Re: Function to find the greatest or smallest value

I may stand corrected. Or maybe not. Needs more investigation, because I
know that doing a test on each pass through the loop is costly. The sort can
be optimized.

If the array needs to remain in the original order, I agree the for loop
would be better. But we were talking about a max function that would either
compare all parms looking for the max or load and load and sort an array
with the parms. If the sort is not desirable, doing a loop for n = 1 to %PARMS
would do just fine. No array, no SORT.


In a message dated 4/24/2009 3:27:37 P.M. Jerusalem Daylight Time,
adamglauser@xxxxxxxxx writes:

CRPence wrote:

Adam Glauser wrote:

Personally I prefer Asher's [sort] solution, which is both
more elegant and more efficient. To each their own.

Just curious. Why would a sort be more efficient than spinning the
array in the /for loop/ as above? Does the sort not require effectively

all the same work?

Erm, I was wrong. I was thinking of quicksort, which is O(n log n).
Somehow my memory had turned that into O(log n), which as you pointed
out is not possible. I think sort, by definition, can not be better
than O(n). The max loop is clearly better.

Your point later about maintaining the order of the original arrays is
also well taken.

Thanks for the corrections.