Arimaa Forum
        (http://arimaa.com/arimaa/forum/cgi/YaBB.cgi)
      

        Arimaa >> Events >> 2013 Postal Mixer
        
(Message started by: rbarreira on Mar 20th, 2013, 11:02am)
Title: 2013 Postal Mixer
Post by rbarreira on Mar 20th, 2013, 11:02am
In this year's postal mixer, there are only 10 players registered so far (3 bots and 7 humans). The registration ends this month.
Title: Re: 2013 Postal Mixer
Post by ChrisB on Mar 20th, 2013, 11:06am

on 03/20/13 at 11:02:05, rbarreira wrote:
In this year's postal mixer, there are only 10 players registered so far (3 bots and 7 humans). The registration ends this month.


The link to register is: http://arimaa.com/arimaa/tourn/postal/2013/

You can choose the number of games you play and it looks like it's free!  :)
Title: Re: 2013 Postal Mixer
Post by odin73 on Mar 20th, 2013, 12:10pm
I didn´t know that the registration is already open, more players may be non-informed either.

Anybody can put a hint to the announcement menue of the gameroom?
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Mar 21st, 2013, 8:44am
I linked registration from the announcements.
Title: Re: 2013 Postal Mixer
Post by robinz on Mar 23rd, 2013, 6:29am
Excellent - glad there is no entry fee. Thanks for letting me know via this thread - I have now signed up  ;D
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Mar 26th, 2013, 9:14pm
Looks like we are on track to beat our 42 participants from 2010, making this year the biggest tournament ever, although I never updated my table for 2012.  Note that 2010 was the other year that we didn't have a registration fee, and the percentage of unfinished games that year was not too bad.


Year    .    .  2005  2006  2007  2008  2009  2010  2011
Players .    .    16    19    20    15    14    42    29
Games   .    .    80    95    85    84    70   147   141
Mismatches   .     9     5     3    21    19    23    16
Not Finished .    20    31    18     8     4    31    49

% Mismatches .    11     5     4    25    27    16    11
% Not Finished    25    33    21    10     6    21    35
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Mar 26th, 2013, 9:17pm
By the way, Omar, don't forget to do the pairing by WHRP this year.  If players don't have a WHRP, you can fall back on game room rating or something else, but it seems like obviously most sensible to pair a postal tournament primarily by postal rating.
Title: Re: 2013 Postal Mixer
Post by Nombril on Mar 31st, 2013, 7:38pm
:-/

I had meant to register, I thought I had until the end of the month.  But I guess I had until 37 minutes ago.  Serves me right for waiting until the last minute, I guess.
Title: Re: 2013 Postal Mixer
Post by thomastanck on Apr 1st, 2013, 1:10am

on 03/26/13 at 21:14:08, Fritzlein wrote:
Looks like we are on track to beat our 42 participants from 2010, making this year the biggest tournament ever, although I never updated my table for 2012.  Note that 2010 was the other year that we didn't have a registration fee, and the percentage of unfinished games that year was not too bad.


Year    .    .  2005  2006  2007  2008  2009  2010  2011
Players .    .    16    19    20    15    14    42    29
Games   .    .    80    95    85    84    70   147   141
Mismatches   .     9     5     3    21    19    23    16
Not Finished .    20    31    18     8     4    31    49

% Mismatches .    11     5     4    25    27    16    11
% Not Finished    25    33    21    10     6    21    35


With 45 (possibly 46 if Nombril becomes an exception) players this year, three of which are bots, we just beat the record!

Can't wait to see how trash fails. Trash's gameroom rating is very inaccurate because he's either played by people who can easily beat him or people who can't play lightning.
Title: Re: 2013 Postal Mixer
Post by browni3141 on Apr 1st, 2013, 3:13am
Come on, let Nombril in! :)
Title: Re: 2013 Postal Mixer
Post by browni3141 on Apr 1st, 2013, 3:17am

on 03/26/13 at 21:14:08, Fritzlein wrote:
Looks like we are on track to beat our 42 participants from 2010, making this year the biggest tournament ever, although I never updated my table for 2012.  Note that 2010 was the other year that we didn't have a registration fee, and the percentage of unfinished games that year was not too bad.


