Wil: Thanks for the link. Lolol - it's the one I gave in my reply to you earlier. ;-)) It's good but it's not quite as rigourous as I'd like - it's geared towards explaining how to play rather than enumerating and defining the rules (something that would tend to frighten beginners away, lol).

That bear-'em-all-off-first move breaks the rule about moving one man for each of the dice values and moving by the exact number of pips on the dice except when bearing off, when the pips may be more than required. Allowing one "illegal" move in a special circumstance where it makes no material difference to the outcome of the game doesn't mean that any illegal move is okay at any time.

Thanks Pedro, I've seen that one but it's more pretty pictures than careful language. I'm after something like their tournament rules but for the game itself.

I can't find a link to the official rules (something with lots of numbers and paragraphs and sub-clauses, lol). Does anyone have one?

Wil: For the sake of expediency because it doesn't affect the outcome of the game. Or because the last man going off the board ends the game regardless of whether there are dice values left or not.

I skipped ChessM's point about bearing off all pieces on the first roll because I didn't understand what bearing it had on anything. Care to explain?

Chessmaster1000: Yep, you're right about that particular conversation but you wouldn't have it with me. ;-D

With me it would go:

A1. Well, firstly, I wonder what you mean by "move"? Is that a single man moved using one of the dice values or all the moves that the user sees fit to make?

If it's the single-move-at-a-time case then the 2-off would be legal and it would also end the game.

If not then the 2-off would still be legal and it, too, would end the game. The other dice value would be deemed impossible to play, which is fine.

A2. Of course!

A3. Nope. It's a valid move with one dice value legitimately discarded because it's impossible to play once the piece is off the board.

Or:

A1. Well, that's not quite how I'd do it. Mine would be:

1 - Make a list of all the possible and legal combinations of moves.
2 - Get the User's move-combination.
3 - If the User's move-combination does not match one of those listed,
....... - Inform the User.
....... - Go to 2.
4 - Make the move.
5 - If the move ends the game ...

And step 1 would include moves that use a single dice to remove the last man from the board.

If you like, I'm saying that the MDU rule has a special case for the last man. Why? Because the outcome is exactly the same except that the rigid application of the rule requires the user to make more effort to finish the game. Allowing the single move is expedient without loss of fairness.

WhiteTower: Saying it's the way it works doesn't mean there's no issue but it can certainly be used to state that your exploration stop there. (Such action is just one of the options so no criticism is intended.)

Lolol. I reckon maybe he does! ;-)

WhiteTower: "Let's just accept ..." Lol. Why? I'm having fun debating with Abigail and my guess is that she's enjoying it too. So are some spectators. Others are bored by it and some couldn't care less. That's just how it goes.

"Let's just accept ... Fencer's rules" I wish that people would, lolol. Perhaps you've missed what this debate is about. I'm saying that, for various reasons (see below), Fencer's rules include MDU, even if a bug prevents that rule from being enforced. So accepting Fencer's rules means accepting the MDU rule. But some folks disagree with that ....

If you play against me and do what BK allows you - ie. exploit the bug to gain an unfair advantage - when it's against Fencer's rules (and I have provided evidence of that) ... what illusions are you talking about? ;-)


AbigailII: The fact that the game doesn't enforce MDU because of a bug doesn't invalidate that rule. True, MDU isn't written into the rules explicitly, but, as I have explained, it is implied by the phrase in question. Simply saying that you "doubt that" isn't providing any additional information and so I'm keen to see you provide some inkling as to what illegal moves that phrase could be referring. This is a key point in my current argument and this is the third request for you to address it but you seem unable to provide an interpretion. Without that we cannot continue this discussion (much to the relief of WhiteTower and any one else who groans with the appearance of each message, lolol)

Note that the pass situation has nothing to do with this debate as it refers to the impossibility of making a move as opposed to there being alternate possible moves, one or more of which are illegal. ;-)

AbigailII: "Then you tell me what's meant with the illegal moves in the rule that states the swap dice link won't work.". I thought that's what my post did. Was it unclear? As far as I'm concerned the swap link implies the MDU situation and no other. That's why I asked you to suggest a different kettle of fish.

Chessmaster1000: "The rule is simple without using the maximize word anyway:". Yep, but we've talked about this rule so many times that an abbreviation is required. Try saying it with just three words or less. ;-p

Lolol re "Serious about the tricky one". Pgt is right, of course. Having legally taken the last piece off with the higher dice it is then impossible to use the other value. Impossible isn't illegal - so you have a choice with that case.

Interestingly enough (and annoying, too) is that at VogClub you arerequired to use the smaller dice first before bearing off the last man. Too strong an implementation of the MDU rule, lolol.

Wil: Indeed; we have no disagreement. The maximise dice usage rule is about using the dice values, not the pips (which would be the nonexistent maximise pip usage rule).

