[Shootout-list] commercial benchmarks

[Brandon J. Van Every]

Brent Fulgham wrote:
> Brandon J. Van Every wrote:
>
> I have no objection to creating language inclusion
> policies or benchmarking standards.  We can probably
> even provide reasonable publishing dates, or at
> least cut testing of on a date certain and present
> a snapshot of the results.

That is good to hear.

> I'm less sure what benefits we would derive by
> affiliating with a business entity.  I have no problem
> exploring this, assuming we understand what is being
> proposed.

Then leave it for the future.  No business entity wants to deal with an
ill-defined benchmarking effort anyways, so first things first.

> > I can't address your core culture.  I can only ask
> > if anyone else on the list shares my culture.  If
> > it's just me, I'll go find another way to pursue my
> > goals.
>
> My question was not whether you would address our
> "core culture".  I was asking what you proposed to
> contribute.

Right now, I'll give you my brainpower, such as it is, for hammering out
a more equitable and well-classified set of scoring systems than CRAPS.

If that goes somewhere, I will run Windows benchmarks and perform
necessary ports.

But first I have to secure my interest: a better scoring system.
Something worthy of commerce and PHBs.  Otherwise there is no point for
me.  I'm not in this 'for fun', I'm in this to promote some languages at
the expense of other languages.

Now I will candidly reveal my biases.  I want C++, Java, and C# to die.
I think they suck, and I want industry to move on.  My current favored
replacement language is OCaml, but I've been stymied by tools and
libraries issues on Windows.  In theory, OCaml could be great; in
practice, OCaml is always an invitation to do lots of extra work on
Windows.

Recently I've decided that the Nebula2 3D engine
http://nebuladevice.cubik.org is more important than my choice of
language, and that whatever works best in the Nebula2 toolchain is what
I need to do at the moment.  At present that means Visual Studio
support, with Eclipse possibly in the future.

Anyways, tools and libraries support are very important dimensions of a
language.  Commercially, they are probably the critical dimensions.
But, these are difficult to quantify and benchmark, so for now I
entertain the more tractable step of running this-or-that program
against the other to see which is faster.

Because of my biases, I see a garbage collection benchmark as rather
important.  I would contribute to garbage collection benchmarking
efforts.

If I ever get off my ass and learn digital art skills, and ever make a
real website for myself, I will loan whatever visual arts polish I can
to the enterprise.  At present I have none.  I did, however, help try to
get a Python marketing effort going.  That's what the following website
mockup was about:
http://www.pollenation.net/assets/public/python-main-oct05.png  To
protect the guilty, I won't reveal my own logo submissions... as I said,
I need to learn digital art skills.  I'm just reasonably ok with
traditional media.

Anyways, for me the action item at present is: an equitable,
better-categorized, standard scoring system.


Cheers,                     www.indiegamedesign.com
Brandon Van Every           Seattle, WA

When no one else sells courage, supply and demand take hold.