If you are reading from a myisam variable length row table the table handler actually reads the entire row from disk regardless of the columns you choose. I would say store them in separate columns so you can use those in other parts of your queries (where statement ect) Also you don't incur the overhead of sending the entire row each time back to the client for further parsing.
-Eric On Thu, 24 Jun 2004 09:03:37 -0500, Eamon Daly <[EMAIL PROTECTED]> wrote: > > 22 quatloos for you! > > Yeah, in retrospect, I imagine Dan's on the money: the seek > time is where the money's at. Thanks for the feedback, all! > > ____________________________________________________________ > Eamon Daly > > > > > ----- Original Message ----- > From: "Frank Bax" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Sent: Wednesday, June 23, 2004 9:55 PM > Subject: Re: Fetching 12 columns or 1 TEXT field? > > > At 03:29 PM 6/23/04, Eamon Daly wrote: > > >Which do you folks think is faster: randomly accessing a > > >table with a primary key and a dozen CHAR columns or a table > > >with a primary key and a single merged TEXT column? The data > > >in the 11 extra columns will always be fetched as a single > > >request. > > > > Both the same - with extremely minor variance. > > > > >I rolled my own benchmarking program and 10,000 runs each > > >came out just about equal, which surprised me a little. > > > > Wow. I'm right - what's the prize? > > -- > MySQL General Mailing List > For list archives: http://lists.mysql.com/mysql > To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED] > > -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
