note 9324 added to odbc_result

FWIW, ADO and similar ODBC-using interfaces also cannot distinguish between A.id and B.id. The easy solution is aliasing in SQL: Consider a table "A" consisting only of the column "id". Next, consider the following query: SELECT * FROM A JOIN B ON A.id=B.id With ODBC, you'd be forced to use the numerical index, rather than the name. However, you could rewrite the query, too: SELECT A.id AS A_id, B.id AS B_id FROM A JOIN B ON A.id=B.id This can be better on three counts: Firstly, unless you really want everything in the result set, it might be faster. SQL servers can be faster with "*", but often the networking will benefit from less data (Or a smaller tuple width, if you're into database jargon). Secondly, since you're forced into thinking about what you want out of the query, you'll probably write better SQL as a result. Thirdly, if you change the query - or tables - slightly, you don't have to revisit all your code to cope with the change. Aplogies for stating the obvious, and having no imagination with my examples. http://www.php.net/manual/function.odbc-result.php