I and others have already defined the rule in previous discussions (which you most likely haven't seen) which is why I use the shorthand rather than spell the rule out again and again.

For anyone who wants to know, here's a full version (see section Movement of the Checkers, part 4).

Pedro: Lol. Return quibble: I've declared that Fencer has defined it as a bug. ;-)

Pedro Martínez: the English rules should take precedence with regard to the other translations: Other translation being French, German, etc, but not presumably Czech. It was just a minor point. Why would Fencer write something in English and then translate it back to his own language? Not knowing the guy, who knows? lol. But the reason I wondered is that you've told us that the Czech version doesn't have the word legal while the English rules do, so it was a question of whether the word was lost going from English to Czech or added when going frmo czech to English, unless they were done independantly. Again, just a minor curiousity.

Quote: "I think that Fencer said somewhere that it was a bug" Lol. I don't know how many times I've declared/stated/affirmed that it is a bug, given the bug-tracker link and more ... I agree that having a bug means that it's up to the players what to do - it has to be, for the possibility to exploit it exists - but this discussion is more about what they should do; whether it is legal or not. I reckon that the fact that it's a bug means that the move in question is illegal. But I've said all this before so Ende.

Hrqls: That's a separate issue and certainly a reportable bug.

AbigailII: Er, that's an example of the kind of illegal move that we are already been discussing, ie. those which go against the maximise-dice-usage rule. I'm after a new class of illegal move that the English rules (lol, Pedro) could be implying, because without a new class the rules are referencing the MDU rule (even if it's not implemented correctly).

Given that it's an example of the MDU rule, isn't you saying "(or at least, according to the rules, you shouldn't be able to swap the dice)" an agreement that these moves are against the rules? I'm quite happy if so, but I'd just like to clarify that. ;-))

Pedro: That's interesting. I wonder which set of rules would take precedence. I believe the English ones should be primary with regard to the other translations but Czech is Fencer's language so I wonder whether the English set is a translation of the Czech, vice versa or are they independant? Lol.

Let me think ..... Today I'll play by the, mmmmmm, Czech rules!

Wil: Aye, to follow the rule doesn't require Sharon's input at all. If fairness is your priority then taking advantage of the bug certain would. Respect to you, as I know that move loses you the last wee chance in that tourney. Good luck in the next one. :-)


Abigail: 1) It's funny that you brought up the dice swapping thing:
... click on "Swap dice" .. [except when] ..
    a) Both dice show the same value.
    b) The player could not make a legal move with the second die..
That's because I almost included it as a point in my own argument (too much effort at the time). ;-)

When can you not swap dice? One time is when one number cannot be used (eg. when trapped by a prime, 1-6 is rolled and there's room to move up with the one, or with two men on the bar and one's re-entry is blocked) but this is not a case of an illegal move, this is an impossible move - it's a separate class from the two mentioned in the rules (and shouldn't have been omitted).

The only other cases that I can think of are those that are the subject of this discussion - the illegal ones. Fencer has included them in the rules, albeit in a generalised way, for what does The player could not make a legal move with the second die. mean? It means that the BrainKing rules acknowledge the existence of illegal moves. They aren't described explicitly, nor is the maximise-dice-usage rule stated, but the mention of legal moves implies illegal moves, so what moves are they? Can you describe a class of illegal moves that is different to what we're discussing?

The rule is implied rather than explicit, but it is there, at least as I interpret it. The fact that the implementation doesn't detect this class of move and act according to the rules is a bug. It's not a permission to make the move.

2) You said you've not played backgammon on a physical board. Let me echo your words to Hrqls: Are you trying to make a point? If so, what is it? ;-)

3) It still stands with me that if someone knows the standard rules of a game yet prefers to use a site's E&O (Errors and Omissions) as loopholes then they are showing a poor attitude.

Wil: I abstain from this vote. It's yours and Sharon's game; it should be your decision. (Is she aware of this debate?)

If this were a vote for us all to play according to the standard rule then of course I would vote for the rule. As Wil says, if he were playing on a physical board he would do the right thing. To me a player who needs a web page to tell them that ... pah!

It would be different if the web page were to explicitly state that either or both dice may be used (though I expect there'd be much concern among the players), but someone who uses an omission as a loophole is skimming close to cheating as far as I'm concerned.

AbigailII: That's fine. Remind me not to play you until the bug's fixed. ;-p

ps. "Pro backgammon" is on its way. The doubling cube has been mentioned. Presumably it includes gammons and backgammons. Hopefully it includes this bug.

AbigailII: I use the following logic: It's acknowledged as a bug. If it were within the rules then it wouldn't be a bug. Therefore it's illegal.

