Hi Robert

>I was under the impression that when you run a VFP view, the 1st step is for the back-end server to send only index info to the VFP data engine from which VFP then determines the records it needs based on the index info provided by the server and then re-send the SQL statement requesting only the needed records based on data it had read from the index info. Thus, the VFP engine was doing much of the selection work on the local workstation and because of the first step of sending the index info to VFP, you were increasing the traffic over the LAN/WAN.

That's Rushmore technology. It only applies when working with VFP data with SQL or SET FILTER, or any FOR clauses, not remote backends although a similar optimization occurs inside the backend database server.

>Futher, I thought that I had heard (or read) that by using SQL Pass-through you eliminate the first set of sending index criteria to VFP from the back-end server because your are sending the "packaged" SQL statement to the server for it to process. Since all of the processing is done by the server data engine and only the requested records are sent back to VFP, you limit the traffic across the LAN/WAN.
>
>If I am totally confused on this issue, let me ask it another way. Does a view vs. a PT statement generate the same amount of traffic accross the network?

Remote views or SPT basically send a query to the backend database server (not the file server) which processes the query and returns a result set. No index data is returned.