>I just ran a simple select * query in the Query Analyzer. It reported an execution time of 00:00:04. I ran the same query in VFP and it was so fast it wouldn't even register on the clock. You really don't need the clock at all to tell which is faster. VFP is so fast that the results are in face your instantly. In SQL Server you can actually see the clock counting.
>
>Then try to scroll through the result sets. VFP moves so fast the gird lines are a blur, in SQL Server it moves much slower, and if you move the scroll bar too far you can overload it and it will jump back.
>
>My machine is a pentium 550 with 250 RAM.
>

Charlie,

Your test is totally invalid for as a measure of teh speed of the two data engines. The difference between .04 seconds and "so fast you can't tell" is meaningless. Just, exactly, how long is "so fast you can't tell"? I would venture that you would have a very difficult time saying "No" twice in .04 seconds. You may just be seeing that SQL Server is able to measure smaller time intervals than VFP can. VFP may actually be taking .05 seconds to do the query and that would negate your conclusion.

As for browsing the data later, that is not comparing the VFP vs. SQL Server data engines, it is comparing the VFP grid/browse vs. that abomination that SQL Server uses for displaying data (remember SQL Server is not designed or intended to display data). If you want to test the VFP vs. SQL Server data engines with grids or browses, then you would need to neutralize the influence of the control you use. That would mean either using a VFP grid to view both sets of data and seeing how fast you can scroll, or using some third party grid control. With the influence of the grid/browse object in the tests the results are meaningless in respect to the hypothesis.