Wil: It's against the rules and not being forced to move using both dice values is a known bug. Until implemented properly it's either a question for your own conscience and sportingmindedness (to play by the rule), or it's a decision that both players should agree to (to take advantage of the bug). both of those are fair ways to deal with it. I know it's a tourney loser but I'd make the correct move 24/20/15 (but then, I was on the wrong side of this same situation and wasn't pleased that the guy took the illegal route without hesitation. He's just been banned for cheating - multiple nicks and rate fixing - but that's a different story, lol).

You could put the question to Sharon - ask her is she's willing to take the gamble. Your chance of winning would be about 1/40. ;-)

If you want to see what thoughts there were last time, stick bug into the search box on this page. It's the same rule but in hypergammon.

Is there anyone who knows hypergammon well enough to explain why they choose their moves and would enjoy doing so?

Pedro: Thanks. :-)

AbigailII: While this site has a small clientele the overlap of IP addresses is reasonably unlikely to occur (eg. 600 UK registrations, a mere 70 for Italy). Agreed, when we have hordes of people all from the same source then IP address distinctiveness becomes more tenuous.

IP address as an indicator of personality? Lol. I usually use behaviour for that ... sometimes foot size ... maybe house number . . . but never IP address. ;-p

Andre Faria: That's not a good method. There's a new guy at #2 - el diabolique, rating 2587 - who's got there by winning his first 4 games. This was in a tournament against 4 different opponents so fair play is to be assumed. There are three other players in the top 20 who have had early and legitimate success. Such gains are to be expected from any expert player who joins BrainKing.

Chessmaster1000: Aye, the IP address can be used to help in these matters. At one forum site that I enjoyed for a while there was a self-appointed "IP Police" who took it upon herself to detect multiple nicks by collecting posters' IP addresses. It's fine when the user has a static IP address but there are many ISPs which assign a new IP address at the start of each session. (Very common with dialup and is still pretty common with broadband.) Even so, there will be a commonality. For instance my IP address is never the same but does always start with the same two numbers.

ArtfulDodger: The maximise-dice-usage rule hasn't been implemented (but this is acknowledged to be a bug).

In our game wayney was able to move a single man with a single dice value which allowed him to stack all three men safely on one point. Further movement off the stack, using the other dice value, was blocked by my men. He should have been required to use both dice values which would have meant splitting the two that were already stacked, leaving me a spread of blots to hit. Subsequent dice rolls showed that I would have hit two of those blots but as a result of his illegal move he was able to make a game winning block on his ace point.

The rule is also part of standard backgammon and is probably needs to be addressed there as well. I expect that it's on Fencer's plate as he says on the Features requests board that "Pro backgammon (with doubling cube)" is in development.

Chessmaster1000: The simple pattern seems to be a pair who play together with a very unbalanced win/lose ratio brought about by a suspicious amount of resigning and with one or both having high ratings. As more nicks are brought into the scam, detection would become increasingly difficult but a significant level of resignations would still be a good indicator. It would ease the server burden if checking were restricted to the top X% of players.

ArtfulDodger: I presume you mean "wayney" rather than "Wayne". How are they no longer here (password changed, account locked out?) Any idea what happens to the games that they're playing?

The reason I ask is that I'm playing against him at the moment ... The hypergammon game that he brought to this board about him having taken advantage of a bug to make an illegal move, and me not appreciating it .... the game that I wanted to call a draw but he refused .... the game that, as a result of a conversation in which he made it clear that sportsmanship was not a priority, I'm playing in protest one move at a time but only towards the end of the timeout period.

ArtfulDodger: Who? When? Sounds like a rating fixer.

What's a blitz?

This is a devastating attack, often starting with a double against split backrunners, in which the table gets rapidly closed. In the worst case (for the victim) they get to make one move at the start of the game and then have to wait until their opponenent is bearing off before they get another go.
___________________

Hrqls: I'd have to see the board position to know what I'd have done in your game but generally I wouldn't offer a double if I thought the game were balanced. One of the considerations in offering a double is how you'd feel if it were taken. If you'd groan and say "Oh, no!" then it's not perhaps a good time to double.

But if your "hoping he would decline" is based on knowing his bravery and style, and has more expectation than vague hope, then that changes things; it can make it a canny move to double even when the percentages are less than optimal.

Walter: In the voice of Mr T: Danofool !! ;-)

Thanks for a full and well considered response. It was very persuasive. And then I checked the status of my bug report. Tulip reported the same bug back in March with regard to this game. Her opponent's next roll was a 6-5 and he should have had to break his block but got away with moving a different man using just one of the dice values. Fencer marked that bug as open with a priority of 3 and my bug as a duplicate. This tells me that Fencer recognises the maximise dice usage rule and wishes, when he has time, to implement it correctly. Moves made as a result of the bug would therefore be illegal under the BrainKing rules - else it wouldn't be classed as a bug.

