Fortran

[Marko Rauhamaa]

Steven D'Aprano <steve at pearwood.info>:

> On Wed, 14 May 2014 07:35:35 +0300, Marko Rauhamaa wrote:
>> I'm having a hard time finding information on how well it performs wrt
>> Java, for example. Did PyPy truly find the Philosophers' Stone?
>
> [...] do you mean, how does PyPy compare with <insert name of Java
> compiler here? I don't know. There are some old benchmarks here:
>
> http://attractivechaos.wordpress.com/2011/04/25/my-programming-language-
> benchmarks-plb/
>
> but the person doing the benchmarks is by his own admission not a
> skilled Python programmer.

That's what I'm after. The answer from the page is:

   Java   1.7  2.6  67.1  6.8  13.4   6.7  314.8
   PyPy  19.5  8.5  84.1  4.0   7.3  12.3  236.0

None of the benchmark tests seems all that relevant, but the one that
comes closest to being interesting is Column 1: Sudoku. There, Java is
an order of magnitude faster (1.7 s vs 19.5 s).

When Python equals Java performance, Java will lose its raison-d'être.


Marko