Year    .    .  2005  2006  2007  2008  2009  2010  2011
Players .    .    16    19    20    15    14    42    29
Games   .    .    80    95    85    84    70   147   141
Mismatches   .     9     5     3    21    19    23    16
Not Finished .    20    31    18     8     4    31    49

% Mismatches .    11     5     4    25    27    16    11
% Not Finished    25    33    21    10     6    21    35

Would it not be better to consider the number of players that didn't finish at least one of their games due to their own time-out if you're trying to assess how an entry fee affects player commitment? One player playing a lot of games and timing them all out would mess up the data if this is what you're concerned with.
Title: Re: 2013 Postal Mixer
Post by rbarreira on Apr 1st, 2013, 3:27am

on 04/01/13 at 03:13:10, browni3141 wrote:
Come on, let Nombril in! :)


I don't know if Nombril wants to ask for an exception, but if he does it's probably best to contact Omar directly, as Omar does not always keep tabs on the forum.
Title: Re: 2013 Postal Mixer
Post by Nombril on Apr 1st, 2013, 9:36am

on 04/01/13 at 03:27:26, rbarreira wrote:
I don't know if Nombril wants to ask for an exception, but if he does it's probably best to contact Omar directly, as Omar does not always keep tabs on the forum.

No tournament director is listed as a contact, so I decided not to email anyone directly.


on 04/01/13 at 03:17:34, browni3141 wrote:
Would it not be better to consider the number of players that didn't finish at least one of their games due to their own time-out if you're trying to assess how an entry fee affects player commitment? One player playing a lot of games and timing them all out would mess up the data if this is what you're concerned with.

If we are trying to accurately measure "commitment", would you rather want to measure how how many players abandoned the tournament (% players abandoned), as indicated by timing out for the last game(s) finished by the player?  (This would specifically exclude those games timed out when the player just missed one game out of many, but was still continuing to play others.)
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Apr 1st, 2013, 4:18pm

on 04/01/13 at 09:36:54, Nombril wrote:
If we are trying to accurately measure "commitment", would you rather want to measure how how many players abandoned the tournament (% players abandoned), as indicated by timing out for the last game(s) finished by the player?  (This would specifically exclude those games timed out when the player just missed one game out of many, but was still continuing to play others.)

Yes, I was trying to measure commitment, and this seems like a better measure than I was using.  The case it misses, however, was when MrBrain quit by resigning all of his games rather than timing out.
Title: Re: 2013 Postal Mixer
Post by Hippo on Apr 2nd, 2013, 12:40pm
And this year the fastest loser is bot_trash. Finishing 23 turns during a day.
Title: Re: 2013 Postal Mixer
Post by rbarreira on Apr 3rd, 2013, 4:32am
In the space of 5 minutes, bot_Hippo moved in both games vs bot_briareus. So it's not always thinking for 8 hours?
Title: Re: 2013 Postal Mixer
Post by Hippo on Apr 5th, 2013, 2:41pm
Ricardo, I have said it thinks for 8 hours single thread, but 5 copies of the bot are run in parallel. I would run 6 copies (on 8 core machine), but the memory consumption of the copy version is enormous so I am afraid of swap possibilities ... . I hope to debug the undo version early and it could solve the memory problems to allow 6 parallel runs.

When I last checked it the bot was gaining reserve as it was thinking in all 5 of its games where it was its turn.
Title: Re: 2013 Postal Mixer
Post by mistre on May 5th, 2013, 8:15pm
The link in the menu still points to the 2012 postal tournament and not this years - http://arimaa.com/arimaa/tourn/postal/2012/showGames.cgi

No big deal, but I keep forgetting to click on the one in the middle of the page...
Title: Re: 2013 Postal Mixer
Post by Hippo on May 6th, 2013, 7:53am

on 05/05/13 at 20:15:56, mistre wrote:
The link in the menu still points to the 2012 postal tournament and not this years - http://arimaa.com/arimaa/tourn/postal/2012/showGames.cgi

No big deal, but I keep forgetting to click on the one in the middle of the page...


