>>>So it is faster for 4-5 rows??? What about 10 or 15? My results with 11 rows:
>>>Using COPY TO ARRAY 10000 passes took 0.330.Records in result = 11
>>>Using SQL SELECT 10000 passes took 0.327.Records in result = 11
>>>Using SQL SELECT to cursor 10000 passes took 0.440.Records in result = 11
>
>Newly restarted VFPA running the code below: Note the table is closed between tests to try to minimize benefit of caching for tests 2 and 3,
>
>Using COPY TO ARRAY, 10000 passes took 0.073; records in result = 12
>Using SQL SELECT, 10000 passes took 0.528; records in result = 12
>Using SQL SELECT to cursor, 10000 passes took 0.463; records in result = 12
>

I can't confirm these results on VFP SP1. Exact same code, copy to array takes 139 ms, while others 240 and 219 ms on my system. Yesterday, I got the results with "set talk off", on data\Employee, but was on batteries, don't know if it was the reason. Today tried on fresh boot of windows. Such an inconsistency also made me think this test is not reliable.

Probably VFPA means what you call "VFP Advanced", which I don't have and don't use. That is interesting in VFPA, SQL-SELECT operations seem to be slower than xbase commands.

Anyway, in either case, I don't see the results worthy enough to choose one over another. They have pros and cons.

PS: For cursor test, you should include NOFILTER for a true cursor and remove "release aFrance..." which explicitly makes it lose more time for no reason. And you should include Local array declaration. That is what one normally want to do in real world app.