7 ms·

Rediscovering rules for ancient board games

54 points by elorant 7y ago

sndwnm 7y ago
As more powerful AI computation capabilities become available to hobbyists, I hope someone takes up researching new board games similar to Go, Chess and the other giants of abstract strategy games (ASG). Basically, train really strong AI players using the AlphaZero method with the rules of the game changed by a human author (either some variant of Go, or a completely new game). Then take a look at the games the AI plays at its strongest level and let people decide if the game looks interesting enough for them to try play competitively. With the strong AI player already in place, human players would have something to learn from and a serious competitive culture could be born. In the absence of something like this, I don't think it's surprising the only popular ASGs are very, very old. Modern commercial board games are fine, but they supply a different demand than ASGs.
- Marazan 7y ago
  There are countless Abstract Strategy Games out there.
- kd5bjo 7y ago
  You’re unlikely to see another game very similar to Chess or Go, because they have each evolved over time to be the pinnacle of their particular niche, and there’s really only room for one king of each of these hills. There are newer games that are starting to show similar signs of longevity, though: Scrabble, Poker, and Bridge come to mind. Modern commercial board games and some genres of video game have communities that could continue for a similar length of time, but they’re still too young to have reached a consensus on one or two games; I suspect that’s a few decades out. The good news is that there are some older titles that are still holding their own in the market— to beat out the newcomers hints that there’s some kind of staying power there.
  - disordered 7y ago
    I don't agree with your premise that chess has evolved to be at the pinnacle of its niche (I don't know enough about Go to comment on it). There are a number of interesting variations in chess-like games that, while maintaining the general theme of killing the enemy king, offer aspects of gameplay that chess does not. Two of the more well known variations are shōgi and xiangqi: Shōgi allows the redeployment of captured pieces, whilst xiangqi limits the movement of a few pieces and has a mechanism for representing artillery. I find that each game has a very different character to that of western chess, even despite the fact that they share a large number of tactical motifs. Western chess may be the most popular variant world-wide, but that's no indication that it is at any kind of pinnacle.
    - michaelscott 7y ago
      I think it's fair to call it the pinnacle when talking about the Western world. Both shogi and xiangqi developed from the same base game as chess historically, so I'm not sure how far they venture from the theme compared to something like hnefatafl (which is chess-like but varies considerably in game mechanics when compared with shogi and xiangqi).
      - kd5bjo 7y ago
        Just like Like Go and Poker each have a number of variant rule sets in common use, I mentally group all of these as “chess” based on their common ancestry. I haven’t really studied them enough to know if this is truly justified.
reificator 7y ago
I've been wondering, how closely could a future/alien civilization recreate Magic the Gathering's rules given nothing but a collection that included every card and several (100? 1000?) valid decks for different formats. I imagine they could get fairly close, especially thanks to reminder text. But I wonder how older/erattaed cards or old templating might hinder those efforts. For those unfamiliar with Magic, the comprehensive rules are about 190 pages of 10-20 rules each, plus a glossary/index. On top of that, the golden rule of Magic is that if a card and the rules disagree, the card takes precedence. And there are over 20,000 unique named cards, though a small number are functionally the same card as another with a different name. See this link for the comprehensive rules, but note that the vast, vast majority of games don't require an exhaustive understanding. You can learn enough to play in 10 minutes, though you'll probably need to learn something new each game for a little while. https://magic.wizards.com/en/game-info/gameplay/rules-and-formats/rules https://magic.wizards.com/en/game-info/gameplay/rules-and-fo...
- Marazan 7y ago
  IMO you wouldn't be able to come close. I'm not familiar with how much reminder text modern cards have but surely there is no way you'd be able to work out anything close to the timing rules from just cards alone.
  - reificator 7y ago
    There are plenty of cards that reference sorcery speed, there's a card that lists all the phases in order and has players skip one, there's the Split-Second keyword which says you can't put things on the stack until it finishes resolving. The fact that the game is fundamentally about overwriting its own rules is why I think it might be possible.
  - bsder 7y ago
    Older cards have more expressive text. So, you can probably link modern vocabulary to older vocabulary to explain play and rules. However, there are certain rules that are arbitrary. For example, the fact that resolution order follows a "stack" is specified nowhere on cards.
drdeadringer 7y ago
> the oldest known board game is an Egyptian game that dates back to 3100 BCE called Senet. “We almost never have the rules for these early games,” Browne said. “The rules have never been recorded, so our knowledge is largely based on historian’s reconstructions.” I live near the Rosicrucian Egyptian Museum in San Jose, CA. The group behind it owns the whole square city block, so besides the museum itself there's also a research library, other related buildings, and a nice park area all around. Included in the park space is a reconstruction of the Senet game, along with an approximate reconstruction of the rules; it's explained that the rules are an approximation with a rough reason as to why. Whether the rules were simply "lost to time" or due to a subset of this as "they were never written down because everyone knew them" I'm not sure, but it was nice to see the game available in person.
ArtWomb 7y ago
Ludii was presented at 2019 IEEE Conference on Games this past week in London. Lot's more interesting stuff in the rest of the conference program ;) http://ieee-cog.org/assets/program-full.pdf http://ieee-cog.org/assets/program-full.pdf
huehehue 7y ago
I spent a couple of months last year researching Liubo and constructing/play-testing a crude recreation. Of course, we don't even know what a full game set looks like, much less how it was played. But the result was a decent game, and I like to think there's enough overlap between board game enthusiasts and history nerds that someone could crank a few of these out on Kickstarter.