That doesn't mean that we are necessarily obliged to enforce the rule ourselves. To me it's an opportunity to be sporting. But if we were playing for money, or I didn't like my opponent, then I'd be more likely to shove sportsmanship to the side and take the advantage.

grenv: Hey, good buddy, are you and I actually agreeing to something here?! lolol. My opinion revises upwards. ;-) (I assume that by ".. needs all the help they can get" you mean those who don't know the rules well enough need help in winning the game rather than all the psychological help they can get - which is what it sounded like at first, lol).

Hrqls, skipinnz: I'm torn between asking for a draw, knowing I'll be pissed off with losing, and playing on to satisfy my curiosity. I wish it could be made unrated but unfortunately there's been no response from anyone with the ability to make changes to the game. Big Bad Wolf, are you one of those persons?

WhiteTower: Pawn Power Rules!! ;-)
You're correct in saying that .. there is no rule that says "when a game implementation is faulty, try to stick to the rules of real world play". However, how many gentlemen's agreements are? By their nature they tend to be implicit and "understood". A rule that exists in the mind is still a rule, even if it's unenforceable and for adoption only by those who prefer to live by it.

Hrqls: Thank you, too, for a considered answer. I like your approach to refereeing and it fits this situation. I'm not here to find out what to do. (As you say, that's up to Wayne and myself). For me, this is more about bringing a potentially interesting topic to the board.

Lol. Wayne has chosen to bring this to the Board of Backgammoners. That wasn't particularly my wish but, as the hand has been played, let me lay out all the information that we have at present:


This is the position that Wayne was in when he rolled a 4-2.

He moved 11/7 as he said, stacking all the men in a temporary position of safety. This is an illegal move under the accepted rules of hypergammon.

Wayne knew this because his message to me was:
look carefully at my last move.
because of the flaw in Fencer's programming, I only had to move the 4. Normally I would have had been forced to move both 4 and 2
Naughty Fencer



My reply:
Hmmm. naughty Fencer indeed. It's a shame that you pressed the [Move] button before doing the correct move. What move would you have done if you'd used both the 4 and the 2?



Then Wayne:
I would have had to move 7-3 with the 4 and then 11-9 or 3-1 with the 2 which obviously would have opened me up. Probably 3-1 if I didnt know your dice roll



Then me:
Aye, I'd have done 7/3/1* as well, I reckon. If nothing hit it then it would be safe in the corner with a good chance of that 7-point man covering it is he survives.

This bug makes for a very interesting situation. The problem is that you knowingly made an illegal move. Now some would say that that was cheating but I don't see it that way. A proper cheat would have kept quiet and hoped for it not to be noticed. You, having very clearly pointed it out, are not a cheat.

What you did do was take advantage of a bug that you discovered. You got to make a move that could turn out advantageous (depending on what happens when you split them next go) but, if made properly, and now that we know my dice, would have seen your man get sent to the bar with the others possibly to follow.

More importantly, at least to me, is that you missed a wonderful opportunity to score a very good sportsmanship point.

If I'd been in your situation I'd have made the correct move and told you about the bug and how I'd dealt with it. That way I may have lost the game but I would certainly stick in your mind as a fair-minded player and a good sport. That would have had more value to me than the result because getting the opportunity to win happens every game but getting to show that you're a good sport doesn't. (Some people would see it as spelling chump and loser but I don't think you're one of those.)

The situation leaves us with the question of how to proceed. As I see it there are several possibilities:

6) You, having taking advantage of the bug, resign. I say it merely because it's a possibility. I'm definitely not in favour of this solution.

5) We play on. If I lose, I will feel that I lost unfairly. If I win then I'll wonder whther I should have lost. Either way it won't be the game that we should have played.

4) We make the game a draw. It gets added to our history but our ratings remain the same.

3) We ask the powers-that-be to make the game unrated and play on. This would turn the game into an interesting but painless what-if.

2) We ask that the game be deleted. This is the 'cleanest' solution.

1) We ask that the game be put into the state it would been had you made the move that you would have done (hence me asking you what that would have been). I'm in favour of this one because we get to keep the game after an interesting interruption.

I'm in favour of options 1, 2 or 3.
What thinkest thou, Wayne?
:-)



And me again:
Lol. I've just thought of another possibility.

4a) We report the bug and Fencer declares it Not a bug. We play on.

That's acceptable to me too, for this situation, but not for the game as a whole.
;-)


And Wayne:
actually this has been mentioned so many times in the past and Fencer's opinion has always been that this is ok to do in games on BK. Remember too we are playing HIS hypergammon and not backgammon. I would NEVER have done this in a game of "backgammon" however, this is Fencer's Hyper Gammon and hence it is not a bug, only a nuance


And Wayne:
I would have said option 1 if I did not know your future dice roll.
If we can have option 1 with the future still to be undetermined, that would suit me.
Feel free to ask Fencer to do this and you have my ok on that



Unfortunately I seem to have lost a bit of 1) when I was editing for it includes me keeping the dice roll that was given me after Wayne's move. (It's a 3-1 which would have hit the blot that he'd have left on the ace point.)

