>>maybe create an index based on entered_dt and status would make it quicker
>>index on status+dtos(entered_dt) tag stat_date
>>set order to stat_date
>>set filter to status=[R]
>>go top
>
>The filter still wouldn't be optimised and so I'd doubt that an index would do very much. If you have a relatively small population of records that satisfy the filter then I'd use a query rather than a filter. The query can be optimised and would return all the records far faster than using FILTER.
>
>Simon

I agee - my method of choice would be a permanent index, SEEK and SCAN into a cursor structure or OCX list server. Even Queries might be sluggish compared to that!