Autotrace query results question

Hi All,

I am trying to understand the difference between the results of my autotrace  (see results attached)

I am basically running a very simple query, and although the differences don't seem drastic I want to get a handle on the results:

The 2 queries are:

Query 1 -->     select 'x' from dual where exists (select 'x' from codeuser where cdus_user = user);

Query 2 -->     select count(*) from codeuser where cdus_user = user;

The CPU cost is more on the Query 1, although it seems to be reading less bytes.

The CPU cost is less on Query 2 plus it seems to be sending less over the network (bytes sent via sql*net to client). During the executing phase it seems to be reading more bytes.