"Stembridge, Michael" <[EMAIL PROTECTED]> wrote on 03/14/2005 02:18:25 PM:
> I noticed another listmember used COUNT(fieldname) instead of COUNT(*). > > Is there a noticeable performance increase with COUNTing a column name > instead of all columns? (ie, like SELECTing specific columns instead of > using SELECT *) > > Thanks! > COUNT() only counts non-null values. A long time ago, I once had a table where every column could have been null (yes it was a bad design but I was much younger then). COUNT(*) did not count those rows that were COMPLETELY NULL. I haven't tested this recently so I no longer know if it's still true (and it wasn't a MySQL database I was using, either) More on topic...if you say COUNT(fieldname), you say that you want to count all of the non-null values in that column. Is it faster than select(*)? Probably not but you are asking for something different when you phrase your question that way, aren't you. You are asking "how many non-null values are in this column" and not "how many rows are not null". Shawn Green Database Administrator Unimin Corporation - Spruce Pine
