A CTE with DENSE_RANK, should i use a Subselect within OR a JOIN Condition - tsql

Question

In the original table there are a lot of records. Is it better to filter within the CTE (example B) or should this be done in the JOIN condition (example A). Or is it possibly all the same, both are equally good/fast?
I guess in the CTE the whole table would have to be prefiltered first and in the JOIN only the corresponding records would be affected.
DECLARE #Orders TABLE(
orderid int NOT NULL,
orderdate datetime NOT NULL,
empid int NOT NULL,
custid varchar(5) NOT NULL,
qty int NOT NULL
);
-- in original lots of data
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(30001, '20020802', 3, 'A', 10);
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(10001, '20021224', 1, 'A', 12);
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(10005, '20021224', 1, 'B', 20);
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(40001, '20030109', 4, 'A', 40);
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(10006, '20030118', 1, 'C', 14);
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(20001, '20030212', 2, 'B', 12);
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(40005, '20040212', 4, 'A', 10);
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(20002, '20040216', 2, 'C', 20);
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(30003, '20040418', 3, 'B', 15);
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(30004, '20020418', 3, 'C', 22);
INSERT INTO #Orders(orderid, orderdate, empid, custid, qty) VALUES(30007, '20020907', 3, 'D', 30);
-- example A
WITH CTE_
AS(
SELECT
orderid
,orderdate
,empid
,custid
,qty
,DENSE_RANK() OVER(PARTITION BY custid ORDER BY orderid) AS [Ranking]
FROM #Orders
)
SELECT
d.*
FROM CTE_ AS d
WHERE d.[Ranking] = 1;
-------------------------------------
-- example B
WITH CTE_
AS(
SELECT
orderid
,orderdate
,empid
,custid
,qty
FROM(
SELECT
orderid
,orderdate
,empid
,custid
,qty
,DENSE_RANK() OVER(PARTITION BY custid ORDER BY orderid) AS [Ranking]
FROM #Orders
) AS d
WHERE d.[Ranking] = 1
)
SELECT
d.*
FROM CTE_ AS d

Answer 1

Generally, it's better to filter within the CTE (example B), especially in case of your tables have indexes in filter conditions, these indexes could be applied to be pre-filtered data first. Since a CTE is a temporary view, it can't have indexes, so in case of example A, your main query could not use indexes (if have) for filter condition in WHERE statement.