Welcome, Guest. Please Login or Register.
Mar 29th, 2024, 8:40am

Home Home Help Help Search Search Members Members Login Login Register Register
Arimaa Forum « Essay by Christian Freeling on inventing games »


   Arimaa Forum
   Arimaa
   Off Topic Discussion
(Moderators: christianF, supersamu)
   Essay by Christian Freeling on inventing games
« Previous topic | Next topic »
Pages: 1 ... 4 5 6 7 8  ...  78 Notify of replies Notify of replies Send Topic Send Topic Print Print
   Author  Topic: Essay by Christian Freeling on inventing games  (Read 520650 times)
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #75 on: Apr 26th, 2009, 2:16am »

on Apr 25th, 2009, 9:10pm, Adanac wrote:
I think HaniBall has tremendous potential and I’m very curious as to whether it’s possible to score in these 10 scenarios.  If the game needs a only minor rule tweak to make it work, I won't lose any sleep over it  Wink

Neither will I, the game is after all more important than my ego, and I'm as curious as you are. I'll have to take a closer look at the scenarios first though.
 
I'm flattered by the 'tremendous potential', but the question is 'as what'? On our homepage is stated:
Quote:
"Strategy games have strategies varied enough to allow different styles of play, tactics varied enough to induce their own terminology, and a structure that allows advantageous sub-goals to be achieved as calculable signposts along the way.
Tactical games have strategies that are either fairly obvious (however deep), like Pente, or fairly obscure, like Othello."

These 'advantageous sub-goals' ... Chess is full of them. But HanniBall must do without a strategical framework such as is for instance provided by pawns. The only permanent sub-goal would be winning a piece, but barring that, the game is far more volatile in that it is difficult to imagine any more or less permanent sub-goal. Like a soccer match really.
 
So I would label HanniBall a tactical game with a 'fairly obscure' strategy. This may have some commercial potential (whishful thinking, more likely) but for me it remains a possible intelligent recreational pastime rather than a 'strategy game'. No value judgement implied.
« Last Edit: Apr 26th, 2009, 4:27am by christianF » IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #76 on: Apr 26th, 2009, 4:25am »

on Apr 25th, 2009, 9:10pm, Adanac wrote:
Choose one of the following scenarios (or multiple scenarios if anyone wishes) and try to score on me:
 
1.  White: E-E on d2 & f2  // Black:  E-E on d10 & f10  
2.  White: L on e2  // Black:  E-E on d10 & f10
3.  White: L on e2 // Black:  H-H on d10 & f10
4.  ...

Adanac, I've looked at the first couple (I'd add White: L on e2 // Black: H on d10, E on f10) but wouldn't it be better to have info on one-on-one endgames first? Here's a table I tend to fill. The defender (White) using the Keeper and a piece. The attacker (Black) using only a piece and starting with the ball from the center spot.
Say + if the attacker wins, = if it's a draw.
 
 
Attacker Defender result
1. Elephant Elephant
=
2. Horse Horse
?
3. Horse Elephant
?
4. Elephant Horse
?
5. Lion Lion
?
IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Cancelling a rule / Zillions game available!
« Reply #77 on: Apr 26th, 2009, 7:41am »

I've had a look at an attacking Horse against a defending Elephant and the outcome leads me to the conclusion that there slipped a superfluous rule into the game.
 
Guilty as charged! Embarassed
 
The rule that players are not allowed in the goals appears to needlessly favor the defender.
So I've omitted it.
It boils down to cancelling one sentence. The rule slipped in on a misplaced sense of esthetics.
 
It's omission is in line with a point already made by the more sceptical section of the forum, by Fritzlein for instance, that a defensive strategy may lead to problems.
 
On top of that, Ed just called with the announcement of the first moves made by Zillions.
Zillions, according to Ed's first impression, favors a defensive strategy as well*. It's to soon to say to which extend, because dwindling material makes a game more 'decision prone', but there's no reason for an explicit rule that favors defensive play, in particular if the game is better off without it.
 
The omission of the rule implies that all players now may be anywhere on the field, including the goals. Nothing changes with regard to a ball in the goal, with or without a player of whatever side: the game ends in a win for the other Keeper's side. A piece other than the Keeper in the own goal has no great defensive potential, because a ball shot at it lands in the goal just the same, and the opponent wins. It at most can prevent an attacking piece from entering the goal, ball and all.
 
