Author |
Topic: Essay by Christian Freeling on inventing games (Read 522343 times) |
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender: 
Posts: 804
|
 |
Re: Essay by Christian Freeling on inventing games
« Reply #645 on: Aug 16th, 2011, 9:38am » |
|
on Aug 16th, 2011, 7:40am, christianF wrote:| Symple © Mindsports / Benedikt Rosenau |
|
Benedikt has agreed to my using our recent correspondence, so I'll start as recently as yesterday with part of his mail:
Quote:Benedikt:
Why people do not see the greatness of Symple or Sygo? That question has lingered long in me. Last year, there was the umpteenth challenge to design a game of type Bla, with an award going to the best one.
What happened? Sure enough, many games were submitted, and one got the price. The jurors, however, complained at a certain stage how difficult it is too assess a game. It takes much more time than inventing one.
That was the clue we have missed. Designing a game is not easier than assessing it, but, averaging over the designers, it takes less time. |
|
Quote:christian:
Except for Sygo all games I made in the last run were ignored. And I had to work for Sygo. People can't appreciate a game without playtesting, and great games elude playtesters far more easily than simple tactical ones.
Actually I see Grand Chess, Dameo, Emergo, Sygo and Havannah as the games that matter in te long run. The rest is ornamental. But to tie them to my 'brandname', the ornamental ones are important because of their number and the overall fair quality.
"This guy knew what he was doing", that's the message they convey (hopefully).
There's no alternative either. Tactical games can get more or less immediate acclaim, strategy games ... well, hope springs eternal. |
|
Quote:Benedikt:
I checked About Symple again. It is too much "tidbits about Symple". I think it is better to say something about gameplay there and move most of the current stuff to a Further Reading section.
So, what do we know about Symple? It is a territory game with a major overlap to Go. If the reader knows Go, the approach of taking strategically important points and extending from them is similar. The similarity extends to the places where the points are taken and the order in which it is done: corner, edge, center.
Furthermore, there seems to be a basic order: first place stones, then grow. At least it helps to approach the first games that way.
The rule is not cut in stone. There may be exceptions to this for two reasons:
1. the balancing rule - obvious, but it should be expanded on, and
2. urgently needed counter-moves.
However, if the latter happens, it might indicate less than ideal moves by either player so far.
Growth phase: Here, your groups comepete with the opponents groups for the free space. Walling in own territory and cutting off opponent groups from entering them appears similar Go. It is good to place your stones in the opening in such a way that you are quicker and greedier than your opponent at this walling in of free territory. The square grid and cross cuts allow to cut off the opponent without having to connect.
Depending on the group penalty setting, the tactics differ in how one blocks an opponent who is worming into your territory.
Endgames: here drops reappear. If you can grow less than a dropped stone can take, then the drop is clearly the right decision. However, drops might be worthwhile before. The mathematics are interesting, but not formally solved.
Sygo: is largely similar, BUT groups do not just live. That makes for much more tactics which in turn re-influence the strategy: groups do not just live, they can die, incurring a high cost. Actually, groups seem to have a harder time living than in Go (meaning they need more territory or quicker connections). This applies to both players, so mutual fights in semeai can appear - all in 10 moves or so. You have written more insightful stuff about Sygo, like that capture does not create eyes. It should go well together with the stuff above.
So much for my sketch. Last night at 4 o'clock (tells you about my sleeping habits), I realised that there is no handicapping mechanism to Symple/SyGo. I actually feel like a minor of Symple co-inventor.
Hence, an idea: Symple might do with a different group penalty. But I bet you will come up with something brilliant for Sygo.
> There's no alternative either. Tactical games can get more or less immediate acclaim, strategy games ... well, hope springs eternal.
The correct answer is to do both. We could sift through the recent inventions and guess at rewarding gameplay. I think we will gain some insights. |
|
Quote:christian:
Symple with a different group penalty seems splendid. All kinds of ratios might be tried to establish their relative value, but of course you need a lot of solid players and dependable ratings. A problem in the long run.
>Depending on the group penalty set, the tactics differ on how one blocks an opponent worming into your own territory.
