[Shootout-list] Java benchmarks

[James McIlree]

Brent,

	I think this should be a clear case of
"which performs better on the shootout tests?",
rather than a popularity or consensus vote.

	I will take the time to run all the current
Java tests in both server and client versions,
and report the results to the list.

	I will also run the IBM jdk and report those
numbers.

	James M

On Mar 29, 2005, at 11:19 AM, Brent Fulgham wrote:

>
> --- James McIlree <ovrskeek@mac.com> wrote:
>>
>> 	I'd like to raise an issue about the Java
>> benchmarks. It looks to me as if they are all being
>> run with the "-server" option.
>
> Yes, because that is what the majority of Java
> advocates demanded that we use.
>
> I do not have time to switch back and forth between
> different versions of JDK and server/client settings.
> I am happy to use whatever configuration people
> want, but I refuse to switch based on the current
> batch of e-mail.
>
> If you can create some kind of consensus on this
> issue, I'll go with the desired choice.  But we have
> already been hammered with hate mail from people
> wanting it done as '-server' and I'm not inclined
> to change it unless the Java Lobby, Sun, IBM, or
> some representative group identifies the set of flags
> they want to run under.
>
>> As you can see, it would be really nice to have the
>> IBM jdk available for testing as well. It seems to
> be
>> generally faster than the Sun jdk.
>
> We have the following java implementations in the
> test:
>
> GNU Compiler and Interpreted
> Kaffe
> SableVM
> SUN Java
>
> Which of these should be replaced with IBM's compiler?
>
> -Brent
>
> _______________________________________________
> Shootout-list mailing list
> Shootout-list@lists.alioth.debian.org
> http://lists.alioth.debian.org/mailman/listinfo/shootout-list
>