Important
The cancelling of the rule will certainly have consequences for the scenarios that Adanac suggests to test defensive strategies.
More precisely: it will turn out to favor the attacker!
 
Now back to the one-on-ones. The attackers main concern is of course to avoid being captured while approaching the goal:
 
Attacker Defender result
1. Elephant Elephant
draw
2. Horse Horse
?
3. Horse Elephant
win
4. Elephant Horse
?
5. Lion Lion
?

 
* The Zillions game is available in a bèta-0.3 version, albeit without implementation of the obstruction rule, but also already without the rule I omitted above.
 
download HanniBall (beta 0.3) from the rules page
download HanniBall (beta 0.3) directly
 
Meanwhile Ed has reason to reconsider the degree to which to program 'plays defensively' - if given half a chance it does indeed attack.
« Last Edit: Apr 26th, 2009, 10:35am by christianF » IP Logged
Adanac
Forum Guru
*****



Arimaa player #892

   
Email

Gender: male
Posts: 635
Re: Essay by Christian Freeling on inventing games
« Reply #78 on: Apr 26th, 2009, 11:37am »

on Apr 26th, 2009, 4:25am, christianF wrote:

Attacker Defender result
1. Elephant Elephant
=
2. Horse Horse
+
3. Horse Elephant
=
4. Elephant Horse
=
5. Lion Lion
=

 
IMPORTANT NOTE:  I examined these scenarios before seeing Christian's new rule regarding nets.  The key difference is that Horses can now win against Elephants and Horse versus Horse is much easier.  Otherwise, the same analysis applies.
 
Horse vs. Elephant
 
Attacker: Black Horse & ball on e10
Defender:  White Elephant on e2
 
With perfect play I believe this is a draw.  A ball-carrying horse is a threat to score anywhere from rows 2-6.  However, if the white elephant steps up to e3 it benefits from Keeper protection while covering all 35 squares in b2:h6.  The black horse will need to score between a2:a6 or i2:i6.
 
1.  Ee2-e3    H*e10-a5
2.  Ke1-d2    H*a5-c4-a3
3.  Kd2-c2/Ee3-c4  Draw  
 
In this position, the horse flees with the ball and the elephant returns to e3.  The key is that the elephant did not de-centralize until required.  It would have been a blunder to make an unnecessary early decentralization:
 
1.  Ee2-e3    H*e10-a5
2.  Ke1-d2/E3-c4   H*a5-g4  Goal-in-one
 
Elephant vs. Horse
 
Attacker: Black Elephant & ball on e10
Defender:  White Horse on e2
 
A defending horse can capture a ball-carrying attack over the vast majority of squares over a 5 square radius.  However, there are key squares that a horse cannot reach within its 3 step allotment:
 
2-2  (two steps in one direction, two in the other)
4-4
5-1
5-3
 
So, with a defending horse on b8, for example, an attacker should carry the ball to a square such as d6 or f4.
 
I think the ideal square for a defending horse is e5.  From that square, the key weak squares are c3, g3, c7 and g7.  However, the keeper is protecting c3 and g3 leaving very little room for infiltration.  That forces the Elephant to infiltrate via c7 or g7 and then later kick the ball into the corner upon further advancement.  I cannot find a forced win for the attacker but I haven’t examined all possibilities yet.
 
1.  He2-e5    Ee10-g7
2.  Ke1-g3    *g7-h5/Eg7-g5
3.  He5-g4
 
At this point, I tried various strategies for the horse to try to score but could find anything.  This looks like a draw to me, but I’m not 100% sure.
 
Lion vs. Lion
Attacker: Black Lion & ball on e10
Defender:  White Lion on e2
 
This is a curious stalemate as neither piece can end the turn with possession nor kick the ball within 1 move of the opposing lion, lest it be captured.  The best I can come up with here is a perpetual dance in which the lions can trade roles as attacker and defender but neither side makes any real progress.
 
1.  Le2-e4   L*e10-c6/*c6-a5
2.  Le4-c3   Lc6-a5/L*a5-b3/*b3-a3
3.  Lc3-a3/*a3-b5  Lb3-b5/*b5-c5
4.  La3-c5/*c5-d7  Lb5-d7/*d7-e5  
Draw
 
Horse vs. Horse
 
Attacker: Black Horse & ball on e10
Defender:  White Horse on e2
 
This is the toughest scenario.  An attacking horse can drive a defending horse absolutely crazy by constantly kicking the ball one orthogonal space beside the defending horse and then setting up to repeat the process on the next move.  Or sometimes it can just pick up the ball and move 2 diagonal steps from the defender.   Note that neither horse can capture the other unless one carelessly hold onto the ball on the 3rd step.
 