Wayne wants version 1a) where I lose that dice roll. ;-)



So, ladies and gentlemen of the board, these are the facts, opinions and wishes of the two players. What are your own feelings about these 8 options?

WhiteTower: Too long since I played chess so I didn't anticipate that one. Lolol. Nice one, very true. ;-)

WhiteTower: It seems there's no way to persuade you to accept draws and no way that we (of the Anti-NDPA) can accept having no draws.

So let's call it a draw shall we?

Chessmaster1000: Lol. 14:04! How localtime() of me! ;-D
I'm glad that the initial impression of No draws! was incorrect. And of course I fully accept your preference not to involve yourself in draws. Ha-hah, an opportunity to tease! Instead of doubling before I bear off the very last piece (why do some people do that?) - I'll off you a draw! hee hee ;-))

nb. That title? Corny British humour, can't help it.

Chessmaster1000: RYTFBE! ?? I gotta know!

And are you going to answer my previous question? (24. June 2005, 14:04:21)

Chessmaster1000: Lol. Then there's that other variation of the game - played with a cat in a box. If, on your turn, you look into the box and the cat is dead .... ;-))

But back to the main point: WhiteTower is obviously putting forward that no draws is the best way for him and that being fair to other players isn't an idea that he'd entertain (which is reasonable enough as being nice can't always be high on people's list of priorities), but are you saying, George, that everyone should be deprived of the ability to offer a draw?

CM1000, WhiteTower: It's true that it's the environment that allows draws and for some people it's a useful option on occasion. Calling it ridiculous and an abomination? Wow! Is 'despise' a good word here? It's obviously something you feel very strongly about? Enough, it seems, that you would deprive those who might enjoy that benefit.

danosham: Still hiding my posts, eh, chum? Would someone tell him he's dribbling again? Or if he wants to waste "air space" with personal remarks (like I'm doing, he he) then the word he's after is "imagination".

Agreeing a draw is rather the obvious answer, especially with that link that says Offer draw, lol. Admittedly few would realise it but at Vog there is no option to draw; it's win, lose or resign. As I was unable to see the matches referred to, I was unable to determine haow they were drawn and was wondering whether I was unaware of some quirk of the game that that made a draw possible through play.

WhiteTower: One possibility is that the two players are more in it for the sociability, and the friendship outweighs the competitive aspect. In the holiday case, or any similar, they might agree a draw because a timeout win would be taking an unfair advantage.

Another, perhaps, is if there is a strong competitive aspect and the draw is agreeable because they want their matches to reflect only their playing and not spurious wins. In the latter case the overall rating would matter less than the match history between them.

Then I imagine there's a whole bunch of players who don't really care about statistics or ratings. For them it might be as simple as "Draw?" "Okay, why not."

One reason for drawing that I have: I occasionally give backgammon lessons at Vog and when the main points of the lesson have been given, or my student's reached the brain-full-need-to-digest stage, lol, then I will resign the game or match. (It's just a politeness and doesn't affect my rating because we play unrated). If there were a draw option then I could use that.

Putting Abigail's explanation of why a win and a loss is not equal a different way (for those who see 0 q, p r s and find their eyes glazing over and their mind going blank, lol): The points per game is based on the difference between the two players' ratings. After the first game that distance will have changed and thus the second game will be worth a different amount.

The backgammon formula gives a match a maximum value equal to the square root of the match length. Whether that's the "correct" function to use or not, it makes sense that longer matches earn more beans; more effort/risk, more reward/loss.

I've often been puzzled by this one. I've seen it in player's profiles at Vog but not been able to retrieve any match from the archive. How can you have a draw in backgammon?

AbigailII: That's filled a gap. I'm not a chess player and only read as much of that Chess Rating link's info as needed to work out how my provisional ratings were being generated, and to note how much more complicated their scheme is than that in backgammon! Now that you mention matches I've reread that page and seen where match length comes into it. Thanks Abigail :-)

The chess formula is based on single games where skill is the only factor. A player deemed better than another is expected to win by skill alone and the gains from winning are meagre and the losses from losing are punitive for the better player when the difference between them are large.

In the chess formula, a rating difference of 400 points favours the expert who is expected to win 9/10ths of their games against the average player. In the backgammon formula, the effect of luck is such that experts (500 higher than average) are expected to lose in the region of a third(!) of games against an average player. The losses and gains are much less per match to account for this luck effect.

In chess I believe you play only single games and each game is worth 1 or 1/2 a point whereas in backgammon there are matches worth multiple points. Though the expert backgammon player is expected to win only 2/3s of their single games against an expert, in an 7-point match that goes up to around 80%. So an expert is expected to win a decent length match but the chances of the beginner's lucky win are by no means negligible.

Chess maxes out at 2700 or something, with backgammon 2200 is unusual.

