In help it says that empty() is not optimizable and that "#" is.

I originally was thinking this wasn't correct because I had a select-sql clause that was much faster if my expression was "ino_acct = nAcct and !empty(ino_total)" instead of "ino_acct = nAcct and ino_total # 0.00".

My invoice table has around a half million records.

Anyway, after awhile I came to the conclusion, (whether its correct or not I don't know), that the "#" operator is IN FACT optimizable and that is why its slower! Is this possible? When I use the non-optimizable empty() function rushmore only acts upon my first expression. Is this correct?

If so, I wonder if there are any golden rules of thumb as to when I should or shouldn't use optimization.

I hope this makes sense,

Thanks,
Paul