Yes, that's a truly innovative parameter in the realm of abstract board games, 'scalable tension'.
>2. urgently needed counter-moves.
However, if the latter happens, it might indicate less than ideal moves by either player so far.
Probably, but not necessarily. Tactical attacks may occur in balanced games as well.
Sygo? Let's first say I dislike handicap games. And I see nothing even remotely brilliant, but I'll think about it.
Semeai play a prominent role in Sygo throughout the opening and into the middle game, probably even more so when players get better.
>You have written more insightful stuff about Sygo, like capture not creating eyes. It should go well together with the stuff above.
I've written about Othellonian capture in Go variants and shown that Sygo is the only variant that doesn't need an artificial life condition despite the fact that capture doesn't create eyespace. That in itself should be telling enough. |
|
|
| « Last Edit: Aug 17th, 2011, 3:03am by christianF » |
 IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #646 on: Aug 17th, 2011, 10:48am » |
|
>Depending on the group penalty set, the tactics differ on how one blocks an opponent worming into your own territory.
Yes, that's a truly innovative parameter in the realm of abstract board games, 'scalable tension'.
I stand corrected by coincidence. Luis Bolaños Mures posted a new game at rga and Nick Bentley linked it to one of his called Odd, which turns out to have a 'scalable tension' parameter. Just my luck 
|
|
IP Logged |
|
|
|
MarkSteere
Forum Guru
Finite games rule
Gender:
Posts: 289
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #647 on: Aug 17th, 2011, 11:22am » |
|
on Aug 17th, 2011, 10:48am, christianF wrote:
I give up.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #648 on: Aug 17th, 2011, 12:55pm » |
|
on Aug 17th, 2011, 11:22am, MarkSteere wrote:
Not that I would mind, but it's Odd because it's so Symple:
Note: One can change the character of the game by increasing or decreasing the minimum size a group must have in order to be counted at the end. Try minimum sizes 4, [5], 6 or 7.
Odd
Note: This parameter P is central to the theme. The applet allows it to be set at 2, 4, 6 ... 32, but values between 4 and 12 would seem to give the most interesting play. The nature of the game's strategy changes with the choice of the parameter.
Symple
|
|
IP Logged |
|
|
|
NickBentley
Forum Senior Member
Arimaa player #6704
Gender:
Posts: 44
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #649 on: Aug 17th, 2011, 4:06pm » |
|
Doesn't *Star also have a similar thing with its group penalty?
http://www.boardgamegeek.com/thread/404644/re-worded-rules
Each player gets -2 points for each group he has on the board. I think Ea Ea, the designer, wrote somewhere about changing this number (natch, I can't now find it)
Speaking of which, I've never really understood all the weird esoteric details of *Star. I've played a version where you simply get one point for every edge space occupied and -2 points for every group on the board, regardless of whether it touches an edge, and it played great. In fact I enjoyed this simple version more than the original.
|
| « Last Edit: Aug 17th, 2011, 4:47pm by NickBentley » |
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #650 on: Aug 17th, 2011, 5:07pm » |
|
on Aug 17th, 2011, 4:06pm, NickBentley wrote:Doesn't *Star also have a similar thing with its group penalty?
http://www.boardgamegeek.com/thread/404644/re-worded-rules
Each player gets -2 points for each group he has on the board.
Speaking of which, I've never really understand all the weird esoteric details of *star. I've played a version where you simply get one point for every edge space occupied and -2 points for every group on the board, regardless of whether it touches an edge, and it played great. In fact I enjoyed this simple version more than the original. |
|
Imo. it was all meddling with a concept that was falsely perceived to be about "groups with certain connections" with a group penalty, Star, *Star, Superstar and YvY alike. This thematic hierarchy of penalizing a set of groups satisfying certain connections kept all these games from reaching the essence of either the group penalty or the connection theme. The latter already knows Y and Hex as quintessential implementations, so at that end it's no surprise. At the 'group penalty' end Symple emerged as the quintessential implementation by abandoning the connection theme ad simply penalizing 'groups'.
That made the penalty the central parameter and making it variable was an inevitable step, although it took me some time to realize that.
