1a. test perl loop with $i=10

$ time perl -e 'for($i=0;$i<=10;$i++){}'

real    0m0.003s
user    0m0.000s
sys 0m0.000s

1b. test PHP loop with $i=10

$ time php -r 'for($i=0;$i<=10;$i++){}'

real    0m0.027s
user    0m0.020s
sys 0m0.004s


in average test 1a and 1b, perl is 9x or 10x faster than PHP

------------------------------------------------------------

2a. test perl loop with $i=10000

$ time perl -e 'for($i=0;$i<=10000;$i++){}'

real    0m0.005s
user    0m0.004s
sys 0m0.000s

2b. test PHP loop with $i=10000

$ time php -r 'for($i=0;$i<=10000;$i++){}'

real    0m0.027s
user    0m0.024s
sys 0m0.000s

in test 2a and 2b, perl still much faster than PHP

------------------------------------------------------------

3a. test perl loop with $i=1000000

$ time perl -e 'for($i=0;$i<=1000000;$i++){}'

real    0m0.091s
user    0m0.088s
sys 0m0.000s

3b. test PHP loop with $i=1000000

$ time php -r 'for($i=0;$i<=1000000;$i++){}'

real    0m0.045s
user    0m0.044s
sys 0m0.004s

in test 3a and 3b, PHP already faster then perl

------------------------------------------------------------

now, use extreme loop number like all benchmark code  
in alioth benchmark game:

4a. test perl loop with $i=100000000

$ time perl -e 'for($i=0;$i<=100000000;$i++){}'

real    0m7.310s
user    0m7.304s
sys 0m0.000s

4b. test PHP loop with $i=100000000

$ time php -r 'for($i=0;$i<=100000000;$i++){}'

real    0m1.624s
user    0m1.616s
sys 0m0.004s

in test 4a and 4b, PHP is much much faster then perl