Yes and now the link from announcements disappeared, so the only remaining link is the mentioned link to 2012 from pulldown menu events>postal mixer.
Title: Re: 2013 Postal Mixer
Post by mistre on May 6th, 2013, 8:28am
If you click the text "games are underway" under 2013 Postal Mixer in the announcements, it takes you to the correct page.

The correct link is http://arimaa.com/arimaa/tourn/postal/2013/showGames.cgi just in case that announcement goes away.
Title: Re: 2013 Postal Mixer
Post by Alfons on Jul 6th, 2013, 11:56am
What's upt with bot_trash?
It seems it does make any more moves...
(and he's the only relaxing oponent I have  ;) )
Title: Re: 2013 Postal Mixer
Post by supersamu on Jul 7th, 2013, 3:07pm
Thomastanck, the developer of bot_trash was last seen 2 months ago on Arimaa.com.
This is a game where bot_trash already timed out: http://arimaa.com/arimaa/gameroom/comments.cgi?gid=269572
You will probably win your game on time.
Title: Re: 2013 Postal Mixer
Post by Hippo on Jul 24th, 2013, 3:53pm

on 07/07/13 at 15:07:21, supersamu wrote:
Thomastanck, the developer of bot_trash was last seen 2 months ago on Arimaa.com.
This is a game where bot_trash already timed out: http://arimaa.com/arimaa/gameroom/comments.cgi?gid=269572
You will probably win your game on time.


So my bot (bot_Hippo) as well as bot_trash finished their postal mixer games. My bot lost no game on time in contrary to the most games of bot_trash.

Unfortunately my bot won only one game on the board and was probably winning it's games against trash as well. I have converted my bot to do/undo rather to copying bitboards version during the mixer, but it had no notable influence on the search speed. I almost let the evaluation function unchanged. Trap controll understanding, mobility understanding (and forks) are the most obvious problems of it.... Very long way to make it play well.
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Jul 27th, 2013, 1:21pm
Sadly for carbon-based intelligence, it looks like bot_briareus will eclipse its 10-4 showing from last year.  So far briareus is 10-1 and the worst it is likely to finish is 13-5, which is counting on both browni3141 and me to win from positions that are currently materially even with no (bad for bot understanding) elephant deadlock.  Only ChrisB and Alfons have dominating positions.

I continue to be amazed that briareus doesn't seem to suffer from the longer time controls in the way that most bots do.  Even given the relative time handicap that human players aren't thinking as long per move as bots, most bots seem to slightly underperform their 2-minute-per-move rating when they play in the Postal Mixer.  I wonder what briareus's secret is for ourperforming.

In any case, I offer my premature congratulations to Ricardo.
Title: Re: 2013 Postal Mixer
Post by rbarreira on Jul 28th, 2013, 4:44pm
Thanks Fritz! I don't know what the difference is to other bots, if there's indeed a difference. bot_hippo seems to have better WHRP than gameroom rating as well?

Maybe next year marwin, sharp or quad will participate in the mixer as well (hint, hint, hint), it would certainly be interesting to get more data about this.

PS: BTW, bot_briareus is currently on vacation from the postal, due to my own vacation. It'll be back in around two days (leaving the computer on 24/7 seemed like a bad idea with so few games left, there would probably be big periods of inactivity).
Title: Re: 2013 Postal Mixer
Post by browni3141 on Jul 28th, 2013, 7:44pm

on 07/27/13 at 13:21:54, Fritzlein wrote:
Sadly for carbon-based intelligence, it looks like bot_briareus will eclipse its 10-4 showing from last year.  So far briareus is 10-1 and the worst it is likely to finish is 13-5, which is counting on both browni3141 and me to win from positions that are currently materially even with no (bad for bot understanding) elephant deadlock.  Only ChrisB and Alfons have dominating positions.
In any case, I offer my premature congratulations to Ricardo.

Personally I would discard my timed-out game for the purpose of comparing records. I had a significant advantage when it ended and would most likely have won.

I find it strange to call DD for hr even, although it seems roughly so much of the time. The game has been kind of crazy and I've given up on winning quickly, but I think you can count on me winning eventually ;)
Title: Re: 2013 Postal Mixer
Post by rbarreira on Aug 23rd, 2013, 2:17pm

on 07/27/13 at 13:21:54, Fritzlein wrote:
So far briareus is 10-1 and the worst it is likely to finish is 13-5, which is counting on both browni3141 and me to win from positions that are currently materially even with no (bad for bot understanding) elephant deadlock.  Only ChrisB and Alfons have dominating positions.


