>So...are JOINS generally faster than using the subset version?
>
>I was just going to ask if the following would be better implemented with JOINS.
>
>

>SELECT * FROM crsMatchResults ;
>INTO CURSOR crsMatchResults ;
>WHERE ;
>(Sequence in (SELECT Sequence FROM crsSurvivors) OR ;
> Sequence in (SELECT Sequence FROM crsDuplicates)) AND ;
>(GroupNum in (SELECT GroupNum FROM crsSurvivors) AND ;
> GroupNum in (SELECT GroupNum FROM crsDuplicates)) ;
>ORDER BY Sequence ;
>READWRITE NOFILTER
>

>

James,

There's no simple answer. Often JOINs are faster but only tests can give the answer in each particular case.

SELECT * FROM crsMatchResults ;
	INTO CURSOR crsMatchResults ;
	WHERE ;
		Sequence in ( ;
			SELECT Sequence FROM crsSurvivors  ;
			UNION ALL SELECT Sequence FROM crsDuplicates) AND ;
		GroupNum in ( ;
			SELECT GroupNum FROM crsSurvivors ;
			UNION ALL SELECT GroupNum FROM crsDuplicates) ;
	ORDER BY Sequence ;
	READWRITE NOFILTER
	
	
SELECT * FROM crsMatchResults mr ;
	INTO CURSOR crsMatchResults ;
	JOIN ( ;
			SELECT Sequence FROM crsSurvivors  ;
			UNION ALL SELECT Sequence FROM crsDuplicates) dt1 ;
		ON dt1.Sequence = mr.Sequence  ;	
	JOIN ( ;
			SELECT GroupNum FROM crsSurvivors ;
			UNION ALL SELECT GroupNum FROM crsDuplicates) dt2 ;
		ON dt2.GroupNum = mr.GroupNum ;	
	ORDER BY Sequence ;
	READWRITE NOFILTER