Oleg wrote:
>  I thought it was shown that old style shift
>  is more efficient than infix. 

Nope, other way around.

Scalar functions like  - :

   http://www.jsoftware.com/help/release/infix2.htm
   http://www.jsoftware.com/pipermail/general/2000-October/004573.html
 
Any function  f  :  

   http://www.jsoftware.com/help/release/infix2a.htm
   http://www.jsoftware.com/pipermail/general/2005-June/022851.html

Though I don't see this reflected in my benchmarks:

           ts =:  6!:2 , 7!:2@:]
            Y =:  i.1e7
           
           ts&> '(-~ }. , >:@{:) Y';'(1 ,~ 2 -~/\]) Y' 
        0.188479 1.34219e8
        0.185833  1.3421e8
           
I think this is because we had to compensate for curtailment due to infix with 
a catenation.   Perhaps  x , 2 f/\ y  and  x ,~ 2 f/\ y  are candidates for 
special code?  Tacitly, we could optimize  (noun , 2 f/\ ])  and  (noun ,~ 2 
f/\ ])  .

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

Reply via email to