Author |
Topic: Essay by Christian Freeling on inventing games (Read 522068 times) |
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender: 
Posts: 804
|
 |
Re: Essay by Christian Freeling on inventing games
« Reply #435 on: Jun 3rd, 2011, 9:45am » |
|
A good friend of both Ed van Zon and me, Gerard Dijkman, astrologer extraordinaire, died yesterday. It's becoming a trend. We're from the "People try to put us down, talking 'bout my generation" generation but there's hardly any need to push it, nowadays.
He was extremely well-read, erudite and funny, and an alcoholic, and that's what did him in. Now the rest of us who ruled in the seventies make arrangements to go to the funeral together and talk about old times and who's most likely to go next. The last one will know for sure 
I'll be back later.
|
| « Last Edit: Jun 3rd, 2011, 9:46am by christianF » |
 IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #436 on: Jun 3rd, 2011, 11:10am » |
|
on Jun 3rd, 2011, 9:45am, christianF wrote:
A good friend ... died yesterday. It's becoming a trend ... make arrangements to go to the funeral together ... who's most likely to go next. |
|
Anything else we can talk about besides Mt. Doom and old people dying?
on Jun 3rd, 2011, 9:45am, christianF wrote:
Well, yeah, if you're still alive.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #437 on: Jun 7th, 2011, 5:53am » |
|
To keep this on target, what I'm saying about Symple's and Sygo's move order advantage balancing mechanism is that, given a regular boardsize, it isn't possible to argue one way or another. I'm not talking about "the truth" because the truth, although known to exist for any move in any position in any two-player abstract perfect information game, cannot be determined.
I'm not claiming a revolution here either. The pie in Hex goes a long way in the same direction.
Gametrees
The truth of any game is contained in its gametree. If all possible legal positions and how they result from one another are mapped, the result in terms of win and loss can be tracked backwards from the leaves, leaving the remaining connections as draws, should these be possible. Only relatively small games have been completely mapped: Tic-tac-toe, minimancala, awari, fanorona, checkers, to name a few.
Here's the minimancala gametree. It has two leaves. It also has cycles and is thus called 'infinite'. Some games, like Hex or Othello, don't and are thus called 'finite'. Most dominant games, Chess, Go, Draughts, Shogi, Xiangqi, have cycles in their respective gametrees. Saying that ‘finite games rule’ is like Castro saying ‘communism rules’.
In the gametree of any size Hex the truth of every position is win or los. Hence any move in any size Hex is always winning or losing, never 'advantageous'.
Truth and the human condition
But that doesn't help a human player when faced with an opening move in 19x19 Hex and having to decide whether or not to swap. Here we already have a case where it is difficult to 'argue one way or the other'. The relative merits of a move cannot be considered in terms of 'perfect play', because that would be based on complete access to the gametree, and nothing would be 'relative' anymore.
Humans would have to rely on a multitude of imperfect games between 'consistently high level players' with that particular opening move to get to the 'most likely' result.
That is true for any opening move, so given enough high level matches the 'most likely truth' of each and every move would emerge.
Convergence and divergence - a timeline
More importantly: the overall results of each particular move will eventually drift away from 50/50, not converge on it, because the move is either winning or losing and that fact will eventually 'shine through' in the results.
In this world Hex probably isn't played extensively enough to become 'problematic' on 19x19. Players make mistakes and the 'truth' may favor one player or the other several times during a game without the players being aware. And where the result of any particular opening would eventually diverge from an equal score, it initially will converge on it because that's the very point of a swap. Divergence only takes place once its intricacies have been extensively explored.
A high resolution refinement
Suppose extensive play would eventually have zoomed in on the truth of every one of the 181 opening moves in 19x19 Hex, then a refinement could be introduced in the form of a 3-stone swap. Now players would have to zoom in on the truth of more than 20 million positions. That would mean that the procedure would hardly get beyond the ‘convergence’ stage. That subjected to extensive play each of the positions would eventually diverge from an even result towards ‘the truth’ has become meaningless in view of the sheer number of positions to be considered. Arguing one way or the other would be impossible for a large subset of the 20 million plus positions.