1.  He2-e5   H*e10-c7
2.  Ke1-c3   H*c7-d5/*d5-e6/Hd5-f4
3.  Kc3-e3   Hf4-e6/*e6-d5/He6-c7
4.  He5-e4    Hc7-d5/H*d5-c6
5.  Ke3-c2/He4-e2   H*c6-g4   Goal-in-one
« Last Edit: Apr 26th, 2009, 11:40am by Adanac » IP Logged


christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #79 on: Apr 26th, 2009, 12:08pm »

on Apr 26th, 2009, 11:37am, Adanac wrote:
I examined these scenarios before seeing Christian's new rule regarding nets. The key difference is that Horses can now win against Elephants and Horse versus Horse is much easier. Otherwise, the same analysis applies.

Taking your much appreciated analysis into account the preliminary result is:
Attacker Defender result
1. Elephant Elephant
draw
2. Horse Horse
win
3. Horse Elephant
win
4. Elephant Horse
?
5. Lion Lion
draw

 
Thanks for all the work Smiley My mind is poorly wired for deduction Undecided
 
Note: The Zillions game is available in a bèta-0.3 version.
(Implemented by Ed van Zon)
download HanniBall (beta 0.3) from the rules page
download HanniBall (beta 0.3) directly
 
« Last Edit: Apr 26th, 2009, 1:13pm by christianF » IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #80 on: Apr 27th, 2009, 2:20am »

I'm a poor analyst, but here's how far I came with the Elephant versus Horse:
 

diagram 1
It's not too difficult to get the ball to h2 and the Elephant next to it, say h3.
It's also clear that the Keeper on f1 and the Horse on g2 are the only defense.
 

diagram 2-1
If Black moves ... Eh32/E*h2i4, both the Keeper and the Horse are pinned.
 

diagram 3-1
If the Horse moves, it must end on h3 to prevent the Elephant from reaching g2 with the ball in two steps, and score. However this triggers ... E*i4f5 and Black can score next move.
 
So white can't move either piece, but then, he doesn't have to, does he?
A player may make up to 3 'moves' per turn, which up to now included 'zero moves'.
Of course this endgame probably isn't the most frequent to appear, and I don't consider it a big problem that its outcome hinges on the possibility of passing. However ...
 
Guilty as charged, again Embarassed
However, it may be indicative of a more general problem concerning passing on one's turn. I have therefore changed the move rule to "On his turn a player is allowed to make up to 3 (4, 1-6) moves, making at least one change".
 
Taking this change of the move rule into account the preliminary result of the one-on-one endgames is:
Attacker Defender result
1. Elephant Elephant
draw
2. Horse Horse
win
3. Horse Elephant
win
4. Elephant Horse
win
5. Lion Lion
draw

 
The next update of HanniBall for Zillions (beta 0.3 download) will have the rule implemented.
« Last Edit: Apr 27th, 2009, 9:05am by christianF » IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #81 on: Apr 28th, 2009, 9:23am »

To revert to what this was all about, here's the content of my original mail to Ed van Zon, Arty Sandler and Benedikt Rosenau, sent April 7:
Quote:
Hoi Ed, Hi Arty, Benedict,  
 
So this is what I dreamed up yesterday evening and this evening. No tries, straight from the mind.
 
HANNIBALL (board attached)
 
Knights move as in Chess, but cannot jump, so if both squares in between are occupied, they can't get to a target square.
Elephants move like the king in Chess.
Lions combine the above.
 
Knights kick the ball using the king's move.
Elephants kick the ball using the knight's move - here a jump.
Lions combine the above.
 
The goalkeeper moves as a knight (no jump) or queen, but is restricted to the goal and the goal area. It kicks a ball up to 5 squares straight or diagonal.
 
A turn consists of up to 3 moves distributed over pieces (whether or not carrying the ball) or kicking the ball (to any square, vacant or occupied by either). Moves may be with different pieces or the same.
A piece with the ball can also choose to move and leave the ball behind.
 
If on a player's turn, an opponent's piece has the ball, then this piece can be captured chesswise and the captured piece is removed, while the captor gets the ball. It is permitted to kick the ball to an opponent and next capture the piece.
 