In the other games it was set at 2 or 4 because it could hardly be recognized as the essential variable, the focus being on the connection aspect. The arbitrariness of it was one of the nagging thoughts that made me feel I missed something, the same feeling Benedikt conveyed in his mail that started the whole thing.
So yes, you could make the penalty variable in Star et all, but it would be the wrong environment, I feel, and the connection aspect would get in the way of clarity.
|
|
IP Logged |
|
|
|
GregS
Forum Newbie
Arimaa player #6740
Gender:
Posts: 1
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #651 on: Aug 19th, 2011, 8:15am » |
|
on Jun 27th, 2011, 12:50pm, MarkSteere wrote:
The "asymmetry" you're referring to is more of a normal thing than a bad thing. Possibly even a good thing.
I hate to see a designer bowled over by program data. And I don't even think Greg was trying to bowl you over. You're bowling yourself over at his expense.
|
|
(I realize I'm coming in late on this thread)
I'd just like to mention that I had not planned on performing any computer analysis of Grabber until Christian noticed that during computer vs. computer self-play, the 2nd player was predominantly winning. At that point, we suspected a bug in the program. I searched high and low and could not find any problems and began to suspect that the behavior was inherent in the game. That in turn led to a number of experiments in trying to solve (i.e. exhaustively search) various sized games. The 6x6 indicated a win for 2nd. Solving a game through exhaustive search should remove the AI from the equation and other artifacts such as the order in which the moves are generated. Having done so should also raise the confidence that the behavior is due to the game itself, not it's implementation. Still, there's a nonzero probability that the results are off (and the last thing I would want to do is to draw faulty conclusions about someone else's game based on my own error) , however I've run out of further avenues to explore. About the only thing I can think of at the moment, would be for someone else to independently program the game and attempt to confirm or refute my results.
-- Greg
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #652 on: Aug 19th, 2011, 10:27am » |
|
on Aug 19th, 2011, 8:15am, GregS wrote:| The 6x6 indicated a win for 2nd. Solving a game through exhaustive search should remove the AI from the equation and other artifacts such as the order in which the moves are generated. Having done so should also raise the confidence that the behavior is due to the game itself, not it's implementation. Still, there's a nonzero probability that the results are off (and the last thing I would want to do is to draw faulty conclusions about someone else's game based on my own error), however I've run out of further avenues to explore. |
|
Thank you for running the tests. So 6x6 is a second player win. I would have been very surprised indeed if it had been the other way around. Don't worry about faulty conclusions. I'm convinced it's the game. What I don't understand is the why. It seems counterintuitive considering the tempo switches that occur (a simple choice between single or multiple capture with the same piece may imply such a switch).
So it's good to realize one's intuition may be totally off the mark.
on Aug 19th, 2011, 8:15am, GregS wrote:| About the only thing I can think of at the moment, would be for someone else to independently program the game and attempt to confirm or refute my results. |
|
I fear the result would be the same. Thanks again for your efforts, I'll include the provisional conclusions in the Grabber section at mindsports.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #653 on: Aug 22nd, 2011, 5:58am » |
|
How about that, another contest. It's called
and it's organized by Daniel Solis who has quite a few games to his credit himself, this challenge listed among them.
The challenge isn't limited to abstracts, in fact it is:
Quote:| "Create a game. The game can be of any theme or genre you desire, but there is one restriction: You're creating a "new classic," like Chess, Tag or card games. So, create a game to be enjoyed by generations of players for a thousand years." |
|
I've entered Sygo and added a game Ed and I were playing at the time as an example game. The game has finished now.
Entries are limited to one per contestant and Mark entered Flume which I consider a good choice.
Mark didn't fail to mention that all his games are made for the centuries and that all the other contestants' games may as well be disregarded:
Quote:| "I haven't looked at the other games in the contest, but, that being said, I'm sure they will all be obsolete in well under 100 years. Never mind 1000." |
|
Which should make it easy for Daniel and his wife  .
