Subselects better than multiple wheres?

[jennyflower]

Hi, I have a query like this:

select tb1.fd1, tb1.fd2, tb1.fd3, tb2.fd1, tb2.fd2, tb2.fd3 from tb1 join tb2 on tb1.fd1=tb2.fd1 where tb1.fd1 = 1 and (tb1.fd2 = 'P' or tb1.fd2 = 'Q' or tb1.fd2 = 'R') and tb1.fd3 > CURRENT_DATE and tb2.fd1 = 3 and tb2.fd2 <= CURRENT_DATE and tb2.fd3 != NULL order by tb1.fd1

would this type of query be better optimised and probably run faster if i were to split it into 2 subselects like

select tb1.fd1, qrytb1.fd2, qrytb1.fd3, qrytb2.fd1, qrytb2.fd2, qrytb2.fd3
from tb1
join (select * from tb1) as qrytb1 on tb1.fd1=qrytb1.fd1
join (select * from tb2) as qrytb2 on tb1.fd1=qrytb2.fd1
order by tb1.fd1

Hope this makes sese to someone!!

Jen

 

[ceco]

i can't tell for sure, because these two queries look to me different, but you can use

EXPLAIN ANALYZE and the server will tell you, what is doing and how long does it takes