If the goalkeeper gets a ball kicked to it by an opponent's piece, then the ball ricochets off up to five squares in a straight line, direction and distance being decided by the keeper.
 
If the goalkeeper gets a ball kicked to it by afriendly piece, then the ball is catched normally, and the keeper can, if moves are left, kick it or move with it (or without it).
 
The game starts with a swap. One player makes up to three moves, the other chooses which side he'll play.
 
Result of an hour's listening to whispering Smiley
 
cheers, christian

 
So how much has changed since?
 
major
1. The 'ball to the keeper' rule has been generalized, courtesy of JDB, to apply to both keepers the same way.
2. Obstruction has been recognized as possible, and defined by Adanac, and it has been made a 'red card' offense.
3. There are now 3 variants, depending on the number of moves allowed per turn - this is not a rule change however.
 
minor
1. Pieces other than the Keeper have temporarily been banned from the goal. This restriction has been lifted again.
2. After any turn, there must be at least one change in the position.
3. The Keepers shot & ricochet range was extended from 5 to 6 squares.
 
That was all. For now we haven't seen "a whole lot of rule adaptions induced by playtesting" as was the prediction of some of the more sceptical posters.
Now I'm curious about the game's behaviour because my prediction was and is that it will behave properly. And it is this prediction that is at stake here.
 
So, did anyone try the Zillions Game? (download HanniBall bèta 0.3)
 
P.S. Maybe it is worth noting that HanniBall has an unusual theme and unusual mechanics. It's not as easy to predict it's behaviour as, say, Grand Chess, which is basically Chess, or Dameo, which builds on an established and well tested framework. So I'm not betting on the safe side with this one.
 
Nor is it in the same league as Grand Chess and Dameo. A game idea develops on its own terms, and HanniBall turned out to be a tactical game with a loose and largely unknown strategical framework, that might solidify somewhat over time, if indeed it is played enough for that, but hardly enough to qualify as a 'strategy game' by mindsports standards (see point 4 of the 'acknowledgement').
 
« Last Edit: Apr 29th, 2009, 2:02am by christianF » IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #82 on: Apr 30th, 2009, 3:04am »

You know the Zillions Machine:
"It doesn't feel pity, or remorse, or fear, and it absolutely will not stop. Ever. Until you are dead."
 
So I'm only too glad I'm facing a bèta version here. No swap. It's by mail, because I'm on a mac.
It gets 45 sec per turn (actually 15 per move, and it works up to seven ply each time).

Zillions (bèta) - Christian
position after black-1

(download Zillions HanniBall bèta 0.3)

1. Hh4-f5/Hf4-g6/Hb4-d5
 
* = ball
rules

Hh14-f10/Hb14-d13
 
^ = ricochet
history

 
 
« Last Edit: Apr 30th, 2009, 3:47am by christianF » IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #83 on: Apr 30th, 2009, 3:41am »


Zillions (bèta) - Christian
position after black-2

(download Zillions HanniBall bèta 0.3)

1. Hh4-f5/Hf4-g6/Hb4-d5
2. Hf5-f7/Hd4-c6
 
* = ball
rules

Hh14-f10/Hb14-d13
Eh15-f13/Hd14-e12
 
^ = ricochet
history

 
 
« Last Edit: Apr 30th, 2009, 3:48am by christianF » IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #84 on: Apr 30th, 2009, 5:51am »

Zillions prefers defensive play, which is good inasfar as the merits thereof were a discussion point.

Zillions (bèta) - Christian
position after black-3

(download Zillions HanniBall bèta 0.3)

1. Hh4-f5/Hf4-g6/Hb4-d5
2. Hf5-f7/Hd4-c6
3. Hf7-e9/H*e9-c5
 
* = ball
rules

Hh14-f10/Hb14-d13
Eh15-f13/Hd14-e12
Lg16-e11
 
^ = ricochet
history

 
 
IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #85 on: Apr 30th, 2009, 6:53am »


Zillions (bèta) - Christian
position after black-4

(download Zillions HanniBall bèta 0.3)

1. Hh4-f5/Hf4-g6/Hb4-d5
2. Hf5-f7/Hd4-c6
3. Hf7-e9/H*e9-c5
4. Lc2-d4/*c5-d4/L*d4-c4
 
* = ball
rules

Hh14-f10/Hb14-d13
Eh15-f13/Hd14-e12
Lg16-e11
Hf10-e8/He12-d10/Ed15-c14
 
^ = ricochet
history
IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #86 on: Apr 30th, 2009, 7:54am »

The little sneak is stealing my ideas Angry Time to keep an eye on every piece that can get the ball from the Keeper, in particular my own. Meanwhile I'll try 'coordinated progress'.  

