ID:41582
 
I once played a game that was based on the rock-paper-scissors concept, but taking it to the next level and turning it into a board game. It was simple, a series of rock, paper and scissors pieces laid out in formation on a checker board. A rock could take scissors, scissors could take paper, and paper took rock. You could move one piece per turn, and once you'd taken all the other player's pieces, you won.

So I thought, "Hey, why not take that idea and expand on it further?" In Chess, you don't have to be troubled with hunting down all of the other player's pieces because the only piece that really matters is the King. So what if we added a "king" piece to this rock-paper-scissors game? Perhaps the "king" can take any piece, but can also be taken by any other piece in the game. If you lose your "king", your game is over.

Rock-Paper-Scissors board layout with King pieces!

Adding a king piece would allow for some more strategy, faster games and less "wrap it up after you've already won" games, but it would still not give it a whole lot of extra depth, since the primary gameplay would still be the same. So lets add another element: taking multiple pieces at once.

(Additionally, in order to make sure that there is an even number for each of the rock, paper and scissor pieces, we're going to add a "defender" or "queen" piece in front of the "king", which acts the same way that the "king" does. It can take any piece, and be taken by any piece.)

Since the game will take place on a large checkers-style board, with more than enough empty tiles, we'll let the players take advantage of those empty spaces in their strategy. When your piece takes one of your opponent's pieces, your piece will take the other piece's spot. Now, if from that spot your piece can take another one of your opponent's pieces, then it can take that piece in the same turn. This can continue until your piece reaches a tile from which none of your opponent's pieces are vulnerable.

An example of how to take multiple pieces in one turn.

That means players can use sneaky combination attacks on their opponents. For example, lets suppose that you're the red player in the depiction above. Notice the paper where the green arrow starts. Using that piece of paper, you could take the neighboring rock. That would result in your opponent using his or her scissors to eliminate your paper, since it poses a threat to your opponent's "king". It would be a 1-for-1 deal, with neither player gaining any real advantage from it.

However, since your opponent has two rocks lined up, your piece of paper can take both the adjacent rock, and once it has taken that rock's spot, it can then take the next rock down the line as well. Plus, it is now next to your opponent's "king", which means that from three tiles away you can take your opponent's "king" and win the game. 3-for-0. A much better deal.

That new rule will add some more strategic depth to this game beyond the "my piece takes your piece, your piece takes my piece" mantra that it would otherwise consist of. It might be good to throw in a few additional rules, such as allowing any piece to move two tiles on its first move. Its just a matter of play-testing it out to see which way is more strategic and fun.
The only problem I see is that there are only five paper pieces on either side, with six rocks and six scissors. In that regard, the major piece of the game is the scissors piece, as it outnumbers the paper pieces and is kept in check (not surpassed) by the rocks. (Actually, thinking on this more, the rock is perhaps the better piece, since it matches against the scissors and has few natural enemies.)

How about adding yet another piece: a wild "queen". The queen can take any piece if she attacks, but can also be taken by any piece if she is landed on. She takes on the quality of the tool used to destroy the first piece, so the queen can kill two papers or two scissors in a row but cannot kill a paper and a scissors in a row (alternately: she takes on the nature of the tool she destroys, so she could kill a paper, then kill a rock, then kill a scissors in a row).

Then the board could be 10x10, and you could have 20 pieces to a side, consisting evenly of six papers, six scissors, six rocks, a king, and a queen.
The "queen" piece could also be just the equivalent of a "king" piece but without the "game over" prospects if lost. Basically just a simple takes anything, can be taken by anything setup, which means it would be taken easily out in the field but is useful for guarding the "king" piece back at home.

Still, if you wanted to change the theme to something besides Rock-Paper-Scissors meets Chess, you'd have to come up with some cool things that would fittingly represent a "king" AND a "queen".
Foomer wrote:
The "queen" piece could also be just the equivalent of a "king" piece but without the "game over" prospects if lost. Basically just a simple takes anything, can be taken by anything setup, which means it would be taken easily out in the field but is useful for guarding the "king" piece back at home.

That's what I meant; she'd take on properties for the given turn only, then revert to being wild for the opponent's turn (so she could be taken by anything) and her next turn. I was referring to combo attacks only.
Jtgibson wrote:
That's what I meant; she'd take on properties for the given turn only, then revert to being wild for the opponent's turn (so she could be taken by anything) and her next turn. I was referring to combo attacks only.

Oh I see what you mean. That's a good point, and it should apply to the "king" piece as well, otherwise the "king" will end up wiping out half the board in one turn.
Jtgibson wrote:
The only problem I see is that there are only five paper pieces on either side, with six rocks and six scissors. In that regard, the major piece of the game is the scissors piece, as it outnumbers the paper pieces and is kept in check (not surpassed) by the rocks. (Actually, thinking on this more, the rock is perhaps the better piece, since it matches against the scissors and has few natural enemies.)

