[Numpy-discussion] Speed degression?

[Steffen Loeck]

On Thursday 06 July 2006, Travis Oliphant wrote:
> This can test the speed of the implementations.

With the latest modifications i get the following

Results (in usec per loop):
                                     sin-array            mod-array  
			              use_scalarmath/use_pythonmath                                    

 Numeric                                134                   18         
  
 numpy 0.9.8                             97                   55

 numpy 0.9.9.2764                      114/78                55/21
   
 numpy 0.9.8 + math                      38

 numpy 0.9.9.2764 + math                38/23
  
 Numeric + math                          23


The newest version seems to be faster then Numeric in all sin-array tests. 
Also the speed of mod-array has improved a lot.

Many thanks,
Steffen