More info:
The US chess formula

Rating calculator for backgammon
Complicated looking but also informative.

Press the [Newbie] button (it uses 1600, not 1500) and the [500] button and look at the percentages in the first table to see that the expert, P2, should only win 64% of single games but 82.1% of 7-point matches and 90.3% of the 25-pointers.

alanback: FIBS uses a backgammon formula which is used on many sites. Like the chess formula that is used here, it encapsulates the entire playing history. The 400 figure that you're remembering is used when a new player is establishing their rating. For the first 400 experience points the amount gained or lost by a match is multiplied by a number proportional to how many of the 400 points are left. The multiple is 5 at the start and 1 by the time the player has reached 400.

I also don't care to see the same 6/0/0 at the top of the rankings table. Perhaps the cocktail has lost his bottle? I like pgt's suggestion of using a limited history but it might be expensive to administer. A reasonably easy to write method would be that the ratings are recalculated every day for every player (and presumably for every game type). I don't know how much server time that would take but it would certainly be a growing amount as the site gains in popularity. Doing it monthly would be a reasonable compromise; a different set of players could be done on each day of the month.

Plies: These are single moves (replies). Gnubg was written by a computer scientist and it starts counting at zero. [rolls eyes + shrug]. So at 0-ply it is considering all the moves that it can make with each of the 21 possible rolls. 1-ply is the player's responses and 2-ply would be GnuBg's replies to those, etc. However a ply isn't quite the same as thinking ahead in chess or as we would think ahead.

You may remember me saying that the neural nets work by amassing huge amounts of statistical data. That data allows it to say with some degree of accuracy that getting to a particular board position gives a certain winning chance. In that sense it's actually "looking ahead" from that position to the end of the game. This evaluation isn't perfect, however, because perfection requires the right values for all the possible board positions - and that's just too much. The reason that neural nets are used is because they are the best mechanism, so far, of making good approximations for data of this nature. Like us they can look at a pattern and say "hmmm, that reminds me of something very similar, I'll use that as a guideline" except that they are geared to look specifically at backgammon patterns, and can do so with great accuracy.

The way plies work is that they take the board further towards the end of the game where these estimations are (generally) more accurate. It's a bit like running to the top of the next hill and the next to see what's out on the horizon.

Ideally the program would always work at 4-ply or better and calculate every possible roll and every possible move at each level. There are 21 possible rolls and anything from zero to umpteen moves (me's no computer, lol) at each ply. This degree of "branching" is much more than in chess and the reason why chess programs can look ahead further; the processing required in backgammon, even at 2-ply, is huge. In order to cut down on the processing and maximise looking ahead, the bg programs utilise a filtering system. The initial 21 rolls are always considered in full. (This is 0-ply). The worst moves are discarded and the remainder examined for responses to the next 21 possibles rolls. (1-ply) The top moves are kept and the next ply examined, and so on. (This, for those who have recently acquired GnuBg is what the filter settings refer to).

At each level, and for each roll and possible move, the board is evaluated. The board evaluation isn't done in a dumb sense of just saying how many pieces are there on each point. This will only be possible when a database can be constructed which holds every position in backgammon (about when "Beam me up, Scotty" is possible). Instead the program does what we do - it consider what elements are present: how many blots and points made in each home table, how many points made in the outer field, is there still a midpoint, how many spares are there on the points, how many builders are there and where, how many runners, what's attacking what, what is the balance is across the board, is there still contact, is there a prime, a broken prime, a closed table, etc, etc, etc; whatever the designers can think of. These elements are what the neural net considers when it's looking for patterns (and partly what makes the differences between the programs). Then, the statistical weightings that it has generated from the thousands of games that it's played against itself say that a given set of positional elements (or, more likely, a set with a given (and high) degree of similarity) has been found to produce a win in such and such a percentage of the games that were played from that position.

Because the winning chances for these positional sets are determined from self-play, you can imagine that situations that turn up again and again will be more accurate. This makes sense as it is the same for us, too. The "degree of similarity" of the set of position elements will improve, approaching an exact match, and the number of games that have been played from that position will be higher too. The quality of the bg programs is still high in the lesser known positions, however, simply because the programs get to discover and play through a lot more of them than we do. A well designed bg program will seek to ensure that the bg "state space" is explored adequately.

Using self-play has an interesting aspect. The evaluation of any position is based on the premise that the opponent is as good as the program. The moves made will thus be on the assumption that the responses will be "perfect" and the program will do its looking ahead amongst the best moves for each side. There is an occasional advantage, then, in playing the unexpected dodgy move because you will be leading the program along a game path that it might not have considered (having filtered out that move and path as being too poor). But making poor moves in order to fool the program may lose more than is gained, simply because they are poor moves. [Ignore this paragraph if it comes across as confusing. ;-) Hey, ignore the whole article! ;-D Lol]


Hopefully you can see that at the furthest extreme it's not even necessary for the programmer to know how to play backgammon! They can simply encode every possible board position and assign it the results of playing every possible game. Hey presto - the perfect robot player. This has in fact been done for hypergammon and for the ending positions in backgammon (the bearing off stage). The robots can play these absolutely perfectly with no consideration required other than looking up an exact board position. The next stage will be to encode every non-contact position (all my pieces have passed all yours, let's race) but that's still too big a number of positions to calculate and store.