How about adding yet another piece: a wild "queen". The queen can take any piece if she attacks, but can also be taken by any piece if she is landed on. She takes on the quality of the tool used to destroy the first piece, so the queen can kill two papers or two scissors in a row but cannot kill a paper and a scissors in a row (alternately: she takes on the nature of the tool she destroys, so she could kill a paper, then kill a rock, then kill a scissors in a row).

Then the board could be 10x10, and you could have 20 pieces to a side, consisting evenly of six papers, six scissors, six rocks, a king, and a queen.

The problem with this is that it still leaves one of them three (rock/paper/scissors) at an advantage. You either have to have two queens or wild card pieces, or change the layout on the board to allow for removing/adding the extra pieces. The thing about Rock-Paper-Scissors is the equal balance between winning and losing between the three, and that is a concept I'd assume you want to keep in a game based around Rock-Paper-Scissors.
The game could also be played on a 7x7 board instead of a 9x9 or 10x10 board. It would put the pieces closer together, but there would be equal numbers of all pieces.

http://www.byond.com/members/Foomer/files/images/rps3.png

(The shield in this picture represents what I'm calling a "defender" instead of a "queen", since its basically only good for protecting the "king" piece.)

Having less space to travel across the board might be better, since unlike Chess, this game doesn't have any pieces that can move across the board in one turn.

Still, there's no reason why the formation couldn't have more than one defender or queen piece, if the board size calls for it.
If your interested, I could try creating this in Law of the Board of mine? :p
Acebloke wrote:
If your interested, I could try creating this in Law of the Board of mine? :p

If you want. It couldn't hurt to have a proof of concept version.

Pretty cool concept, although paper seems a bit outclassed with only 5 vs. the 6 of everything else. Perhaps this would work better on a traditional 8x8 board since that gives you room for 5 of each piece plus a king.

I would despair of ever coming up with a good AI for this game though.
Lummox JR wrote:
Pretty cool concept, although paper seems a bit outclassed with only 5 vs. the 6 of everything else. Perhaps this would work better on a traditional 8x8 board since that gives you room for 5 of each piece plus a king.

I was actually considering a 7x7 board as seen here.


I would despair of ever coming up with a good AI for this game though.

I don't imagine AI would be that difficult since you could just loop through each of the computer's pieces, check all of its possible moves, search for any combos that can be made with those moves, and then assign a score for each move based on:
- Whether or not that move will take the opponent's piece.
- Whether or not that move will put its piece in jeopardy.
- How close this move will place it to pieces which it can take.
- How close this move will place it next to pieces which can take it.
- Whether it will take it closer to or further from the opponent's "king".
- Whether or not it will be able to do any combos from that move.

Something like that anyway. It should work similar to AI for a checkers game, so I guess it'd just required some research into how that kind of AI works.
There, I adjusted the images and accompanying text a bit to fit the most up-to-date concept.
On the old brunching.com site there was a Flash game called Rochambeau Run that was a puzzle game. You were trying to move a little figurine toward a coffeee mug, while any scissors, paper, or rocks in your path would try to move toward you. The game required you to use walls to your advantage and also trick the pieces into attacking each other. For example if scissors headed toward you and then lined up with paper, they'd head over for the paper instead. But if scissors headed toward you and lined up with a rock, the rock would get the scissors and then it would head toward you. While pieces were on the move, you had to stay still, so it wasn't a matter of timing. It was a very clever game, though short.
Yeah I can imagine making levels for a game like that would be a major headache. That's what keeps me away from doing puzzle games of that nature.
For puzzle games I prefer if there's some way to generate a puzzle. For a game like that, I think it might possibly be doable to create a generator. It's not an easy project of course, but it'd be interesting to attempt. I think the way to do it would be to work backwards from the goal to a starting point, inserting the killer objects as you went along.
I'd thought about doing something like that in a puzzle game idea that I was bouncing around in my head, something along the lines of Theseus and the Minotaur. Only, instead of walking toward the exit, you walk toward the starting point, and instead of walking toward you, the minotaur would walk away from you. And you wouldn't be allowed to move in a direction if the minotaur couldn't step back from there.

For most puzzle games though, if they have a lot of different elements instead of just one big element to work with, its not really feasible.
I'm thinking as far as AI goes for this game, you should be able to use A* to have each piece forge a path from its location to the opponent's king, which is the ultimate goal. Each possible move would be given a weighted value based on whether that move would place the piece in jeopardy, whether another piece blocks movement to there, and whether it would take an opponent's piece by moving there. Also, proximity to piece to which this piece is vulnerable would be a factor. Scissors would want to move further away from rocks, not closer.

We'd also have to consider whether moving to a certain point would allow the piece to use a combo, and if so where could it end up?

Once each piece has determined its ideal path, then we just need to determine which piece has the better next step based on the value that the piece's first step was given in the initial calculation. The piece with the most valuable move gets to go. Unless there's a tie, then we'd just pick one randomly.

Assuming this works out to create a competitive AI, we could then dumb it down by either having giving the AI a percent chance of skipping the best move and moving onto the next (and the same chance of skipping the next move and going to the move after that...), or we could just have the pieces randomly ignore things such as moving into a vulnerable position or skipping a move where they could take an enemy piece.