John,

>
>If VFP7 now provides a significant performance benefit for using m. on the right side of variable assignments, I'd have to ask WHY. How does that help us?
> < snip >
>I do not want to revert to m. and would like to know what VFP7 improvement has taken place with the unfortunate side effect described!
>

Here are the results he posted:

>Now my results (Athlon 650Mhz, 192Mb,WinXP pro):
>VFP6 :
>11.59 - 11.92 (ignorable)
>VFP7 :
>11.95 - 37.01 (oops. both are slower than VFP6 results AND difference is not ignorable)

It looks more like a penalty for *not* using m., rather than a performance benefit or improvement. Rather odd, I think...