Re: [GENERAL] Two slightly different queries produce same results,

2005-11-11 Thread Andrew Schmidt
Hi Richard, While in the process of responding to your email I found out what was doing it. It was the grouping of the team_players team_id instead of the team team_id. Here's some table numbers: team => 31720 rows, team_players => 464896 rows, player => 948 rows player_updates => 5414 row

Re: [GENERAL] Two slightly different queries produce same results,

2005-11-11 Thread Richard Huxton
Andrew Schmidt wrote: However, I've run into a problem where one query took about twice as long as innodb. Some investigation and playing around with the query, I ended up with a slightly different query but was about 3 times as fast as innodb (and 5 times faster than the original query). I d

[GENERAL] Two slightly different queries produce same results, one takes 1/4 the time.. bug in planner?

2005-11-10 Thread Andrew Schmidt
Summary: Two queries that are nearly identical AND use the exact same indices take much different times in execution. Both queries produce the exact same results. One takes 1072ms to execute, the other 262ms. Although I have a fix, it seems more of a hack, and I'd also like to know why it'