? tTime1= Seconds() *Select Count(*) from UniqI Group by iUniq into cursor intcount nowait && 20.828 secs no optimiz *Select Count(*) from UniqI into cursor intcount nowait && 4.391 secs no optimiz *Select * from UniqI where iUniq = 79421453 into cursor intcount nowait && .016 secs partial optimiz tTime2 = Seconds() ? "Integer SQL results - " + Str( tTime2 - tTime1 ,8,3 ) ? Sys(3054,1) ? tTime1= Seconds() *Select Count(*) from UniqI Group by iUniq into cursor charcount nowait && 21.219 secs no optimiz *Select Count(*) from UniqC into cursor charcount nowait && 7.156 secs no optimiz *Select * from UniqC where cUniq ="79421453" into cursor charcount nowait && .016 secs partial optimiz tTime2 = Seconds() ? "Char SQL results - " + Str( tTime2 - tTime1 ,8,3 ) ? Sys(3054,1)I understand that the Group By is not optimized (and why isn't it in VFP7?), but why is Count(*) not optimized? And why am I only getting partial optimization when I am querying for a single record only? I also noticed that while the UniqI table is 2/3 the size of UniqC, it is not singificantly faster on the query times.