compare sql tables displaying different result

The query below returns all records that are not registered in the release table:

SELECT
    cadastro.id
 FROM
     cadastro
 WHERE
       NOT EXISTS ( SELECT * FROM liberar where liberar.aluno =  cadastro.id )
 GROUP BY
     cadastro.nome
 ORDER BY
     cadastro.nome

However you can not filter by the release.title because the title exists only in the release table, and in the records you want there are no items in the release tables.

I'll put two examples where I want customers who do not have sales

1 - 

SELECT DISTINCT.* FROM CLIENTES C
LEFT JOIN VENDAS V ON(V.CODID = C.CODID)
WHERE (V.CODID IS NULL)

2 - 

SELECT DISTINCT C.* FROM CLIENTES
WHERE (CODID NOT IN(SELECT CODID FROM VENDAS))

Both of them bring the same result, what changes is the performance depending on the amount of records in each table

Assuming that the CLIENTS table is huge, the first example will be slower, if SALES is the second example will be slower

You should take into account the appropriate indexes created in each table and the additional filters in the WHERE clause

Solved with the tip of Marciano Machado, thank you!

  

NOT EXISTS (SELECT * FROM release WHERE student = cadastro.id)

"SELECT cadastro.*, liberar.* FROM cadastro INNER JOIN liberar WHERE liberar.title = ? AND NOT EXISTS (SELECT * FROM liberar WHERE aluno =  cadastro.id ) GROUP BY cadastro.nome ORDER BY cadastro.nome"
$verifica->execute(array($identificador));

I have adapted this way: includes the join of the tables to retrieve the title because it is the title that I retrieve by the $ identifier variable in the GET, this variable filters the query by title, since the application will have titles based on school subjects.