subselect vs. no subselect sql with analytical functions

[wakedba]

The optimizer in the explain plan's cost and rows are the same for the two following queries however one is efficient (a) and the one using a subselect is extremely slow (b). Any ideas why this would occur, it seems like it's trying to satisfy the subselect before filtering it with a literal.

(a)
select * from (
select a.a, b.b, a.c, rank () over (partition by b.b order by a.c desc)
from a, b
where a.x=b.x
and a.x = 555)

(b)
select * from (
select a.a, b.b, a.c, rank () over (partition by b.b order by a.c desc)
from a, b
where a.x=b.x)
where x = 555