Author |
Topic: Essay by Christian Freeling on inventing games (Read 522351 times) |
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender: 
Posts: 804
|
 |
Re: Essay by Christian Freeling on inventing games
« Reply #330 on: Oct 31st, 2010, 6:32pm » |
|
on Oct 31st, 2010, 6:04pm, SpeedRazor wrote:| Christian: please take back your thread; I promise I won't bother you anymore. |
|
Fair enough, and you won't bother anyone if you focus on content instead of displaying ignorance. Here's something about the balancing rule.
A kiss is still a kiss, a sigh is just a sigh, but a swap ain't a pie 
The first quote is copied from Byg (p)review of Symple(x), by Benedikt Rosenau.
Quote:The current rule for Symple is that Black gets an extra (sub-)move for planting a new stone, if and only if he grows first. In other words, if Black decides to grow and he is first at that, he may grow all his existing groups and start a new one.
Assume we are at move 8. Noone has grown so far. So, we have 8 White groups and 7 Black groups on the board. Black grows now and adds the extra-move. After that, White has 8 single stone groups, whereas Black has 7 groups of 2 stones and 1 single stone group. Since Black can still place single stone groups on the board until he has reached the desired number for growing, this is likely in Black's favor.
So, White has waited too long. At what move should White grow in order to prevent Black from doing so? At move 1, it is not possible. At move 2, it gives a one double stone group facing an opponent with a single stone group and the right to move. In the discussion of the three move swap protocol, I have argued that this is lost. Likewise, will Black wait until White can pull a growth on him or will Black secure the extra points and the increased options of the groups grown due to the balancing rule?
Symple is a double Chicken now. The first is when to grow first for the sake of either using or, well, disabling the balancing rule, and the second is inherent in the game mechanism itself - when do you have generally enough groups to grow from?
In any case, we can infer that the balancing rule succeeds at balancing the game. And the result is an extra twist. Since the extra rule "empowers" judgment respectively skill in the growth phase - that is where the result of the first chicken game will show -, I argue it gives an even better game. |
|
Benedikt does in fact explain the balancing principle adequately, but I had to understand the use of 'Chicken' from the context. Moreover, there are other opinions to be taken into account, for instance this one of Mark:
Quote:| Byg, Symple, etc. are based on an unstable, divergent principle. If they're in the same class as Go Moku, only worse, which I strongly suspect, advancing skill will always be accompanied by advanced tweaking of the balancing rule set. Your special balancing rule, the Nagasaki of Symple, makes the the pie rule look like a firecracker in comparison. |
|
I attribute the comparison with Go Moko to Mark's somewhat eccentric sense of humor and his reference to 'divergence' to not quite seeing black's privilige working as a pie-rule.
Not a swap, a pie-rule, despite the fact that wiki treats them as one and the same. So what's the difference? This is what wiki says:
Quote:| The rule gets its name from the divide and choose method of ensuring fairness in the division of pie between two people; one person cuts a pie in half, then the other person chooses which half to eat. |
|
You have to eat it too, obviously  but what I'm getting at is the order 'divide and choose'. A slightly different protocol allows to reverse the order. One person moves the knife over the pie, from side A to side B. Both can say 'cut' at any moment, but the piece chosen must be at the 'A' side. It doesn't matter who yields the knife, other than that the person doing so can refrain from saying 'cut', and simply cut.
The result of this 'choose and divide' protocol is the same: both get about half the pie. But there are two differences:
a: The element 'timing' is introduced.
b: The protocol works for any number of participants: 'n' people can divide a pie in 'n' pieces that all converge around about the same size.
We're concerned with the first one, because the 'timing' aspect is exactly what guides Symple on two different levels:
1: When to stop placing isolated stones and start growing.
2: When to either cash in or prevent black's privilige.
How to place and where to grow is a different matter. These are strategical and positional aspects and I'll leave them outside the equation for now.
Both 1 and 2 are different for white and black. Without 2, white would have the advantage, but still neither can start too early because the later the start, the bigger the impact of the first growth and the stronger the initiative. An early start has little enough impact to allow the opponent to postpone growing for another move, or even two, and reap the fruits of an extra group or two in the subsequent turns.
Within that context, black can't very well start first, because he would be one group down, allowing white one stone extra in every move in the foreseeable future.
