Which pretty much leaves only compiler differences...
That was my conclusion, too.
BTW, on second glance I noticed that your figures are about an order of magnitude slower in absolute terms, compared to what I and the others got — which made me wonder what hardware you were using.
Don't get me wrong, I'm not into some childish "mine is bigger" kind of silly thing... Not at all, it's just that if we presume roughly comparable hardware, what might account for that order-of-magnitude difference? Compiler differences, too?
Just for comparison, the CPU I ran the test on is a 2.3 GHz AMD Phenom 9600 quad-core — which was already pretty "standard" at the time I bought the machine 3 years ago. (The quad-core should be irrelevant here, as the benchmark uses one core only anyway.)
In reply to Re^4: Why does global match run faster than none global?
by Eliya
in thread Why does global match run faster than none global?
by Kc12349
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |