[kde-solaris] Re: fibonacci

[Stefan Teleman]

GCC's optimizations for Intel are pretty good.

Just out of curiosity, if someone who has access to Forte on x86, 
would be so kind and could run this little benchmark on Solaris x86 
with Forte 9 and post the results. I have read that it smokes GCC, 
but i can't confirm it myself, because i don't have a Solaris x86 box 
*yet*.

<GRIN>But i am getting a W2100Z box in January. And an UltraSPARC III 
Blade 2000.</GRIN>. That means KDE for AMD/Opteron and UltraSparc 
III. :-) Yes, with Forte, not GCC. ;-P

--Stefan

-----

On Friday 19 November 2004 07:44, Christopher Layne wrote:
> Kinda funny:
>
> [clayne at build02-sol8-x86 /tmp]$ gcc -O3 -o fib fib.c
> [clayne at build02-sol8-x86 /tmp]$ time -p ./fib 36
> ./fib: calculated 10 fibonacci(36) in 13.000000 seconds
> real 13.72
> user 13.68
> sys 0.00
> [clayne at build02-sol8-x86 /tmp]$ gcc -Os -o fib fib.c
> [clayne at build02-sol8-x86 /tmp]$ time -p ./fib 36
> ./fib: calculated 10 fibonacci(36) in 12.000000 seconds
> real 12.59
> user 12.58
> sys 0.02
> [clayne at build02-sol8-x86 /tmp]$ gcc -Os -fomit-frame-pointer -o fib
> fib.c [clayne at build02-sol8-x86 /tmp]$ time -p ./fib 36
> ./fib: calculated 10 fibonacci(36) in 10.000000 seconds
> real 10.75
> user 10.75
> sys 0.01
>
>
> None of these tests mean much as CPU types are wildly different.
> ___________________________________________________
> This message is from the kde-solaris mailing list.
> Account management: 
> https://mail.kde.org/mailman/listinfo/kde-solaris. Archives:
> http://lists.kde.org/.
> More info: http://www.kde.org/faq.html.

-- 
Stefan Teleman          'Nobody Expects the Spanish Inquisition'
steleman at nyc.rr.com                          -Monty Python