Row comparision for 2 identical tables

[tamilselvan]

=======
(select cust_num, favourite_colour from table1
minus
select cust_num, favourite_colour from table2)
union all
(select cust_num, favourite_colour from table2
minus
select cust_num, favourite_colour from table1);

========
The above SQL is to run slow.
There will 2 full table scans on table1 and another 2 full table scans on table2. Isn't it bad?

This above SQL was modified by Tom and another brilliant Italian (I forgot his name). Their version is given below:

select c1, c2, c3, count(src1), count(src2)
from
(
    select a.*, 1 SRC1, to_number(null) SRC2 
    from   a
    union  all
    select b.* , to_number(null) SRC1, 2 SRC2
    from   b
)
group by c1, c2, c3
having count(src1) <> count(src2)
/ 

Tamil