[Csdp] Puzzling result using icc

[Brian Borchers]

>That said, the p4 single-threaded version is still faster than the icc
>version.  Until you up the dimension: on a similar problem, but with
>about 1600 constrains, the icc version took 5 sec, the p4 took 20 sec.
>I have two dual-core processors, so that's exactly right.  (In fact,
>suspiciously right.)

It's quite common to see little or no parallel speedup on small problems, 
because there is overhead associated with parallelizing the code that 
can't be amortized over the short run time for a small problem.  

In addition to the speedup from four cores, you're probably getting a 
speedup from having the code optimized to use features of the newer
processor.