The current situation is that a neural network can evaluate game situations by recognising the mix of positional elements. The programmer can easily code for these elements without being too good a player (although, in practice, the advice of top players has been readily utilised). The programmer isn't telling the robot how to play, however; he's telling it what to look for when considering how to recognise game situations. And it's the statistics of how many wins were produced from each game position that was met in the course of self-play that tells it what to play. The bg programs can only teach by saying "here's my list of moves"; they still don't really know much about how to play, lol.

Hope that was useful. :-)

Walter, CM100: First off, sorry for missing that bit about the game properties, and thanks for answering it George. (And yes, I analysed it as a single-point match.)

Much of my learning with GnuBg has been of the form: do a move, get told off about it and then, in the absence of any verbal explanation from the program, rationalise the moves that it says are best. Often this is helped by the fact that there will be part of the move which is common in all the top moves, eg getting a backrunner moving or making a point.

That move 6. The best move given by GnuBg was to hit on the 11 as I said. The second was to bounce off the bar to 16 and start getting home. Your move was deemed 5th out of the 6 that were possible. Guessing to the utmost I think there are several factors. Your move hit on the 2-point in your table which is deep. GnuBg doesn't usually care to go deep. And the hit gave no particular advantage given that there were 5 points open. It even thought that simply moving your blot from 10 into 4 would have been better. More importantly, I think, it wasn't comfortable with you having 5 men in my home table. You had anchors on 4 and 5 so one of these is superfluous. the 5-anchor is better, of course, but hitting me on 16 would have started you on the way home and evened the pipcount by 14 reducing my lead to only 10 pips.

Damn right with "Luck beats skill". It takes a lot of skill to overcome a little bad luck and a little luck to beat a lot of skill. ;-) That's very apparent from having watched so many tournaments at VogClub (they're over in a couple of hours). The top players win more often, of course, but they frequently go out to some of the weakest players.

The Backgammon Rating Formula as used by most of the bg sites calculates that top players (2100-2200; the maximum is lower than in the chess system) playing an average player (1500) will actually lose about a third of the individual games played. That indicates how much luck the formula reckons to be in the game.

I think we'd both agree that you won that particular game due to luck. All those hits and me dancing for half the game while you romped 5 men back from my home table without a care in the world?!! Lolol. I was looking forward to a good jousting match between our respective knights but your knights snuck home in the dark of the night while my King was getting drunk.

As for who's better? Time will tell, my friend. You have the lead so far. ;-)

But does this machine know what it's talking about? Well there are certain games plans/styles/situations where it is less accurate than others - mainly because they are less common. Back games and near back games such as ours may well be in that category; certainly it was true in the past, but the databases improve with every release. I don't know enough about that area to state much.

They do play each other. There's a program vs program tournament held every year but attendance is dropping as it's expensive to enter and the existence of GnuBg as a world-class and free program means that revenue has dropped for the professionals.

Walter Montego: The first was move 6 and it was hitting the wrong blot. Should have been hitting on 11 from your anchor on my 5-point. I guess you didn't want to break that anchor?

The second was move 19. Again it looks like you didn't want to bust that anchor. You came off the bar and sent the man out missing my two blots. He got tapped on the head for being nice! ;-)

Walter Montego: Do you really want me to say here? lol. The program graded your play as "Beginner". It is a very harsh judge, though, and all of the top players whose games I've analysed get "Awful!"s and "Beginner"s as well as the "Expert"s and "World Class"s. ("Top players" doesn't mean world-circuit top players, but those at the sites at which I play).

I've made my GnuBg stricter on evaluating individual moves than the standard version (so, for instance a fairly good move on the standard is a Doubtful move on mine, a Doubtful on standard is a Bad move on mine). On that basis you got 6 Doubtfuls, 2 Bads and 1 Very Bad. (Standard settings: 2 D, 2 B and 1 VB)

The two Very Bads were failures to hit blots. The two Bads were both doubles (a 4-4 and a 5-5). It's a surprisingly common error. Many of the games that I've analysed show good play apart from a failure to take the best advantage of doubles.

The bad luck that Walter is referring to is me being hit and sent to the Bar 11 times. And with that last hit I was on the Bar for 13 straight rolls despite the 4-point being open the whole time. Odds of 10,000 to 1 or something. Walter was very hospitable though, and I got very drunk for free in his bar. ;-)
Walter vs playBunny

All that being hit forced my own play somewhat and I made 1 Doubtful move and 1 Bad one. Overall I was judged as "Advanced" (which is the category below "Expert").

