Page 2 of 2

Posted: Wed Apr 09, 2008 11:58 am
by Kef
Dominator wrote:because players generally play bots first... I'm just saying that when bots are in the game humans behave differently then if it was a full house... I know I do.
He's got a point there, might be interesting to see the spread only for 6-human games

Posted: Wed Apr 09, 2008 6:35 pm
by GregM
Kef wrote:
Dominator wrote:because players generally play bots first... I'm just saying that when bots are in the game humans behave differently then if it was a full house... I know I do.
He's got a point there, might be interesting to see the spread only for 6-human games
Makes sense. With Dustin providing game data access we'll be able to run any ranking formula we want on any subset of games, e.g. you could have a ranking list for 6-human Classic 5-10-15 0% exclusively, or weird stuff like only ranking games in which Dustin is a player.

Has Dustin published the current raw formula so we could apply it this way?

Posted: Wed Apr 09, 2008 6:51 pm
by Dominator
GregM wrote:
Kef wrote:
Dominator wrote:because players generally play bots first... I'm just saying that when bots are in the game humans behave differently then if it was a full house... I know I do.
He's got a point there, might be interesting to see the spread only for 6-human games
Makes sense. With Dustin providing game data access we'll be able to run any ranking formula we want on any subset of games, e.g. you could have a ranking list for 6-human Classic 5-10-15 0% exclusively, or weird stuff like only ranking games in which Dustin is a player.

Has Dustin published the current raw formula so we could apply it this way?
I believe the official formula has not been posted...

http://sillysoft.net/forums/viewtopic.php?t=57

Posted: Wed Apr 09, 2008 7:53 pm
by Dominator
the original post was edited

Posted: Wed Apr 09, 2008 9:33 pm
by n00less cluebie
Dominator wrote:the original post was edited
Ok, then let me rephrase my question:

Is it the same as when Last edited by dustin on Sat Sep 30, 2006 5:38 am?

Posted: Wed Apr 09, 2008 9:35 pm
by Dominator
n00less cluebie wrote:
Dominator wrote:the original post was edited
Ok, then let me rephrase my question:

Is it the same as when Last edited by dustin on Sat Sep 30, 2006 5:38 am?
From what I understand, yes. Although only Dustin can answer that one.

Posted: Thu Apr 10, 2008 1:27 pm
by dustin
The algorithm is the same, but some of the values in it may have been tweaked since then.

Posted: Tue Apr 22, 2008 4:37 pm
by SunTzu
Damn! Where was this when I needed it last year?

I had to create a Java program to get all of the data in order to test/tweak my ranking formula.

Whenever I get some free time again I'll take a look at this in more depth...

Posted: Thu Aug 27, 2009 8:24 pm
by dustin
The XML output now includes player ID in addition to player name.

Posted: Thu Aug 27, 2009 8:29 pm
by paranoiarodeo
Woohoo ... thanks d ...

;)

Posted: Sat Oct 31, 2009 1:21 am
by dustin
New limits on the XML access.

It will only return games after 300,000 right now.

It will only return a maximum of 5,000 games at a time.

Re: Online game data XML access

Posted: Tue Sep 08, 2020 2:32 pm
by SecondTermMistake
% curl "http://sillysoft.net/lux/xml/gameHistor ... stGames=10"
<?xml version="1.0" encoding="UTF-8"?>
<gamelist>
</gamelist>%

Am I dong something wrong? =P

Re: Online game data XML access

Posted: Wed Sep 09, 2020 6:30 pm
by SecondTermMistake
For future reference, gotta tell sillysoft.net that I will happily accept XML.

curl -H "Accept: application/xml" "http://sillysoft.net/lux/xml/gameHistor ... stGames=10"