13-5 as a worst case is now a lock. Now to see if it can make it 14-4, 15-3 or 16-2 (unlikely) !
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Sep 15th, 2013, 1:19am

on 08/23/13 at 14:17:01, rbarreira wrote:
13-5 as a worst case is now a lock. Now to see if it can make it 14-4, 15-3 or 16-2 (unlikely) !

Whew, I finally won in 80 moves, the longest game of the tournament so far.  I remember when we thought 80-move games would become increasingly common, but instead they have become rarities.  I shouldn't speak too soon though, as there are games yet to finish in the Postal Mixer.

With browni scoring a sexy win in a goal race, briareus stands at 13-4 with only Alfons's game yet to finish.  There's another position I can't see through clearly, although on a superficial perusal it looks like briareus has staged a comeback over the last ten moves or so.
Title: Re: 2013 Postal Mixer
Post by clyring on Sep 15th, 2013, 7:53pm
Even if there is another 80-move game, it may take months to reach that point still. The large initial reserve and a mutual willingness to use almost all of it have meant my last remaining game has seen only 27 moves played so far. :)
Title: Re: 2013 Postal Mixer
Post by browni3141 on Sep 15th, 2013, 11:21pm
When I win the continued game that I timed-out against briareus do we modify that record, or would that be unfair?  :)
Title: Re: 2013 Postal Mixer
Post by Alfons on Sep 24th, 2013, 9:52am
This game against briareus is driving me nuts. So many times I thought I have it under control but but every time it found a way to sneak out  >:(
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Sep 25th, 2013, 2:18am
Don't worry, Alfons.  If, in the end, briareus escapes to defeat you, you will be in the company of the reigning World Champion. :-/
Title: Re: 2013 Postal Mixer
Post by chessandgo on Sep 25th, 2013, 3:47am
It took me 15 secs to make sure that it was I. What is it that would make Alfons be in my (bad) company?
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Sep 25th, 2013, 1:32pm
Having briareus totally smashed against the wall in a postal game before briareus found a way to wriggle free.  I know that ChrisB and others have had success against briareus by swarming to the point that the bot had no options but to shuffle pieces and await defeat, but in light of briareus's past Houdini moves, I was too scared to start swarming myself in the 2013 Postal Mixer until I had a substantial material advantage.
Title: Re: 2013 Postal Mixer
Post by browni3141 on Sep 26th, 2013, 12:06am

on 09/25/13 at 13:32:17, Fritzlein wrote:
Having briareus totally smashed against the wall in a postal game before briareus found a way to wriggle free.  I know that ChrisB and others have had success against briareus by swarming to the point that the bot had no options but to shuffle pieces and await defeat, but in light of briareus's past Houdini moves, I was too scared to start swarming myself in the 2013 Postal Mixer until I had a substantial material advantage.

In my opinion attacking for trap control is the best way to defeat briareus easily at any time control. Trying to win material first is just making things more difficult for yourself :P
chessandgo's game might qualify to me  as a "Houdini" effort by briareus, although if I recall correctly chessandgo missed at least one way to win solidly. I don't think briareus' comeback in Alfon's game is due to the bot's abilities, however. I don't think either game is reason to worry about swarming against briareus. One just needs to be aware of the opponent's chances and make sure he isn't presented with any decent opportunities for counterplay. Obviously this is easier said than done, but I don't think briareus should be getting the better of us here if we are careful.
The most challenging way to beat briareus is probably an unclear goal attack before a lot of material has been won. briareus is excellent at goal defense...
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Sep 26th, 2013, 12:46am

on 09/26/13 at 00:06:08, browni3141 wrote:
The most challenging way to beat briareus is probably an unclear goal attack before a lot of material has been won.