The balancing mechanism of Symple and Sygo is of a much higher resolution, and that by nature, not by 'amplification' as in the 3-stone swap in Hex. It's not a swap: white is white and black is black and at no time a switch is offered. A further difference is that it is based on the timing of an action rather than an initial choice.
Note: it is inherently based on the move procedure of both Symple and Sygo: you either place a single stone, thus starting a new group, or you grow any or every one of your groups present on the board by one stone. To understand the balancing mechanism, one must be familiar with
the move procedure and the inherent dilemma as set out in About Symple.
The obvious revisited
Since one poster clearly wrestled with the balancing mechanism I'll try a step by step explanation.
In Symple white has the move order advantage so we counter this inherent a-symmetry by giving black a compensation.
The compensation is this: Black may once, and conditionally, place a stone and grow every one of his groups.
If this compensation were unconditional, Black would have the advantage. Usually (not always) when one player decides to start growing his 'groups' (as yet single stones) instead of placing new ones, this gives an initiative that incites the other, at least to a degree, to follow suit.
If Black simply awaits White's initiative, he can follow suit and have an extra group.
If White doesn't grow, Black can eventually decide to cash in his option himself and have a large initiative with an equal number of groups.
So giving the compensation unconditionally doesn't solve the problem.
Pushing the button
The condition under which Black may cash in his compensation is that neither player has grown yet. This means that White can grow first anytime and therewith take Black's compensation from him.
Or that Black can grow first anytime and thus cash in his compensation.
Both players are faced with when to 'push the button'.
Now we get to the point where I fail, where some succeed, to see any lack of clarity. It is copied directly from About Symple:
Quote:Let's first look at this from white's position:
If he grows on his second move, he will have one group of two stones and black will have one stone, black to move. For black this is almost as good as having the first move without any compensation for his opponent. So white must wait if he wants his compensation for terminating black's prerogative to grow. But how long? |
|
Quote:Now let's first look at this from black's position:
If he uses both options on his second move, he will have one stone and one group of two stones and white will have two stones, white to move. For white this is almost as good as having the first move without any compensation for his opponent. So black must wait if he wants the advantage of his prerogative to grow. But how long? |
|
Both players have a clear disadvantage if they push the button too early. However, with the impact of the compensation rising with the number of stones placed, pushing too late will at some point, and not more than a few moves away, give the opponent a clear advantage. So timing is of the essence.
This I put up with the question what , if anything, was wrong with the reasoning. Here's an intesesting reply by dr. Sheldon Cooper of finitegames.nut:
Quote:| Yes, the bizarre question about your presumed faulty reasoning. I can't say I would be shocked beyond belief if there was something wrong... |
|
It shows that the author is waiting for someone else to validate or refute the argument, which I find striking.
Yet, at the same time, I'm honored by his well meant comments, because he considers Symple and Sygo important enough to spend some considerable time on them.
Household announcements
I thank the readers and the posters because I'll leave for now and the foreseeable future. Some readers may have noticed that it wasn't the stacking games contest I can't get out of, so rigorous measures are necessary 
If we have new implementations at mindsports.nl I'll let you know though.
|
| « Last Edit: Jun 7th, 2011, 12:16pm by christianF » |
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #438 on: Jun 7th, 2011, 9:58am » |
|
on Jun 7th, 2011, 5:53am, christianF wrote:
I'm not talking about "the truth" |
|
That much is clear.
on Jun 7th, 2011, 5:53am, christianF wrote:
Saying that ‘finite games rule’ is like Castro saying ‘communism rules’. |
|
I knew there'd be something about me in here, though I never imagined I'd be branded a communist! lol "Finite games rule" is my motto at Christian's game site, MindSports. Finite games, unlike communism, really do rule.
