Alexandre Benson Smith wrote:

> right ?
> Before I saw the behaviour I explained in my previous post (all tables
> before outer join could be tested for join order, but can mix tables
> before the outer join with tables after the outer join even if the
>
>

This paragraph should be:

Before I saw the behaviour I explained in my previous post (all tables
before outer join could be tested for join order, but *can't* mix tables

> Another reason that leads me to think that the statistics are the
> culprit is that if I use all inner joins the plan is perfect, i.e. the
> index FK_PEDIDOVENDAITEM_PRODUTO is used and the table PEDIDOVENDAITEM
> is the first on the join order, look this:
>

This paragraph should be:

Another reason that leads me to think that the statistics *aren't* the


sorry...

--
Alexandre Benson Smith
Development
THOR Software e Comercial Ltda
Santo Andre - Sao Paulo - Brazil
www.thorsoftware.com.br