Re: Population count in SSE2, again

I DO understand that if you have millions of very long bits strings
that you have to count bits in or match mbits in (as in DNA or RNA
matching) you might want a very tight efficient loop algorithm.

But then again, from a practical point of view, you still have to
consider the access to the huge strings of bits millions of times. Ar
we nick-picking in the wrong place?

Are we not seeing the woods for the trees?.