sybase - Is there any performance difference between count(*) and count(1)

Question

Query 1:

SELECT COUNT(1) FROM STUDENTS

Query 2:

SELECT COUNT(*) FROM STUDENTS

Both the queries return the same result, but is there any performance difference between these two ?

What I had heard is the first query would be faster than the second one, but can any one give specific details about it?

possible duplicate of [Count(*) vs Count(1)](http://stackoverflow.com/questions/1221559/count-vs-count1) — a_horse_with_no_name, Aug 27 '12 at 06:40

score 3 · Accepted Answer · edited Aug 27 '12 at 13:45

3

You may use count(*) or count(1), one is not faster than the other. As stated, is just a urban legend :)

One final note, count(*) and count(columnName) may be different!
The first one counts all rows, the second one counts the number of rows where the specified column is not NULL.

edited Aug 27 '12 at 13:45

Jonathan Leffler

698,132
130
858
1,229

answered Aug 27 '12 at 09:13

aF.

62,344
41
131
195

1

Note that COUNT(Column) may require a table scan (slow) where COUNT(*) may not (basic table metadata). – Jonathan Leffler Aug 27 '12 at 13:46

score 1 · Answer 2 · answered Aug 27 '12 at 06:39

1

There is no difference whatsoever between the two statements.

The rumour that count(1) is faster is an urban legend that was never true.

answered Aug 27 '12 at 06:39

a_horse_with_no_name

497,550
91
775
843

1

And this is a DBMS-agnostic answer; it doesn't matter which DBMS you use! – Jonathan Leffler Aug 27 '12 at 06:43
@JonathanLeffler: yes, absolutely. – a_horse_with_no_name Aug 27 '12 at 06:45

sybase - Is there any performance difference between count(*) and count(1)

Query 1:

Query 2:

2 Answers2