Group: comp.lang.forth · Group Profile
Author: Albert van der HorstAlbert van der Horst Date: Sep 3, 2008 12:10
> Both sorts begin to have higher cost in time as the width of the sort
>becomes
>larger. For the distribution sort, the radix 256 is still the most
>prcatical
>although a radix of the square root of the width of the data is
>fastest.
>It would still sort things like hundreds of thousands of 64 bit values
>much faster than something like quicksort.
> In any case, the speed of these two sorts for things like tables of
>values
>is so much faster than things like quick sort, it makes an investment
>in the
>time and memory space practical.
> I once used an optimize distribution sort for a mail list of about
>500 names.
>It ran several hundred times faster than quicksort of the same list.
As an algorithm quicksort leaves not much room for improvement. So I
tend to doubt this. Anyway, I would like to see a benchmark to back
this up. (For sure, a general purpose sort has to compromise, and a
factor up to ten I would estimate plausible. Then, it would be
acceptable in a lot of situations.)
>I'd places
>optimizations in it to deal with empty fields in the names. I don't
>recall
>exactly how it worked but the fellow that was using that mail list was
>happy.
Right!
>Dwight
Groetjes Albert
| 
