Keep in mind my tests were on a screaming Dell Inspiron 1ghz machine running Win2k and PO8. I have tested the ability on a regular desktop against our Oracle server at my office, but no time trials. For what I was doing, the response was most acceptible. When I get into this further, I will do some time trials.

Also keep in mind that the primary reason for using SPs and packages to perform all these functions is for security. In general, security tends to sacrifice some performance. By doing what Larry and I are doing [and others], all we have to grant users is the EXECUTE privilege on our procs and pkgs. This means they have ZERO ability to access and modify data with MS Access, etc., unless they are extremely good guessers at calling these stored procs.

>Well don't go to any special trouble or hurry to test it on my account. I just wondered whether you already had tested SP vs. SPT and noticed similar results. I'm putting together a more extensive test program that will include executing an Oracle SP with the latest MS driver, executing an Oracle SP with the latest Oracle driver, and executing SPT with both drivers.
>
>I'm also doing queries that return about the same number of records from the same table with different query parameters, to minimize caching affects on performance.
>
>Thanks,
>Kelly