on Jun 7th, 2011, 5:53am, christianF wrote:
Any move in any size Hex is always winning or losing, never 'advantageous' |
|
What?? Plenty of moves in Hex are advantageous. In perfect play, moves are winning and losing. In imperfect play, moves are advantageous and disadvantageous. This applies to Hex and to all other games, from which, once again, Sygo is not exempt.
on Jun 7th, 2011, 5:53am, christianF wrote:
If only.
on Jun 7th, 2011, 5:53am, christianF wrote:
Given enough high level matches the 'most likely truth' of each and every move [in 19x19 Hex] would emerge. |
|
Boy that's a lot of emerging truth. Are we assuming unlimited player IQ's now? One loses track of all the surreal parameters.
on Jun 7th, 2011, 5:53am, christianF wrote:
More importantly: the overall results of each particular move will eventually drift away from 50/50, not converge on it, because... |
|
...because nothing "converges" in games! Games are inherently divergent. All games. Including Sygo. The similarity of that word to psycho can no longer be ignored.
on Jun 7th, 2011, 5:53am, christianF wrote:
Divergence only takes place once its intricacies have been extensively explored. |
|
Divergence takes place from the word go, and never lets up. Pastafarianism is the only thing I can think of more absurd than this convergence/divergence nonsense you've whipped up, Christian. You used the words true and truth eight times in this one post. This is like my housekeeper droning on about the truth of the gospel.
Show of hands: Is anyone buying this "convergent Sygo" nonsense?
|
|
IP Logged |
|
|
|
megajester
Forum Guru
Istanbul, Turkey
Gender:
Posts: 710
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #439 on: Jun 8th, 2011, 12:32am » |
|
Hello everyone who's just joined us, yes, right now we're watching Mark Steere's back garden and... Oh there he is, he's coming out now. Might he be about to mow the lawn perhaps?... No he's setting up some mannikin of some sort, or is it a scarecrow?... No it's a straw man... And a very handsome straw man if I say so myself, a worthy addition to any vegetable allotment... Ah but what's he doing now? He's pulling... is that a baseball club, out of his bag? And he's... he's... BEATING THE STRAW MAN OH IT'S SO BARBARIC I CAN'T WATCH... STRAW... EVERYWHERE... HE'S SHOUTING, SHOUTING PROFANITIES AT IT... DEMANDING IT ANSWER HIM... OH, THE STRAWMANITY!
on Jun 7th, 2011, 5:53am, christianF wrote:Players make mistakes and the 'truth' may favor one player or the other several times during a game without the players being aware. And where the result of any particular opening would eventually diverge from an equal score, it initially will converge on it because that's the very point of a swap.
|
|
An equal score, eh? You don't suppose he could be talking about accruing or losing advantage in imperfect play, hmm?
on Jun 7th, 2011, 5:53am, christianF wrote:
Any move in any size Hex is always winning or losing, never 'advantageous'
|
|
on Jun 7th, 2011, 9:58am, MarkSteere wrote:
What?? ... In perfect play, moves are winning and losing. |
|
Oh for pity's sake, you're not supposed to agree with the straw man! Come on Mark, we know you can do better!
on Jun 7th, 2011, 9:58am, MarkSteere wrote:
Plenty of moves in Hex are advantageous. ... In imperfect play, moves are advantageous and disadvantageous. ... Show of hands: Is anyone buying this "convergent Sygo" nonsense? |
|
OK if I accrue an advantage, and then give it back again, we've returned from our short sojourn into the realm of asymmetry to "converge" to the point where it's a straight 50/50 that either of us will win.
So no if we're talking about perfect play. And yes if we're talking about imperfect play.
It's actually fairly obvious what he's talking about where if you stop to think about it for two seconds. But that wouldn't give us any ammo now would it?
