At 02:50 PM 6/11/2006, you wrote:

>Would using derived tables (if available) come up with a plan any
>different to the IN clause anyway?

It depends on the specification of the derived table, I guess...but
this byway of the discussion didn't involve derived tables, it was
about joining to a SSP. (Maybe you are confused in thinking that
"derived tables" is the same thing as the output of a SSP? It's not.)

And the predicate ieldvalue in(<list of literals>) *does* use an
index on fieldvalue, as the otimizer resolves it to fieldvalue =
<first> or fieldvalue = <second> or ....

./heLen