Agreed.  You do like to do things the hard way.  :)
Title: Re: 2013 Postal Mixer
Post by browni3141 on Sep 30th, 2013, 6:19pm
Here is the finished game against briareus that I timed-out and restarted:
http://arimaa.com/arimaa/gameroom/comments.cgi?gid=277321

I played this one very tamely :)
See Fritzlein, I am not always a masochist!
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Oct 2nd, 2013, 8:15pm
Hmm, bot_briareus' position against Alfons now looks overwhelming.  Assuming briareus wins, it is another new record performance.  Here is the updated table of bot Postal Mixer performance based on the game-room ratings of their opponents:

Bot .    Year Record Postal Performance
---      ---- ------ ------------------
Bomb     2005   4-6   1716
Zombie   2007   1-14  1409
Sharp    2008   1-8   1568
OpFor    2008   8-9   2060
OpFor    2009   3-7   1798
OpFor    2010   6-8   1848
Clueless 2010   9-7   2072
Briareus 2011  12-6   2167
Hippo    2012   4-4   1798
Briareus 2012  10-4   2221
Trash    2013   3-15  1431
Hippo    2013   7-11  1735
Briareus 2013  14-4   2265

Admittedly, this year's results were heavily affected by timeouts.  I believe only one of trash's games was not decided by timeout, and flipping briareus's timeout win over browni to an over-the-board loss would pull briareus's performance down to 2186.  Even so, congratulations are in order.  Well done, Ricardo!
Title: Re: 2013 Postal Mixer
Post by supersamu on Oct 5th, 2013, 7:58am
Is there going to be such a thing as a game of the Tournament?
I imagine everyone feels that their own games were very exciting.
Title: Re: 2013 Postal Mixer
Post by Hippo on Oct 5th, 2013, 4:14pm

on 10/05/13 at 07:58:36, supersamu wrote:
Is there going to be such a thing as a game of the Tournament?
I imagine everyone feels that their own games were very exciting.


I must say that Alfons bot_brieareus keeps me excited.
But your game with Boo was exciting as well.
Title: Re: 2013 Postal Mixer
Post by Hippo on Oct 9th, 2013, 3:57pm

on 10/02/13 at 20:15:56, Fritzlein wrote:
Hmm, bot_briareus' position against Alfons now looks overwhelming.  Assuming briareus wins, it is another new record performance.  Here is the updated table of bot Postal Mixer performance based on the game-room ratings of their opponents:

Bot .    Year Record Postal Performance
---      ---- ------ ------------------
Bomb     2005   4-6   1716
Zombie   2007   1-14  1409
Sharp    2008   1-8   1568
OpFor    2008   8-9   2060
OpFor    2009   3-7   1798
OpFor    2010   6-8   1848
Clueless 2010   9-7   2072
Briareus 2011  12-6   2167
Hippo    2012   4-4   1798
Briareus 2012  10-4   2221
Trash    2013   3-15  1431
Hippo    2013   7-11  1735
Briareus 2013  14-4   2265

Admittedly, this year's results were heavily affected by timeouts.  I believe only one of trash's games was not decided by timeout, and flipping briareus's timeout win over browni to an over-the-board loss would pull briareus's performance down to 2186.  Even so, congratulations are in order.  Well done, Ricardo!


So what is the bot's performance now? After the game finished?
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Oct 9th, 2013, 4:35pm
What a finish, Alfons!  I sure spoke too soon :-[, but it makes it all the more pleasurable to see the bot get beaten at a goal race.  8)

Hippo, the corrected standings are below:

Bot .    Year Record Postal Performance
---      ---- ------ ------------------
Bomb     2005   4-6   1716
Zombie   2007   1-14  1409
Sharp    2008   1-8   1568
OpFor    2008   8-9   2060
OpFor    2009   3-7   1798
OpFor    2010   6-8   1848
Clueless 2010   9-7   2072
Briareus 2011  12-6   2167
Hippo    2012   4-4   1798
Briareus 2012  10-4   2221
Trash    2013   3-15  1431
Hippo    2013   7-11  1735
Briareus 2013  13-5   2186
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Oct 9th, 2013, 4:37pm

on 10/05/13 at 07:58:36, supersamu wrote:
Is there going to be such a thing as a game of the Tournament?
I imagine everyone feels that their own games were very exciting.

