[Jprogramming] Computer [|Language] Benchmarks Game

[Joey K Tuttle]

Sorry! This time the 50 50 case was garbled... (left out the hard part!)
It should have been (for the 2.4 Ghz Mac):

     0j7 ": 5 (6!:2) '%. 50 50 ?.@$ 1000'
0.0017531

and on the 3 Ghz Linux machine:

    0j7 ": 5 (6!:2) '%. 50 50 ?.@$ 1000'
0.0029506

The mysterious factor of 2 is still there...

- joey

At 15:18  -0800 2007/12/08, Joey K Tuttle wrote:
>No, actually it is like holding a 1990's mainframe in your
>hand... My expression wasn't garbled (at least by my email
>client) and the one I'm asking about is -
>
>     6!:2 '%. 500 500 ?.@$1000'
>
>the 50 50 was just too brief for comparing --- on the Mac
>
>    0j7 ": 5 (6!:2) '50 50 ?.@$ 1000'
>0.0000700
>
>I agree that using multiple cores would be interesting.
>
>- joey
>
>At 16:02  -0700 2007/12/08, Don Guinn wrote:
>>I assume your test expression is supposed to be 6!:2'%.?50 50$1000' . Must
>>have gotten garbled somewhere.
>>
>>Just to add another timing  - on a Pocket PC - 0.439 seconds for the above
>>expression. That's like holding a 1970s mainframe in your hand!
>>
>>About a PC performing as well as a mainframe - the mainframes always have
>>and still blow PCs (microcomputers) away. It's just that the PC is not many
>>years behind the fastest mainframes. But mainframes and PCs are changing.
>>PCs are dual-core and quad-core. The single CPU mainframe is being replaced
>>by racks of microcomputers working together. Perhaps it's time for J to look
>>into taking advantage of multiple processors available on PCs less than
>>$1000.
>
>----------------------------------------------------------------------
>For information about J forums see http://www.jsoftware.com/forums.htm