Revenge is not happening yet, either, as that dice guy is still working overtime creating doubles for Walter, lolol. He won our last Best of Three and is looking to take this next set, too. [I say that hoping to provoke Fate into doing something before it's too late. ;-)]

Hrqls: There could be no disputing such a stringent delineation. Yet I see no problem with analysing a game that's finished. I'm wondering, then, what advantage there is in denying yourself the analysis of a BainKing match. In fact I've done just that with a game against Walter. It brought me into a game situation territory that I only rarely get into and, to me, it was an opportunity not to be wasted.

If your opponent agrees to the use of a program then it would be okay. "Cheating" should be defined as the (surreptitious) use of such programs with intent to create an unfair advantage. I can't see many players agreeing to the use of programs.

I made a post on the brainking.com board (Re: Backgammon Programs, 8. June 2005, 01:39:07) which included thoughts about cheating with a bg program. :
----------------------
The opportunity to cheat [with a bg program] is immediately obvious - you only need enter the current position and dice roll and ask for the moves. A successful cheat would not play an entire game according to what the machine said but would only use it to advise in tricky situations. In other situations it would be possible to pick the 2nd or 3rd moves, even 7th, etc, if they were not desperately worse than the best. This makes it somewhat difficult to detect a cheat.

In practice, however, if you examine the games of the top players at most sites, you will find that even the best will make bad moves and even huge blunders - as judged by the program. These programs are very exacting judges. The player will sometimes be able to argue why their move is good but more often it will be recognised that the move was indeed a poor one. This makes it somewhat easier, then, to detect a clumsy cheat. [That's in general, though. For Fencer to detect backgammon programs would be impossible as he would have to analyse everything. It takes several minutes per game and that's computer power which is not available - nor, if it were, would it be very productive use of his servers.]

Using a computer when playing against a robot is hardly cheating but when playing against other people it certainly is - unless there is disclosure and acceptance beforehand.

What about using the computer to analyse a move after it has been made? A main reason to do this is to maximise learning while that move is pertinent in the mind. The aim is to improve your play in future games by recognising the type of position and/or knowing how to act given a particular dice roll - building one's intuition or heuristics. But is it cheating, even if not as obvious as examining a move beforehand?

Well, it will have no effect on the game in some situations but in others it will. If you are attempting to trap a piece behind a prime and have been concentrating on adding a block at the front, but the bg analyser marks your move down because it reckons that you should have been trying to close the rear end first, then your next moves will be influenced as you change your plan. The computer's analysis has suggested a tactic in an ongoing situation - and therefore been of benefit even though used in retrospect.

On the other hand there are very many situations, especially at the start of the game where the volatility of the position means that tactics must be kept very fluid and every position examined as if the game were starting from there. In those situations the use of the computer would not be of much benefit in the current game.
----------------------

So, if you want to use your program, you should be aware that analysing the game after each move could still give you an advantage and this would be unfair if such usage is not disclosed. The best way to use a bg program is after the game has finished.

The same judgements could also be made if you use this board to get other people's opinions about a game while it is still in progress. Although this would be very visible to your opponent if they were a reader of the board, it could be seen as cheating.

Hrqls, Abigail: Further to Mike's suggestion. If you find World Class is fast enough, then move up to Supremo. They both use 2-ply but Supremo looks at more moves at each level. (This is the Move Filter shown at the bottom of the player settings.) I use Supremo on chequer play and cube decisions for both Player and Tutor analysis which gives me a reasonably paced game on a 1GHz machine.

When you come to looking at your games to examine your mistakes, click on the move that you made and the ones above it in the Annotation window and then click on the tiny [3] button down below. This will re-evaluate all those moves at 3-ply and give you much better accuracy. (4-ply is just toooooo slow except for end-game situations or when a piece is stuck on the bar).

You must click all the 2-ply moves above yours otherwise the unclicked ones will be displayed below all the 3-ply evaluations regardless of whether they are better or not. That's just how Gnubg sorts them. You might want to click on a couple of 2-ply moves below yours as well if they are evaluated as close to yours; 3-ply may show them to be better after all.

GnuBg is as user-friendly as Mike says: As always, you must use Settings/Save settings after any changes otherwise it'll throw your new settings away when you leave the program. You'll especially appreciate this snippet of advice if you put a lot of effort in creating a nice 3D board for yourself! ;-)

Pedro Martínez: What settings do you have it on? As installed it plays weaker than it's capable of. However, ten 5-point matches is still hot stuff.

I usually advise people to set Supremo for Chequer play and World Class for Cube decisions in both Settings/Analysis and Settings/Evaluation. Go higher if you have plenty of patience or a stonking fast machine.

Don't forget to do Settings/Save settings afterwards otherwise it will lose the changes when you exit.