>>
>>IMHO, Rushmore is about 'Where' expressions, i.e. field list is rushmore-irrelevant. I think, the first internal step is collect all records by joint criteria (here Rushmore used), and then Fox looks at field list.
>
>OK,
>but I my opinion Rushmore SHOULD be about using existing indexes to achieve the fastest query performance possible.
>
>To ignore UNIQUE index tags is a waste and to be able to beat the performance of SQL by an order of magnitude
>with old style xbase code is crazy.

Unless there is only one table in the SQL, the DISTINCT or
GROUP BY can only be aftermath based on what was retrieved.
I don't even think you could use the unique index if you
were to code out multiple table joining.
And if it's only one table, just set the index and your done.
This sounds like taking the car instead of walking across
the street.