiler?

		--Steven M. Bellovin, http://www.cs.columbia.edu/~smb