Sorry hang on just one second... Yes... Yes I believe that was Mr. Steere's mother calling from the kitchen to tell him to clear up all the mess he's made with his straw man...
|
|
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #440 on: Jun 8th, 2011, 8:31am » |
|
on Jun 8th, 2011, 12:32am, megajester wrote:
we've returned from our short sojourn into the realm of asymmetry to "converge" to the point where it's a straight 50/50 that either of us will win. |
|
You've sojourned into the realm of the surreal, converging to a dank, dark place reeking of methane, and sounding of faint, muffled Dutch conversation.
on Jun 8th, 2011, 12:32am, megajester wrote:
It's actually fairly obvious what he's talking about... |
|
...when you inhale deeply and drift off in a methane rush.
|
|
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #441 on: Jun 8th, 2011, 11:05am » |
|
Megajester: "Sometimes I have the advantage and then later I don't. Therefore Christian Freeling's claim that 'Sygo converges to 50/50 in the long run' makes perfect sense."
Brilliant. How do you do it?
|
|
IP Logged |
|
|
|
megajester
Forum Guru
Istanbul, Turkey
Gender:
Posts: 710
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #442 on: Jun 8th, 2011, 2:49pm » |
|
on Jun 8th, 2011, 11:05am, MarkSteere wrote:Megajester: "Sometimes I have the advantage and then later I don't. Therefore Christian Freeling's claim that 'Sygo converges to 50/50 in the long run' makes perfect sense."
Brilliant. How do you do it? |
|
It's one thing to take something somebody said and make a straw man out of it. It's quite another to use something that person never actually said.
It's brilliant. How do you do it?
|
|
IP Logged |
|
|
|
Eltripas
Forum Guru
Meh-he-kah-naw
Gender:
Posts: 225
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #443 on: Jun 8th, 2011, 11:48pm » |
|
Board games. Serious business.
|
|
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #444 on: Jun 9th, 2011, 1:49am » |
|
on Jun 8th, 2011, 11:48pm, Eltripas wrote:
Board games. Serious business. |
|
Only when people make outlandish claims and then other people mindlessly jump on board with said outlandish claims.
Christian and Nanojester are barely capable of formulating a sentence, never mind a coherent logical premise. This is why I have to argue both sides.
|
|
IP Logged |
|
|
|
Fritzlein
Forum Guru
Arimaa player #706
Gender:
Posts: 5928
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #445 on: Jun 9th, 2011, 2:04am » |
|
on Jun 9th, 2011, 1:49am, MarkSteere wrote:| Christian and Nanojester are barely capable of formulating a sentence, never mind a coherent logical premise. |
|
That's not vulgar, but it is a personal attack. If it were a first offense, I would warn you myself. If it were a second offense, I would ask Omar to warn you. In this case, however, given your long history of abusive posts, I hope that Omar, having warned you before, will now ban you from the Forum. Unfortunately, he will probably just warn you again. Oh, well, if a warning will get you to merely ridicule what people say rather than ridiculing people directly, I will go back to gritting my teeth and not responding to your posts.
|
| « Last Edit: Jun 9th, 2011, 2:16am by Fritzlein » |
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #446 on: Jun 9th, 2011, 6:12am » |
|
on Jun 9th, 2011, 2:04am, Fritzlein wrote:
That's not vulgar, but it is a personal attack. |
|
I don't mind being attacked in person all that much (not that I see the point of it), but criticizing games should be done in a fair spirit, and it should at least make some sense.
Here's unfair:
When I say "I'm not talking about "the truth" because the truth, although known to exist for any move in any position in any two-player abstract perfect information game, cannot be determined.", I'm talking about the game theoretical proof that all these games are completely determined. What's taken out is: "I'm not talking about "the truth", and the reply is "That much is clear", referring to a whole different meaning of the word.
Here's no sense (just one example of a recurring theme):
on Jun 7th, 2011, 5:53am, christianF wrote:In the gametree of any size Hex the truth of every position is win or los. Hence any move in any size Hex is always winning or losing, never 'advantageous'.
Truth and the human condition
... The relative merits of a move cannot be considered in terms of 'perfect play', because that would be based on complete access to the gametree, and nothing would be 'relative' anymore.