54 points by elorant 7y ago

As more powerful AI computation capabilities become available to hobbyists, I hope someone takes up researching new board games similar to Go, Chess and the other giants of abstract strategy games (ASG). Basically, train really strong AI players using the AlphaZero method with the rules of the game changed by a human author (either some variant of Go, or a completely new game). Then take a look at the games the AI plays at its strongest level and let people decide if the game looks interesting enough for them to try play competitively. With the strong AI player already in place, human players would have something to learn from and a serious competitive culture could be born. In the absence of something like this, I don't think it's surprising the only popular ASGs are very, very old. Modern commercial board games are fine, but they supply a different demand than ASGs.

Marazan 7y ago
There are countless Abstract Strategy Games out there.
kd5bjo 7y ago
You’re unlikely to see another game very similar to Chess or Go, because they have each evolved over time to be the pinnacle of their particular niche, and there’s really only room for one king of each of these hills. There are newer games that are starting to show similar signs of longevity, though: Scrabble, Poker, and Bridge come to mind. Modern commercial board games and some genres of video game have communities that could continue for a similar length of time, but they’re still too young to have reached a consensus on one or two games; I suspect that’s a few decades out. The good news is that there are some older titles that are still holding their own in the market— to beat out the newcomers hints that there’s some kind of staying power there.
- disordered 7y ago
  I don't agree with your premise that chess has evolved to be at the pinnacle of its niche (I don't know enough about Go to comment on it). There are a number of interesting variations in chess-like games that, while maintaining the general theme of killing the enemy king, offer aspects of gameplay that chess does not. Two of the more well known variations are shōgi and xiangqi: Shōgi allows the redeployment of captured pieces, whilst xiangqi limits the movement of a few pieces and has a mechanism for representing artillery. I find that each game has a very different character to that of western chess, even despite the fact that they share a large number of tactical motifs. Western chess may be the most popular variant world-wide, but that's no indication that it is at any kind of pinnacle.
  - michaelscott 7y ago
    I think it's fair to call it the pinnacle when talking about the Western world. Both shogi and xiangqi developed from the same base game as chess historically, so I'm not sure how far they venture from the theme compared to something like hnefatafl (which is chess-like but varies considerably in game mechanics when compared with shogi and xiangqi).
    - kd5bjo 7y ago
      Just like Like Go and Poker each have a number of variant rule sets in common use, I mentally group all of these as “chess” based on their common ancestry. I haven’t really studied them enough to know if this is truly justified.

reificator 7y ago

I've been wondering, how closely could a future/alien civilization recreate Magic the Gathering's rules given nothing but a collection that included every card and several (100? 1000?) valid decks for different formats. I imagine they could get fairly close, especially thanks to reminder text. But I wonder how older/erattaed cards or old templating might hinder those efforts. For those unfamiliar with Magic, the comprehensive rules are about 190 pages of 10-20 rules each, plus a glossary/index. On top of that, the golden rule of Magic is that if a card and the rules disagree, the card takes precedence. And there are over 20,000 unique named cards, though a small number are functionally the same card as another with a different name. See this link for the comprehensive rules, but note that the vast, vast majority of games don't require an exhaustive understanding. You can learn enough to play in 10 minutes, though you'll probably need to learn something new each game for a little while. https://magic.wizards.com/en/game-info/gameplay/rules-and-formats/rules https://magic.wizards.com/en/game-info/gameplay/rules-and-fo...

Marazan 7y ago
IMO you wouldn't be able to come close. I'm not familiar with how much reminder text modern cards have but surely there is no way you'd be able to work out anything close to the timing rules from just cards alone.
- reificator 7y ago
  There are plenty of cards that reference sorcery speed, there's a card that lists all the phases in order and has players skip one, there's the Split-Second keyword which says you can't put things on the stack until it finishes resolving. The fact that the game is fundamentally about overwriting its own rules is why I think it might be possible.
- bsder 7y ago
  Older cards have more expressive text. So, you can probably link modern vocabulary to older vocabulary to explain play and rules. However, there are certain rules that are arbitrary. For example, the fact that resolution order follows a "stack" is specified nowhere on cards.

drdeadringer 7y ago

> the oldest known board game is an Egyptian game that dates back to 3100 BCE called Senet. “We almost never have the rules for these early games,” Browne said. “The rules have never been recorded, so our knowledge is largely based on historian’s reconstructions.” I live near the Rosicrucian Egyptian Museum in San Jose, CA. The group behind it owns the whole square city block, so besides the museum itself there's also a research library, other related buildings, and a nice park area all around. Included in the park space is a reconstruction of the Senet game, along with an approximate reconstruction of the rules; it's explained that the rules are an approximation with a rough reason as to why. Whether the rules were simply "lost to time" or due to a subset of this as "they were never written down because everyone knew them" I'm not sure, but it was nice to see the game available in person.

ArtWomb 7y ago

Ludii was presented at 2019 IEEE Conference on Games this past week in London. Lot's more interesting stuff in the rest of the conference program ;) http://ieee-cog.org/assets/program-full.pdf http://ieee-cog.org/assets/program-full.pdf

huehehue 7y ago

I spent a couple of months last year researching Liubo and constructing/play-testing a crude recreation. Of course, we don't even know what a full game set looks like, much less how it was played. But the result was a decent game, and I like to think there's enough overlap between board game enthusiasts and history nerds that someone could crank a few of these out on Kickstarter.