> I was wondering if that's really so. Wouldn't the marginal speed improvement > be quite small? Is the percentage of the final rowcount really a criterion?
Each transaction costs at least 1 disk seek. Doing thousands of seeks the result would be very slow. John _______________________________________________ sqlite-users mailing list firstname.lastname@example.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users