Giant database schema confusion
Posted by: lance_cotton
Date: July 04, 2006 09:59PM

I am making a chess application that takes hundreds of thousands of games and records the outcomes. Initially I used a system of creating a table for each unique board-layout encountered. In this table I had a 'move' field and 3 outcome fields, win, lose and draw. With php, I could list the 'tried' moves from a given board setup and the success these moves gave. After a move was chosen, the next board setup was determined, that particular table was accessed, the stats queried for and so on.

It worked but the problem of course was that in having these gazillions of tables, search time was brutal.

I read in this forum that querying within a table has far less overhead than across many tables.

So, now I guess I need one giant table with the unique board positions (specially notated) as the primary key (right?). If this is correct, then how can I support attaching an unknown amount of associated moves and outcomes to this board record?
The only way I can think of is by making a long string of appended moves and breaking it apart later with php. The same goes for the wins, losses and draws.

That seems hokey!

Thanks for your advise.
Lance Cotton

Options: ReplyQuote

Written By
Giant database schema confusion
July 04, 2006 09:59PM

Sorry, you can't reply to this topic. It has been closed.

Content reproduced on this site is the property of the respective copyright holders. It is not reviewed in advance by Oracle and does not necessarily represent the opinion of Oracle or any other party.