Zillions (bèta) - Christian
position after black-5

(download Zillions HanniBall bèta 0.3)

1. Hh4-f5/Hf4-g6/Hb4-d5
2. Hf5-f7/Hd4-c6
3. Hf7-e9/H*e9-c5
4. Lc2-d4/*c5-d4/L*d4-c4
5. Hc5-d7/*c4-e3/Ke1-e3
 
* = ball
rules

Hh14-f10/Hb14-d13
Eh15-f13/Hd14-e12
Lg16-e11
Hf10-e8/He12-d10/Ed15-c14
Hf14-f10/Ec14-b13
 
^ = ricochet
history
IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #87 on: Apr 30th, 2009, 8:36am »

I could capture the Horse of course, but I don't like the exchange. My provisional plan is to have a strong midfield first.
 
For the record: I spotted a possible problem reminiscent of actual soccer: prolongued possession of the ball by the Keeper should not be allowed.

Zillions (bèta) - Christian
position after black-6

(download Zillions HanniBall bèta 0.3)

1. Hh4-f5/Hf4-g6/Hb4-d5
2. Hf5-f7/Hd4-c6
3. Hf7-e9/H*e9-c5
4. Lc2-d4/*c5-d4/L*d4-c4
5. Hc5-d7/*c4-e3/Ke1-e3
6. *e3-b6/Lc4-b6/*b6-c6
 
* = ball
rules

Hh14-f10/Hb14-d13
Eh15-f13/Hd14-e12
Lg16-e11
Hf10-e8/He12-d10/Ed15-c14
Hf14-f10/Ec14-b13
Eb13-b10
 
^ = ricochet
history
« Last Edit: Apr 30th, 2009, 9:25am by christianF » IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #88 on: Apr 30th, 2009, 9:45am »

To and fro with the ball. So much for bot coordination.

Zillions (bèta) - Christian
position after black-7

(download Zillions HanniBall bèta 0.3)

1. Hh4-f5/Hf4-g6/Hb4-d5
2. Hf5-f7/Hd4-c6
3. Hf7-e9/H*e9-c5
4. Lc2-d4/*c5-d4/L*d4-c4
5. Hc5-d7/*c4-e3/Ke1-e3
6. *e3-b6/Lc4-b6/*b6-c6
7. *c6-b6/Hd5-c7/*b6-c6
 
* = ball
rules

Hh14-f10/Hb14-d13
Eh15-f13/Hd14-e12
Lg16-e11
Hf10-e8/He12-d10/Ed15-c14
Hf14-f10/Ec14-b13
Eb13-b10
Hd13-c10/Lc16-d14
 
^ = ricochet
history

« Last Edit: Apr 30th, 2009, 9:45am by christianF » IP Logged
christianF
Forum Moderator
Forum Guru
*****



Arimaa player #4019

   


Gender: male
Posts: 804
Re: Essay by Christian Freeling on inventing games
« Reply #89 on: Apr 30th, 2009, 10:49am »

So I got the midfield covered Tongue

Zillions (bèta) - Christian
position after black-8

(download Zillions HanniBall bèta 0.3)

1. Hh4-f5/Hf4-g6/Hb4-d5
2. Hf5-f7/Hd4-c6
3. Hf7-e9/H*e9-c5
4. Lc2-d4/*c5-d4/L*d4-c4
5. Hc5-d7/*c4-e3/Ke1-e3
6. *e3-b6/Lc4-b6/*b6-c6
7. *c6-b6/Hd5-c7/*b6-c6
8. *c6-b6/Hc6-b8/L*b6-a6
 
* = ball
rules

Hh14-f10/Hb14-d13
Eh15-f13/Hd14-e12
Lg16-e11
Hf10-e8/He12-d10/Ed15-c14
Hf14-f10/Ec14-b13
Eb13-b10
Hd13-c10/Lc16-d14
Ef13-d11/Le11-d9
 
^ = ricochet
history
IP Logged
Pages: 1 ... 4 5 6 7 8  ...  78 Notify of replies Notify of replies Send Topic Send Topic Print Print

« Previous topic | Next topic »

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