>Interesting aside here, George ... I got different results with this when run under FPW 2.6 rather than VFP. And the results were inconsistent ... the order of fastest to slowest was different each time you ran it. Care to hazard a guess as to why?? I have no idea .....
>
Hi Bonnie,

I can't think of why this would happen. Much has changed with the FoxPro internals since FPW (obviously). However, one of the more recent changes sped up VFP's string handling significantly. Add that to the changes in the compiler and...

Just one quick question. Wasn't the function call (in both cases) always the slowest? It should have been.