TEMP TABLE USING PROBLEM

Hi all!

I have a little problem, when using a temp table ....

explain UPDATE cim SET os_cim_id=-1 FROM T_CNY WHERE cim.cim_id=T_CNY.cim_id;
NOTICE: QUERY PLAN:

Hash Join (cost=26986.59..44340.35 rows=243969 width=129)
-> Seq Scan on t_cny (cost=0.00..3635.69 rows=243969 width=4)
-> Hash (cost=5961.65..5961.65 rows=254965 width=125)
-> Seq Scan on cim (cost=0.00..5961.65 rows=254965 width=125)

In this case t_cny is a temp table and update runs very very slow ...

After I copied t_cny a real table named tcny :

select * into tcny from t_cny;

explain UPDATE cim SET os_cim_id=-1 FROM TCNY WHERE cim.cim_id=TCNY.cim_id;

NOTICE: QUERY PLAN:

Nested Loop (cost=0.00..3321.34 rows=1000 width=129)
-> Seq Scan on tcny (cost=0.00..20.00 rows=1000 width=4)
-> Index Scan using pk_cim on cim (cost=0.00..3.29 rows=1 width=125)

Postgres's query optimizer does not determine a temp table field information ???