So white can wait to see the impact of his 'first stake' ripe to its full potential. But even without 2 he cannot wait forever, because the very impact of the first strike will at a certain point surpass the advantage of white's extra group if black decides to go for it. This converges to a certain area of moves, beyond which white's advantage diminishes.
The balancing mechanism
The balancing mechanism of Symple a clear convergence too. As long as no growth has taken place at either side, Black may both place stone and grow all other groups. So this is something black can cash in, and white can prevent. What's the timing?
If black grows on his second move, both will have two groups, that is white has two stones and black has a stone and a group of two stones. For white this is almost as advantageous as a situation without black's privilige, so black won't do that.
From black's point of view he'd rather cash in as late as possible. Having an equal number of groups and a strong initiative is precisely what constitutes white's advantage without 2, so he would like to do that.
But he can't wait that long, because white can grow at any turn, depriving black from his advantage. He's worse off than without 2, but it's better than leaving black exploit his privilige.
So black must cash in fairly early or not at all, grabbing a small initiative with an equal number of groups. This converges to a certain area of moves, beyond which white can grow and take a small initiative in return for black now having 'the move', that is: an equal number of groups on his turn.
Both 1 and 2 are converging, though to what specific area's remains to be seen. At the moment cashing in or its prevention on the 19x19 board appears to be around move 4-5, while the main 'pie' - when to start the growing phase - hovers around move 10-15. Still somewhat unclear, but naturally converging. There is no doubt whatsoever that black's prerogative impicitly evens out any first move advantage, as Benedikt explicitly mentioned, considering it evident, and no 'tweaking' will be required. Symple is complete and consistent.
|
|
 IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #331 on: Oct 31st, 2010, 6:37pm » |
|
on Oct 31st, 2010, 6:16pm, MarkSteere wrote:Symple Newsflash:
"Note: In the first version of the game the 'group penalty' was set at 2 rather than 4, which leads to the possibility of..."
BOOM!! What did I say about the tweaking Do you believe me now |
|
Not below the belt Mark, the parameter was open from the start, in fact a point of discussion from the very beginning of Star. It has no relation to the balancing principle. The applet allows players to experiment with either 0, 2 or 4, but 4 is the best value, and not arbitrary, for the reasons given.
|
|
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #332 on: Oct 31st, 2010, 11:06pm » |
|
on Oct 31st, 2010, 6:37pm, christianF wrote:
Sorry, but times of crisis call for masterful baiting.
|
|
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #333 on: Oct 31st, 2010, 11:25pm » |
|
on Oct 31st, 2010, 6:32pm, christianF wrote:
I attribute the comparison with Go Moko to Mark's somewhat eccentric sense of humor |
|
Christian, the following paragraph from the Symple rule sheet tortures me.
"There's one exception to the above. If, and only if, neither player has grown yet, then black on his turn may use both the above options in the above order: he may place a stone, therewith creating a new group, and he may grow any or all of his other (!) groups."
This is not my "eccentric sense of humor" talking. I had to cover my eyes when I cut and pasted it so I wouldn't accidentally read it again. Please stop comparing your unequal goals rule to the innocuous pie rule. It's so unfair to the pie rule.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #334 on: Nov 1st, 2010, 2:39am » |
|
on Oct 31st, 2010, 11:25pm, MarkSteere wrote:| This is not my "eccentric sense of humor" talking. I had to cover my eyes when I cut and pasted it so I wouldn't accidentally read it again. Please stop comparing your unequal goals rule to the innocuous pie rule. It's so unfair to the pie rule. |
|
Lol, your ignorance has a quality all its own  .
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #335 on: Nov 1st, 2010, 6:20am » |
|
Here's the first Symple variant. It's an interesting idea that leads to an even more explosive branch density and a far more tactical game, in which positional play in the opening bears less weight on the growing phase, because players are free to shift the focus to 'local issues'. Feels like fun  .
|
|
IP Logged |
|
|
|
FransFaase
Forum Newbie
Arimaa player #5815
Gender:
Posts: 1
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #336 on: Nov 1st, 2010, 7:59am » |
|
I think that Christian might be right that the game (Super Symple) becomes more tactical. Now that I have thought a little more about the game, it really feels like a very different game. The strategy of enclosing groups of your opponent is complete absent in Super Symple, which means that the other method for gaining an advantage, creating territory, becomes the focus of the game. I guess that that would makes the dominance of corners and the sides much more greater than with Symple (like in Go). I also guess it becomes easier to connect groups (and/or prevent your opponent from connecting), which might make the moment of connecting more critical. And then there is this nasty thing in the end, that if you end up with too few groups and a too large territory, you face the possibilty of an invasion. I don't know if this shift in focus makes it a far less strategical game, because creating territory is a strong strategical principal.
|
|
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #337 on: Nov 1st, 2010, 8:35am » |
|
on Nov 1st, 2010, 6:20am, christianF wrote:
Here's the first Symple variant.
