[Shootout-list] Ray tracer

[John Skaller]

On Sun, 2005-06-19 at 13:06 -0500, Dave wrote:

> We're all volunteers here - the best way to kill the Shootout would be to
> make things more complicated than they already are.

Indeed .. I've been lazy about writing tests myself,
because it is really complicated thinking about
how to convince the admins to accept them. 

I prefer to spend time working on the tests and
my compiler, and be able to make a simple judgement
about whether my code will be accepted or not.

In addition, perhaps weirdly, the Shootout tests
largely favour my compiler, perhaps unfairly so,
and it concerns me that others may also believe
the Shootout isn't a fair comparison. 

On top of that, it would be useful as the compiler developer
to see where my compiler is not performing as well
as another system, and try to learn how to improve it.
I have done this numerous times -- the first version
of Ackermann I encoded in Felix took 20 minutes 
for ack(3,10) on a friend's work's G4 server and wouldn't
run on my machine at all .. on the x86 it is now
faster than gcc.

-- 
John Skaller <skaller at users dot sourceforge dot net>
Download Felix: http://felix.sf.net