definition of etime(3F)

[Keith Bierman fpgroup]

>processor system a speedup by a factor of about 3.9.  Even on a dedicated
>system from that manufacturer a factor of 3 is the best that can be
>obtained.  The writers of those articles just take total time and

Superlinear speedup can and *does* occur. In my experience this is
usually due to favorable cache (or similar) memory effects. The entire
job, when run on a single processor exceeds the cache; but the N
chunks run totally in cache and do virtually no interprocessor
communication. 

Yes it is rare; but it is incorrect to assert that speedup is at
*best* N-1.