Form groups of consecutive rows with same value

I have a situation I think can be solved using window function but I'm not sure. Imagine the following table

CREATE TABLE tmp (
  date timestamp
, id_type integer
) ;

INSERT INTO tmp (date, id_type)
VALUES
    ( '2017-01-10 07:19:21.0', 3 ),
    ( '2017-01-10 07:19:22.0', 3 ),
    ( '2017-01-10 07:19:23.1', 3 ),
    ( '2017-01-10 07:19:24.1', 3 ),
    ( '2017-01-10 07:19:25.0', 3 ),
    ( '2017-01-10 07:19:26.0', 5 ),
    ( '2017-01-10 07:19:27.1', 3 ),
    ( '2017-01-10 07:19:28.0', 5 ),
    ( '2017-01-10 07:19:29.0', 5 ),
    ( '2017-01-10 07:19:30.1', 3 ),
    ( '2017-01-10 07:19:31.0', 5 ),
    ( '2017-01-10 07:19:32.0', 3 ),
    ( '2017-01-10 07:19:33.1', 5 ),
    ( '2017-01-10 07:19:35.0', 5 ),
    ( '2017-01-10 07:19:36.1', 5 ),
    ( '2017-01-10 07:19:37.1', 5 );

I'd like to have a new group at each change of value in column id_type. E.G. 1st group from 7:19:21 to 7:19:25, 2nd starting and finishing at 7:19:26, and so on. At this moment, using the query below ...

SELECT distinct 
    min(min(date)) over w as begin, 
    max(max(date)) over w as end,   
    id_type
FROM tmp
GROUP BY id_type
WINDOW w AS (PARTITION BY id_type)
ORDER BY begin;

I get the following result:

begin					end                     id_type
2017-01-10 07:19:21.0	2017-01-10 07:19:32.0	3
2017-01-10 07:19:26.0	2017-01-10 07:19:37.1	5

While I'd like:

begin					end                     id_type
2017-01-10 07:19:21.0   2017-01-10 07:19:25.0   3
2017-01-10 07:19:26.0   2017-01-10 07:19:26.0   5
2017-01-10 07:19:27.1   2017-01-10 07:19:27.1   3
2017-01-10 07:19:28.0   2017-01-10 07:19:29.0   5
2017-01-10 07:19:30.1   2017-01-10 07:19:30.1   3
2017-01-10 07:19:31.0   2017-01-10 07:19:31.0   5
2017-01-10 07:19:32.0   2017-01-10 07:19:32.0   3
2017-01-10 07:19:33.1   2017-01-10 07:19:37.1   5

Once that works, I want to include more criteria to define groups, and these others will be nullable. Postgres Version: 8.4. We have Postgres with PostGis, so it is not easy to upgrade. PostGis functions change names and there are other problems, but we are already rewriting everything and the new version will use a newer version 9.X with PostGis 2.x.