andreas : Best known chess variant with hexagonal board is Glinski's Hexagonal Chess. It was very popular once upon a time, with 100.000 board sets selled.

There were 104,000 sold over a period of 22 years.
We are at 63,500 at 4 years 8 months.

When posting "data" the spaces are pretty much filtered out and the columns are not maintained. Would it be possible to support TABLE html tags within posts?

BIG BAD WOLF:

I wouldn't mind seeing the longest winning streaks then.

Something to consider: when you are in long "X wins match" you could have completed many games of a certain type, yet 6 months may have elapsed since a result has been "counted". Recall individual games are not rated in the matches, just one final result.

You might want to avoid dropping active people who are not playing any games except those in matches.

nobleheart: my word game idea should be relatively easy to code.

Maybe not. There are words such as "ai" and "rotl" that really do mean something in the English language (and they are not acronyms either.) You would have to parse an entire dictionary every time a player sets down a letter.

Will Swiss Tournaments be implemented at some point in time?

I don't use the cloak mode myself. One of my opponents and I banged out 15 moves in each of two games when we saw each other online yesterday. I guess as long as they are willing to play, they will "decloak" and let you see them.

I scrutinize every game played by my opponents I am facing for the first time before I ever make a move. I look for weaknesses and try to exploit them. So, for me, cloaking might be a good thing...

...then I started thinking "hmmm...maybe they will freak out if they see me pouring over game after game..."

So, I let them observe me

I suggested to Filip that Gold Rooks be able to see through the cloaks, maybe that would increase sales

"The pawn that roared"

Maybe to help with Gold Rook sales, they can be made to "see through" the cloak mode and tell what cloaked players are doing

Since we have a "cloak mode", I would like to know if there will be a "dagger mode" as well.

fungame:

If we were playing a game of checkers where jumps were not forced, and we were down to the typical minimal endgame where you had 2 kings and I had 1 king, you could not win the game.

I am not making a claim based on supposition, it is a fact.

If I don't have to jump you, when you try to swap off my last man, I just move to some other square.

If there is no way to conclude a game, then what fun is there in playing it? Would you want to play a game for a few months, only to have it drag on ad infinitum?

It would be like playing a game of baseball without walks or strikes. There is no consequence for 'doing nothing'.

Daniel Snyder: played checkers for many years not using the forced move rule

There is only one problem with that, and it is a big problem. When you swap all the way down to 2 kings against 1, if jumps are not forced, there is no way to win the game.

actually, what is interesting is that ANY game where you select a piece to move that has only 1 legal move, the "legal destination square" is selected for you.

If you make a move and the opponent has only 1
"legal move" in reply, it would make sense for that move to be made immediately.

But, that is non-trivial to implement, and could add to the overhead of the site.

Chessmaster1000: wow that is cool!

Would there be a way to show HOW MANY of a particular game type are being played at the present time? Like there are 4000 active Backgammon games, 50 checkers games, etc.

Many ideas about moderation are being discussed here:

http://brainking.com/en/ShowFellowship?fid=371

Everything from auditing moderator behavior, to a proposed new Global Moderator Hierarchy where the Moderator IDs are "non-game playing" so you do not know who they are (should eliminate bias.)

The discussions are quite tame and without so much emotion, so if you want to make a positive contribution, or just read what it going on, join up and make your thoughts known, in complete confidence within the circle.

Some Global Mods are members, and they have responded to queries posted by the memeber community.

This might be very difficult to implement, but it might be cool to have. In postal chess, there are so-called "conditional moves" that you can send. Essentially "if you go here, I go here" is sent with the move.

Maybe after you make a move, before you hit SEND, you can hit CONDITION. In so doing, you are allowed to make one move for the other side of the board, then hit a button, REPLY. At that point, it is your turn again, and you can make a move, then you hit SEND.

Your original move is sent, with no indication there is a conditional move attached. If your opponent makes the move you had specified with the condition, then your move is made instantly and automatically, and it is your opponent's turn to move again.

Just a thought.

you can choose who plays? suppose you want seeds 2,5,7, and 9?

Would it be possible to temporarily have members of your fellowship teams "sit out" for a tournament? Like, perhaps, field your top 4 players and have the others "take a rest"? :)

Also goes by the name Plunder Chess, and the website is PlunderChess.com I think.

I guess opening a file in front of a pawn is worth sacrificing material for, as you could then make successive moves at one point, and just launch it to promote.

George, why don't you mention your cool idea about "Archbishop Amazons"?

I notice many people have "logging off" or "viewing main page" as their last action. How about we see the last game they were looking at before this action?

The people who saw/experienced it may remember how it worked, and might be able to offer their perspective on the merits of the implementation.

Did anyone here every play on Playsite.com? They use to have realtime elimination tournaments that were scheduled in advance on the site, it would be quite a large draw at times.

grenv: Good idea, except that normally 1 plays 4, 2 plays 3.

It depends. In the US Chess Federation, out of a pool of 100, 1 would play 51, 2 would play 52, 3 would play 53.... and 50 would play 100.

