>>I have gone well above 200K records, and found that the TOTAL time to return ALL the records from the query was much faster in VFP. Unfortunately (or fortunately), none of my clients have ever asked to view 200K records at one time.
>
>That's a good point and also an important distinction. In a data processsing environment that is primarily transaction oriented, having a powerful local database engine is likely not the highest-priority issue. However, rather than viewing records, what if you instead needed to do some kind of analysis on all 200K records and pull out a subset, then do another operation on that subset (plus other tables)to create another subset, etc? In a data analysis environment, VFP really shines. However, VFP might be the wrong tool for you.

Mark,

I have found that this ability to manimulate the data and combine successive extractions from the database to be an asset not only in data analysis situations, but also in cases where you want to enhance the user interface by presenting information to the user in the most optimal way. This can enhance his work also in an operational environment.

Marc