Why is this so much faster?

[Ian Kelly]

On Thu, Jun 2, 2011 at 4:38 PM, Tim Delaney <timothy.c.delaney at gmail.com> wrote:
> First of all, do you have the parameters to the lambda the wrong way around
> in the map() version? zip(histogram, range(255)) will return (histogram
> value, index), but enumerate(histogram) will return (index, histogram
> value). But the parameters haven't been swapped, resulting in the histogram
> value being squared in the map version, but the index being squared in the
> manual summing version. Depending on the values, this could result in a
> large performance increase in the second version (if the total value exceeds
> the maximum size of your platform's "long" type).

In addition to what Tim said, I question whether you should even be
multiplying by the index at all.  The usual definition of "root mean
square" is this:

math.sqrt(sum(x * x for x in values) / len(values))

I don't know the problem that you're trying to solve, though, so maybe
I am just being confused by your terminology.

Cheers,
Ian