>Are you using Sql's text indexing features on the fields you're interested in? That will significantly speed up searches.

So, basically, wherever I plan to use LIKE with %, I should do a special setting to avoid that situation. Can you confirm me that in VFP this is something we didn't have to do. But, in SQL Server, doing the same, would require some additional setting as its algorythms of indexation is somewhat different than VFP and we might be better to make some adjustments in the text indexing, such as you just described.

As for the text indexing, I just want to make sure we are talking about the same thing here. I am not in need of fuzzy indexing such as searching for a specific string in a field no matter where the string is. I am only in need of what we have in VFP by obtaining immediate search result when I need to search for a field value starting with something. So, based on that, do I still really need to do some text indexing? If yes, where should I do for that?

>Using named parmeters vs. hardcoded values should make no difference. Maybe the slowness is in your returning the results (ie. returning a large data set to the client code) rather than the query execution.

We tested that the other week directly from the interface. The same SQL, when we use LIKE with % will take much longer, in most cases, based on specific algorythms being used, when being parameterized.