Re: Is there a difference between these SELECTS's?

Nils Myklebust wrote: > > belakimem@aol.com (Belakimem) wrote: > > :select * from A where A.numb in ( select numb from B ) > : > :select A.* from B,A where B.numb=A.numb > There is a difference. If there are duplicate numb rows in B that are also in A, lets say 1 row in A that joins with 3 rows in B, then the first select will only return 1 row whereas the second select will return 3 rows. So it really depends on what you're after when you decide which to use. And I'm not sure why, but for the first select I've found sometimes its faster to say: select * from A where A.numb in (select unique numb from B) Maybe because if theres alot of duplicates in B then it creates a smaller temp table? Cheers, Douglas Wilson > All of these selects should return the same rows even in the face of > nulls in A.numb of B.numb. (In the result there will be no nulls in > the numb column.)