The rational is that the relative "rating difference" among the participants would be more or less constant.

I would just like to add something to the equation here. I get pulled into a great many "3 wins" matches in the finals sections of tournaments. If you look at my profile, it looks like I have only 1 completed game of Gothic Chess for the whole year, when in fact I have some games wrapped up in the envelope of 3-wins matches that have not "counted" yet.

I would like to think that any such completed game, even though not yet "known" to the outside world, should be able to reset such a timer countdown.

I mention this because some of these 3-wins matches are very, very long.

The ratings system is not a so-called zero sum.

If you play only against each other, and split every other game, your ratings will oscillate then eventually become equal.

But along the way, the ratings functions will look like a carrier wave enveloping another signal.

You might be -17, +19, -15, +17, -13, +15 ... with an odd-even cancelation kicking in at some point too.

There are 2 numbers per opponent: rating, and rating deviation (rd). The rating number is published, the rd is part of a hidden calculation.

Your rd governs the rate of change of your rating more than the individual combat against another player.

If you lose your first 30 games against the lowest rated people on here, then win 30 games in a row against the highest rated people on here, your rating would still be near the middle of the pack.

If you won your first 30 games against the top players on here, then lost your next 30 games against the lowest rated players on here, your rating would be much higher.

In the first instance, you have a rating of 1992, with some built in "uncertainty" about how accurate it is. Your opponent has a rating of 1725 which may be a more accurate representation of his "true rating".

So, your 17 point penalty reflects two things:

1. You lost to someone "far" below you with a more certain estimate of their strength. This reinforces the notion that your "true" rating has not been properly tuned yet. So, there was a larger adjustment.

2. Their rating was a more accurate representation of their performance, so you "should have" been able to win if the 1992 rating was accurate, which the systems stipulates is not the case yet.

Your subsequent win of 19 points reflects the fact that the previous "rating correction" was too aggressive. Your 1975 rating was more accurate than your 1992 rating, based on the information known at the time of that previous event. But now your opponent was more accurately rated at 1760, and you won.

Your opponent was penalized 33 points because his gain from the last game was "too high". He lost to a more accurate rating from a lower rated opponent, so the penalty should not be as great.

Has it ever been discussed to track the "history" of an ID's name changes? I just saw games I "played" against names I really don't recognize, and I really can't remember who they may have been, "screen-name-wise".

Yahtzee is public domain now, so there would be no intellectual property dispute concerning an online implementation.

Are there any Amazons variants? How about one where you can pass through the arrows you shoot but not your opponents' arrows?

Maybe call it "semi-zons" or something

My dad has his colon operated on, having some of it removed. I told him he only has a "semicolon" now

We should each take a cluster of games with MJL, accept them, then delete the games after 1 move

Czuch Chuckers:

On a side note, I think it is mostly chess players who skew these results as there are so many variants of chess that you can make the list without knowing any other game types. This is not the same for a line game player or reversi game player as they have to be good at someother game type other than their "main" game because there just arent 11 different game variants for those games.

I saw this and some other comments about games and their "close cousins", so to speak, and I thought I would chime in on this one.

While I think there are some "portable skills" when you compare games such as Chess and Janus Chess, such as being able to play sound tactics while also have a strategical sense of how to plan, that is really where the similarity ends.

Chess and Janus Chess and Gothic Chess are really very, very different. The first game of Gothic Chess I ever played against a chess Grandmaster, Georgi Kackeshvilli (I am probably butchering the spelling of his name) I was able to win. He has a FIDE chess rating over 2700, placing him in the top 60 or 70 in the world at the time.

I would have NO CHANCE of being able to beat him in chess, even if we played something like 1000 games.

In turn, he really pounded me in the next game, so he adapted very quickly. This does not change the underlying premise: pattern recognition is pivotal in reaching the highest level of play in any game.

Despite having "done well" in Janus on here, I believe I play this game terribly. I just don't understand it. I can pull a rabbit out of my hat on occasion, but the game is "too diagonal" for my tastes, so I will never get really good at it.

I think it would be interesting to show the stats that are being mentioned as a function of the number of different games that are involved.

Who is the best when one game is being used in the sample, 2 games, 3 games, 4 games, etc.

Have a pulldown menu or something.

Some people may never play 11 different games on here, just because that is their personal preference.

Sumerian:

I think what you are saying is essentially correct.

There is a way to extend this though without requiring the 'leader' to have finished all of his/her games.

IF, at any point in the tournament, a person in the section has a lead such that:

If he loses all of his remaining games (which can be 0 in the case that he is finished) and if any other player wins all of his remaining games, and the player hypothetically losing all games left to play still has a higher point total than everyone else, then it is possible to include that player in the next round.

