> So it looks like this is not meaningful for such a small matrix with
current machines.

Reminds me of the following story from Chinese history.

A king from a minor kingdom was visiting a rich man, reputed to be very
rich indeed.  So the king wore a special coat made with the finest fabric
(the equivalent of the golden fleece).  After he arrived at the rich man's
estate, he discovered to his chagrin that the rich man's _servants_ wore
such coats.




On Fri, Jul 10, 2015 at 2:45 PM, J. Patrick Harrington <[email protected]>
wrote:

>
> One last one: The JKT benchmark Roger pointed to gives:
>
>    6!:2 '%. ?50 50$1000'
> 0.013092
>
> So it looks like this is not meaningful for such a small
> matrix with current machines.
>
>
> On Fri, 10 Jul 2015, J. Patrick Harrington wrote:
>
>>
>>   But
>>      1000 ts '%. a"
>> 0.00203444 239040
>>
>>     Hmmm...
>>
>> On Fri, 10 Jul 2015, J. Patrick Harrington wrote:
>>
>>>   a=. 50 50$ rand 2500
>>>   10 ts '%. a'
>>>  0.0049867  239040
>>>
>>>  (ts=: 6!:2 , 7!:2@])
>>>
>>>
>>>
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
>>
>>
>>  ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to