Quoth Amir Sela:
> Marc stated he did not benchmark.. Now, I'm not a kernel hacker or even an
> avid programmer, but I'm wondering - How significant are the differences in
> the resulting code, be it user-space programs or kernel-code ? Does it amount
> to anything larger than a 10% difference ?
As I said - I did not bm. Let me qualify. I did NOT bm the k. I _did_ bm
idiotic pieces of code like calculating the \Sigma{1--n}{\sqrt{n}} for
n=100000000 and similar bs.
The Intel compiler gave results from 50% to 300% better.
=================================================================
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