The easy way to do this programmatically:

1. For each player, 1 at a time, give them a "0" for all remaining games in a section. Call this "worst score possible" (wsp)
2. Add up this point total for each player.
3. For every player other than the one being considered for the wsp, given them a "1" for each remaining game. call this the "best score possible" (bsp) for the remaining players.
4. since you are computing many bsp scores for each single wsp score, if you find one bsp > wsp you can stop the bsp computation loop, and compute a wsp for another player.
5. Once you are done, if there exists one wsp that beats EVERY bsp, then that player can advance to the next round.
6. Once every section has a wsp candidate "seeded", then the next round can begin.

I guess the question becomes, do you want the next round to start as soon as possible, or have some "take a breather" time delay (like maybe a few days)?

Here is an extreme case, to make a point. Here is a tournament with 103 participants:

http://brainking.com/game/Tournaments?trg=640&tri=2803&trnst=2

As you can see, all of the next round semifinalists have been determined.

There is only ONE GAME REMAINING, and neither player can advance, as shown here:

http://brainking.com/game/ShowGame?g=300189

I think a great feature would be the starting of the next round when it is mathematically impossible for any of the active games to alter the outcome of the event.

It is easier said than done, but why hold up 18 semifinalists becaue of 1 game that does not matter?

People can abuse the message system now even when they are not running a tournament.

The chainsaw is a good tool for cutting wood. You don't stop selling them because you can hack people to pieces with one also.

BerniceC: I meant message everyone on here, not email them.

In other words, you are TD of a tournament, you type one message, hit SEND TO PARTICIPANTS or something like that, and they all get the message.

In this tournament:

http://brainking.com/game/Tournaments?trg=6703&tri=25674&trnst=0

I am occasionally getting questions for a further clarification of the rules. If there was a way to send a message to all participants at once, that would be a cool feature.

...with all the noise going on in the General Chat, a good feature would be to show all of your nicknames in a list, and the dates they were in use. That way you can see all of the nicks that person has used on the same ID.

Well, I am beating Tripel 2-0 in a 3 wins match where the winner gets a Brain Rook membership. I will not become a 'paying member', but I will become a Rook.

I would like to see a feature where you can HIDE anyone's posts and not have to look at them. Others can see the post if they have not elected to hide them. That way, you can avoid reading what certain people are posting, possibly reducing flame wars drastically.

If you look at:

Summary Page

You can see in the crosstables for the preliminary sections, the draws "filled in" with 1/2 point scores.

At the 3-wins match level of the finalists, the draw score still show "0.0" in the tally.

Two draws should equal 1.0.

If a player wins the first 2 games of a 3-wins match, then the other player cannot win the match. Does the 3rd game need to be played then? If not, is it possible to have the Match Manager not start the 3rd game?

Instead of a 3-wins match, how about a "best of" match arrangement?

You can have varying lengths according to the type of game. For example, in checkers, a longer match is usually required to determine a victor. 24 to 40 game matches are typical at the World Championship level, but this is a bit much for online play. Perhaps 13 games would suffice. Whoever has the most points wins the match.

If it ever reaches a point where it is mathematically impossible for the other side to win, then it can be stopped.

For exmaple, if there are 4 wins and 6 draws in a 13 game match, the last 3 games need not be played since if the winner lost all 3 then he would still win.

This is actually what they do in checkers. Since Alex Moiseyev was up 8-2 over Ron King in the last World Checkers Championship, that match was called off early.

Chess would need fewer games, maybe 7.

Gothic Chess would probably only need 5.

Again, the same rules apply. If a mathematical certainty has been reached, like 3-0 in Gothic or 4-0 in chess, the match can be stopped.

I think in drawish games, you should rate the result 1-0 (like checkers). In other games, you should rate each game as they complete.

Just an opinion, but I think the reasoning makes sense.

I think the whole point of the 3 wins match was for games like checkers. If I win 3 and draw 15 against a player on here, my rating would go way down despite having won by a large margin. But, with the 3 wins format, it is like the draws never happened, and the match is rated in a way to keep the ratings balanced.

A 3 wins match in Gothic Chess, in my opinion, is a great injustice. To win 3 games against one opponent is not easy, and to treat it as just one win is unfair.

Also, opponents in a checkers 3-win match have no way to conceed a draw. After 100 games, all draws, can't we just end it?

This format needs rethinking. It needs at least some sort of "stopping condition". Say after 15 games, the one with the most points wins.

Anything but perpetual play.

Is the 3 wins match unlimited in terms of the number of games? I am about to be in a 3 wins match for checkers, which can go on forever. Many of the strongest players won by scores of 1-0 with 31 draws, or 2-1 with 48 draws. With days to make a move, the odds are a player could go on forever without winning 3 games.

I think the reason why people elect to play in the slower events is that the timer starts ticking as soon as your opponent moves.

So, if you have any type of vacation planned in the months ahead, you have to make sure you have enough time to make each move as if you were actually on your vacation or away from your computer, "just in case" the game spills into your vacation time.

With the N moves in M days approach, you would not have to worry about this at all. Look at the 10 moves/30 days time control. The game starts, you go on vacation for 3 weeks, using up 21 days, and you still have 9 days to make 10 moves when you return. You will not time out.

What would be your alternative here and now? Play 21 days per move.

A big difference.