Do you judge the "best" game by having the fewest errors, or the most drama, or the great move that was hardest to find, or by some other criteria?
Title: Re: 2013 Postal Mixer
Post by Hippo on Oct 10th, 2013, 3:19am

on 10/09/13 at 16:35:03, Fritzlein wrote:
What a finish, Alfons!  I sure spoke too soon :-[, but it makes it all the more pleasurable to see the bot get beaten at a goal race.  8)

Hippo, the corrected standings are below:

Bot .    Year Record Postal Performance
---      ---- ------ ------------------
Bomb     2005   4-6   1716
Zombie   2007   1-14  1409
Sharp    2008   1-8   1568
OpFor    2008   8-9   2060
OpFor    2009   3-7   1798
OpFor    2010   6-8   1848
Clueless 2010   9-7   2072
Briareus 2011  12-6   2167
Hippo    2012   4-4   1798
Briareus 2012  10-4   2221
Trash    2013   3-15  1431
Hippo    2013   7-11  1735
Briareus 2013  13-5   2186


So Brireus performance with loss against Alfons, but win against Browni is same as with loss against Browni and win against Alfons?
And if we consider both games lost for briareus?
Title: Re: 2013 Postal Mixer
Post by Fritzlein on Oct 10th, 2013, 9:04am

on 10/10/13 at 03:19:16, Hippo wrote:
So Brireus performance with loss against Alfons, but win against Browni is same as with loss against Browni and win against Alfons?

Correct.  For performance rating it matters who your opponents were and how many you won, but the distribution of your wins and losses doesn't matter.


Quote:
And if we consider both games lost for briareus?

Then briareus's performance drops to 2112.
Title: Re: 2013 Postal Mixer
Post by supersamu on Oct 11th, 2013, 10:50am

on 10/09/13 at 16:37:48, Fritzlein wrote:
Do you judge the "best" game by having the fewest errors, or the most drama, or the great move that was hardest to find, or by some other criteria?


As I read your post I thought I should have worded my proposal better. But then I realized that I never wrote something about a "best" game.
I don´t think declaring a game the "best" game is desirable.
Instead declaring a game or several games as games of the tournament based on things like "having the fewest errors, or the most drama, or the great move that was hardest to find" is something we might want to do.

There are numerous ways to look at this.

- Is a game eligible for best game of the tournament because a player dominated the position for the whole game and pulled of a good control game?

- Or maybe a game where things get traded down and there is an even endgame and a player sees a goal in 4 that involves sacrificing several pieces?

- What about a game where someone has the positional advantage, but a big material disadvantage and manages to convert that into a material lead about 15 moves later?

There are so many ways to look at this.
When I proposed this I thought about what possible criteria there are, but decided to ask if we want such a thing as a game of the tournament before thinking of ways to determine it.
Maybe one other way of looking at this is:
what game shows several features of what Arimaa is about?
Some essential features of Arimaa are, in my Opinion:
Trap Control, piece alignment, rabbit advancement and goal
race.

In my opinion the following features should not invalidate a game of being eligible for game of the tournament, but are nevertheless detrimental to its eligibility:
- Featuring too many blunders
- The game ended by timeout or resignation (debatable to some)

The following features can be considered positive:
- The players are playing at an equal level in this game
- The game features interesting positions

I have trouble to come up with anything else on this topic at the moment.
Does anyone else have some ideas?
Title: Re: 2013 Postal Mixer
Post by clyring on Oct 13th, 2013, 3:08pm

on 10/10/13 at 09:04:21, Fritzlein wrote:
For performance rating it matters who your opponents were and how many you won, but the distribution of your wins and losses doesn't matter.

It should be noted that this is a useful property of the logistic "elo" model and not something that is true in general for other models.
Title: Re: 2013 Postal Mixer
Post by Hippo on Nov 4th, 2013, 1:10pm
So the postal mixer has ended by the Alfons win over ChrisB.
I would still vote for Alfons's game against bot_ziltoid/briareus for briareus finding of counterplay leading to camel and cat for two rabbits trade and finally for Alfons's following play improving it's positional lead despite being camel down resulting in final goal.


Arimaa Forum » Powered by YaBB 1 Gold - SP 1.3.1!
YaBB © 2000-2003. All Rights Reserved.