going by some tests I did on my quad, it merely spreads out the process over the extra cores but will still only utilise the max process a single core can do.
for example on a single running at max, it will take up 100% of that core, but only 25% overall on a quad (or 16.6% on a hex).
if it's spread over all the cores it will still only utilise 25% (16.6% hex) overall but only use between 1-50% on any 1 of the cores at a time.
so the cores could be running at 4%, 15%, 3%, 3% for a total of 25%.
so as far as I see there is no benefit for the option.