"Only if you can create a group of three stones, there is some gain. That group has a negative score of one, but you take away three points of your opponent, creating a net score of plus two..." |
|
brb, I need a calculator
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #338 on: Nov 1st, 2010, 8:49am » |
|
on Nov 1st, 2010, 8:35am, MarkSteere wrote:
Enter 2 x 3 - 4 and next "="
P.S. the net effect of a group of 'n' stones is '2n-4', but that borders on math and may be a bridge too far  .
|
| « Last Edit: Nov 1st, 2010, 9:06am by christianF » |
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #339 on: Nov 1st, 2010, 9:48am » |
|
Symple borders on math and may be a bridge too far.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #340 on: Nov 1st, 2010, 10:59am » |
|
on Nov 1st, 2010, 9:48am, MarkSteere wrote:| Symple borders on math and may be a bridge too far. |
|
That's inherent to the 'Star' theme isn't it, which is based on getting points for (such and such) stones, but paying a penalty for every group.
I wouldn't know how to find a game on a theme, while avoiding the theme, but maybe you can help out .
|
|
IP Logged |
|
|
|
Fritzlein
Forum Guru
Arimaa player #706
Gender:
Posts: 5928
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #341 on: Nov 1st, 2010, 3:01pm » |
|
Christian, I responded to both of your private messages to the e-mail address you included, but for some reason my letters both bounced. Here is one question I included that might as well be discussed publicly:
You say
"A pie however can be extended to more than two players, a swap cannot."
Are you sure that swap can't be extended to more than two players? I don't play any multiplayer abstracts, but my imagination is failing me here. Why can't a three-player game be balanced by giving the second player the option to swap with the first, and the third player the option to swap with the second player, or if necessary with either of the other two?
|
| « Last Edit: Nov 1st, 2010, 3:25pm by Fritzlein » |
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #342 on: Nov 1st, 2010, 3:33pm » |
|
on Nov 1st, 2010, 3:01pm, Fritzlein wrote:Christian, I responded to both of your private messages to the e-mail address you included, but for some reason my letters both bounced. Here is one question I included that might as well be discussed publicly:
You say
"A pie however can be extended to more than two players, a swap cannot."
Are you sure that swap can't be extended to more than two players? I don't play any multiplayer abstracts, but my imagination is failing me here. Why can't a three-player game be balanced by giving the second player the option to swap with the first, and the third player the option to swap with either of the other two? |
|
Yes, I hadn't considered that. The point I wanted to illustrate is that a pie may involve 'timing' and a regular swap doesn't.
The rule that tortures Mark's eyes, nothing more than given black a one time conditional opportunity to combine both options, how weird can it get , works as a pie and is implicitly balancing. Which is very fortunate, because a swap doesn't work in Symple.
But as I've always said: "if the system is sound, the rule will be there".
Regarding a multi player pie, I must admit that the only thing I visualized, a long time ago, is ten children and a literal pie of around a meter or so, and a knife going slowly from one side to another, and the children who may call 'cut' at any moment. How that would translate to a multi player game is another matter. It may not be very useful I fear.
Phalanx and Mu have balancing procedures, but I never got anywhere near something resembling a swap (or a pie for that matter).
P.S. christian'at'mindsports.nl ('at' being @, but maybe I'm overly careful) should work.
|
| « Last Edit: Nov 1st, 2010, 3:53pm by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #343 on: Nov 3rd, 2010, 9:55am » |
|
I've been notified, by a very dependable source, of a bug in Symple. However, the bug wasn't specified, but 'left for me to find'. The source added reassuringly that 'it shouldn't be too hard'.
My way of finding bugs is to wait till they show up. For whoever is more hasty, 'it shouldn't be too hard'.
|
|
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #344 on: Nov 3rd, 2010, 11:34am » |
|
Sorry to hear that about the bug. Is it a bug where there are no legal moves available or is it an algorithm for first player win?
|
|
IP Logged |
|
|
|
|
Home
Help
Search
Members
Login
Register
Notify of replies
Send Topic
Print