>>Hi Sergey,
>>
>>In message id 648030 Cetin says:
>>"If you don't include m. in for loops VFP7 is slower than previous versions (documented somewhere)."
>>Do you know anything about this?
>>
>Nope, but there's a result of my test in VFP7 SP1 with 5,000,000 iterations on P III 1Gz

>  a = 1    - 2.694 sec
>m.a = 1    - 2.741 sec
>

As you can see the difference is negligible.

Try something different:

ab=ba && Slow according Cetin
ab=m.ba && Should be the winner
m.ab=m.ba && Should be similar to the above

I don't have VFP7, so can not test...