Humans would have to rely on a multitude of imperfect games between 'consistently high level players' with that particular opening move to get to the 'most likely' result. |
|
The reply is:
on Jun 7th, 2011, 9:58am, MarkSteere wrote:| What?? Plenty of moves in Hex are advantageous. In perfect play, moves are winning and losing. In imperfect play, moves are advantageous and disadvantageous. This applies to Hex and to all other games, from which, once again, Sygo is not exempt. |
|
That's repeating what I say while suggesting I meant the opposite! All these games are completely determined, that's common knowledge and I never said otherwise. And it's not what my claim about Symple and Sygo is about. But I'm bored with its deliberate misrepresentation. Players can either try Symple or Sygo or leave it. Opinions won't make them different games.
|
| « Last Edit: Jun 9th, 2011, 8:40am by christianF » |
IP Logged |
|
|
|
ocmiente
Forum Guru
Arimaa player #3996
Gender:
Posts: 194
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #447 on: Jun 9th, 2011, 9:55am » |
|
on Jun 9th, 2011, 6:12am, christianF wrote:
I don't mind being attacked in person all that much (not that I see the point of it), but criticizing games should be done in a fair spirit, and it should at least make some sense.
|
|
Criticizing people should be done in a fair manner, and it should make some sense.
The way this was stated made me think that the writer thought, in some small way, that games deserved more respect than people.
|
|
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #448 on: Jun 9th, 2011, 9:56am » |
|
on Jun 9th, 2011, 6:12am, christianF wrote:
What's taken out is: "I'm not talking about "the truth", and the reply is "That much is clear", referring to a whole different meaning of the word. |
|
That's why it was obviously a joke, Christian. It's called irony. Could you possibly be more sensitive?
on Jun 9th, 2011, 6:12am, christianF wrote:
I'm bored with [Sygo's] deliberate misrepresentation. |
|
So stop deliberately misrepresenting Sygo! Good Gordon! Nobody's misrepresenting Sygo more than you! First we have your outlandish claim that "Sygo converges to 50/50 in the long run." Logic torpedo: If your claim is true in principal, then it should be true for all board sizes, including tiny boards, which it clearly is not. You ran back to the previously unbearable Arimaa forum to duck said logic torpedo in rec.games.abstract. Now that you're here, you're pretending not to notice that the logic torpedo has followed you. It's still there, Christian!
Next on the Sygo ridiculous claim list is the notion that...
"Hex, without the pie rule, is a proven win for the first player. ...
What makes Symple different is that you can't even argue one way or the other"
Hogwash. Hex was not solved, except for on a handful of small boards. There's a difficult, complex, theoretical proof that Hex is a first player win, but nowhere in that proof is any indication of how one goes about winning. Your argument that Sygo is actually impervious to argument is beyond absurd. It's asinine!
This is like Cameron Browne fantasizing about human designed games vs computer designed games, as though humans had nothing to do with the computers designing games. It's psuedo-science and it's offensive.
Christian: "Oh no, that's not what I said. You're putting words in my mouth."
Well, why don't you start making sense, Christian. Then your "words" won't be so subject to misinterpretation. Sygo has no mystical properties that other games such as Hex don't have.
As long as you continue to expound the magical properties of Sygo, logic torpedoes will continue to be fired. If not here, then in Tidbits, and I don't think you want that. Nobody likes Tidbits.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #449 on: Jun 9th, 2011, 11:51am » |
|
on Jun 9th, 2011, 9:55am, ocmiente wrote:
Criticizing people should be done in a fair manner, and it should make some sense.
The way this was stated made me think that the writer thought, in some small way, that games deserved more respect than people.
|
|
I didn't mean to imply that I feel people don't deserve respect. I just feel that criticizing games should be done in a fair spirit. That doesn't mean other actvities should not. Neither does it mean I put games before people. If someone states that he or she cares about art, music, literature or math, would you voice the same suspicion?
|
|
IP Logged |
|
|
|
|
Home
Help
Search
Members
Login
Register
Notify of replies
Send Topic
Print