P.S. I'm glad to see that Nick Bentley entered his game Ketchup. I'd also like to see Luis Bolaños Mures' new game Yodd in the contest, because it is imo. a connection game of some significance, and some independent significance at that.
|
| « Last Edit: Aug 22nd, 2011, 10:40am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Starting with Black, players take turns placing on
« Reply #654 on: Aug 22nd, 2011, 11:25am » |
|
on Aug 22nd, 2011, 5:58am, christianF wrote:| I'd also like to see Luis Bolaños Mures' new game Yodd in the contest, because it is imo. a connection game of some significance, and some independent significance at that. |
|
Why beat around the bush. Yodd is played on a hexhex board. Luis mentioned that base-5 would probably be too small so how about this:
 .
Now the beauty. How's this for simplicity, not to mention originality:
Rules
* Starting with Black, players take turns placing one or two stones of any color on empty cells. On his first turn, Black can only place one stone.
* At the end of each turn, there must be an odd number of groups on the board.
* Players can pass their turn at any moment, unless it violates the previous rule (this means Black can't pass on his first turn).
* When both players pass in succession, the game ends. The player with less groups on the board wins. Draws are not possible.
Note that there's no actual counting involved in the 'odd # of groups' rule: just keep it odd. I'll leave it at that, so as not to spoil anyone's fun in trying to figure this one out.
P.S. I'm glad to say that Luis did indeed enter Yodd in the contest.
|
| « Last Edit: Aug 23rd, 2011, 3:44pm by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #656 on: Aug 22nd, 2011, 1:17pm » |
|
on Aug 22nd, 2011, 12:44pm, NickBentley wrote:
I can't comment on games outside my conceptual framework, but next to the ones mentioned earlier, I particularly like Cartography by Benjamin Mohr.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #657 on: Aug 24th, 2011, 10:55am » |
|
on Aug 22nd, 2011, 11:25am, christianF wrote:| P.S. I'm glad to say that Luis did indeed enter Yodd in the contest. |
|
Stavropol Checkers are checkers or colomn checkers variants where players on their turn are allowed to move with either side. You may move opponent's pieces in the same way as he or she would be allowed to move. It's a game that in my experience induces acute schizophrenia. I hadn't seen this protocol in any other game, till now.
Misere Checkers, or 'give-away checkers' are variants whereby the object is reversed, leading to a dramatically different strategy. Not my cup of tea either.
So if someone had told me about a game that uses the Stavropol protocol (and even extending it to moving with both colors in the same turn), and that has a misere complement (i.e. playing for the most groups) that is in essence identical to the original, I guess I'd have had a hard time believing that.
Yet when it happened it didn't take all that much time (though definitely some time) to grasp it.
On top of that it is arguably the quintessential 'dynamic goal non-parity connection game', so I think we got an important game here.
We'll implement it at mindsports as soon as possible. The rules for Yodd are already present. Note that Luis also provided rules for the square version, called Xodd.
|
| « Last Edit: Aug 24th, 2011, 4:17pm by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #658 on: Aug 26th, 2011, 9:10am » |
|
These are interesting times, both in the Chinese notion (I've been told that "may you live in interesting times" is a Chinese curse) and in the realm of inventing abstract games, in particular quintessential ones.
Quintessential
Benedikt Rosenau recently accused me of coining the term ("An insightful term you introduced to ludology"), but I'm not sure I did. Acually I'm still wrestling with its meaning because its less than sharply contoured. Let's try a loose definition.
Quote:| A game is quintessential if it is the is the most basic implementation of a chosen theme and a chosen mechanism. Such a combination may or may not require a particular grid or a particular boardshape. These requirements should not be too exotic. |
|
Let quintessential in the following mean arguably quintessential, but think in terms of Occam's Razor: all rules should be necessary and sufficient.
Although quintessential games are not necessarily the most rewarding to play, they are always interesting and they have a way of surviving all trends and hypes. Here's a list with examples of what I consider quintessential games. Feel free to make further suggestions or to argue against any of the ones listed. Note that "checkmate" isn't a theme that lends itself to a quintessential implementation, whatever the mechanism.
