Sashikanth Chandrasekaran wrote:   
> (# ; (3!:0)) p1
> +----+--+
> |1393|32|
> +----+--+

32 means you're working with boxed elements.

(As an aside, (# ; 3!:0) should work fine.)

> Assume that I do not know the distributions of p1, p2
> in advance. Is there a way I can get consistent avg.
> performance (i.e. 6!:2 'p1 V p2' = approx. 6 seconds
> and 6!:2 'p2 V p1' = approx. 6 seconds).

Perhaps try the J 6 beta:

http://www.jsoftware.com/books/help/release/idot601.htm

This only seems to apply to i. (not e.) but maybe
that's sufficient for your case?

-- 
Raul

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to