Elimination
Checkers has elimination as its main theme, but cannot do without the "no legal move" safety net. Its mechanism could be made simpler (by making it both single step and single jump) but time shapes games to give maximum reward in play, not to make them quintessential, and this little deviation of 'most basic' doesn't count (say I ) especially in the light of descendants like International Checkers or Shashki, that are about as far away from quintessential as it gets.
Focus, invented by Sid Sackson, is quintessential for the mechanism involved. However, the square grid board is effectively a 5-story high octagon which may be considered rather exotic.
Oust, invented by Mark Steere, has a most elegant mechanism of placement capture, equally suited for the square and the hexgrid, independent of size or shape of the board. The rules are 'necessary and sufficient' and the game is drawless and finite by nature. A perfect mechanism based example.
Emergo, invented yours truly. What can I say? There are five more or less well known games that employ the 'column checkers' mecanism. Bashni is based on the Russian checkers variant Shashki, Lasca is based on Checkers, Stapeldammen is based on International Draughts, Grabber is based on Konane. Emergo is based on nothing but the mechanism and the theme.
Territory
Go is the quintessential game of single placement and capture by eclosure. It is uniquely suited for the square grid. It's rules are self explanatory and so is its main problem: cycles. Rules to prevent cycles are unwanted necessities, but necessities just the same. It doesn't make a game less quintessential, but it does make it less elegant.
Reversi/Othello is the quintessential game of single placement and 'custodian capture by reversal'. A quintessential game with 'capture by enclosure and reversal' doesn't appear to exist, although Sygo comes near.
Dots & Boxes, first published in 1889 by Édouard Lucas (of the series), and Flume, invented by Mark Steere, may both be considered quintessential, using the same theme in combination with a mechanism that is named after the first, but more implicitly and elegantly implemented in the second.
Symple, invented by yours truly, is quintessential for a territorial theme based on 'group penalty': awarding points to every stone, but substracting a penalty for every group. The accompanying mechanism is an inherent move protocol named after the game.
Its descendant Sygo comes very close to being a quintessential territory game of 'capture by enclosure and reversal', but it combines this mechanism with Symple's move protocol, which, however seamlessly, makes it more of a lucky merger than a quintessential game. It relates to Symple much the same way Havannah relates to Hex.
Connection
Hex (Piet Hein, John Forbes Nash) and Y (Claude Shannon, Craige Schensted, Charles Titus) are quintessential connection games for their specific boards. In contrast, Havannah is a lucky merger. The quintessensial connection game for the hexhex board is as yet elusive, and the square board has a load of connection games, but none of them quintessential.
With one notable exception that comes next.
Xodd and Yodd, invented by Luis Bolaños Mures, are also quintessential connection games for their specific boards. It's the same game square or hex, independent of size or shape (don't get finicky on me here). It's theme is 'group count' and the freedom of it's 1-2-2 ... move protocol is astonishing, the only restriction being that the total count be odd at all times.
Breakthrough/Race
In this theme Halma, invented in 1883 or 1884 by George Howard Monks, may be considered quintessential, as may games like Dan Troyka's Breakthrough or Robert Abbot's Epaminondas and doubtless a few others.
How many of the above are younger than a century? And how many were invented this century? These are interesting times  .
|
| « Last Edit: Aug 26th, 2011, 9:27am by christianF » |
IP Logged |
|
|
|
NickBentley
Forum Senior Member
Arimaa player #6704
Gender:
Posts: 44
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #659 on: Aug 26th, 2011, 9:42am » |
|
I nominate Slither by Corey Clark (formerly known as Particle Bond) for quintessential connection game on a square board.
http://www.boardgamegeek.com/boardgame/75957/slither
http://files.boardgamegeek.com/file/download/7no8v61gje/Slither_by_Corey _Clark.pdf?
There have been several equally-simple solutions proposed for the cross-cut problem on square boards, but Slither is the only one that does it with in a way that improves the game. It improves it dramatically, IMO.
As a result it's my favorite game of any kind.
It's weird because it's more or less my Game-of-the-Century, and yet it doesn't get any more attention than a billion other games. Me no understand.
|
|
IP Logged |
|
|
|
|
Home
Help
Search
Members
Login
Register
Notify of replies
Send Topic
Print