Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
xopods
Oct 26, 2010

Personally I do all my prototyping in Illustrator and InDesign, but that's because I own those programs and know how to use them. I certainly wouldn't recommend shelling out for Creative Suite and taking the time to use those applications properly if it's only for prototyping.

But if anyone does have them and a working knowledge of how to use them, let me know and I can tell you a bit about my process. Not that it's much of a process.

Adbot
ADBOT LOVES YOU

Ebethron
Apr 27, 2008

"I hear the coast is nice this time of year."
"If you're in the right business, it's nice all the year."

xopods posted:

First off, let me say that finding mutually agreed-upon definitions for certain terms was something I was hoping we could do in this thread, and part of the reason I left room for a glossary in the second post.

Just wanted to say that this was a very very interesting post that made a lot of good points.

One thing I would like to add is a suggestion for a definition of 'space' as opposed to 'depth'. Donald X Vaccarino described Dominion as a game with a high degree of 'space' because the optimal lines of play change based on the initial selection of action cards, of which there are many. Dominion with different action cards, he suggested, effectively makes it a different game. This is clearly true from a game theory perspective.

In a presentation I watched, Richard Garfield talked about a version of chess in which the starting positions of pieces for each player are random and symmetrical. The space of that game is greater than conventional chess because it is effectively many chess variants embedded in a single game.

Shallow games can reward skill and provide an interesting learning curve if they have a high degree of space. Players have to learn strategies to deal with the many possible alternative game set-ups.

Lots of popular recent board-games have a high degree of space. Maybe it's because lots of gamers like the initial process of discovering basic strategies more than honing and optimising those strategies.

Mark Rosewater argues that the process of discovery, renewed with the release of each set, is one of the reasons why MtG is so popular (the endless Dominion expansions support this sort of argument too). MtG is also a game with a lot of space if we consider the whole constructed metagame, as match-ups between different decks are effectively different games. Players get bored and frustrated when the space narrows because the format has been 'solved' and a single best deck has been discovered(i.e. a shallow metagame results in reduced space across a sequence of games in a given tournament).

PaybackJack
May 21, 2003

You'll hit your head and say: 'Boy, how stupid could I have been. A moron could've figured this out. I must be a real dimwit. A pathetic nimnal. A wretched idiotic excuse for a human being for not having figured these simple puzzles out in the first place...As usual, you've been a real pantload!

Ebethron posted:

Mark Rosewater argues that the process of discovery, renewed with the release of each set, is one of the reasons why MtG is so popular (the endless Dominion expansions support this sort of argument too). MtG is also a game with a lot of space if we consider the whole constructed metagame, as match-ups between different decks are effectively different games. Players get bored and frustrated when the space narrows because the format has been 'solved' and a single best deck has been discovered(i.e. a shallow metagame results in reduced space across a sequence of games in a given tournament).

Almost every magic player i know loves draft a lot more than any other format for this reason. It's funny too because you'd think limiting your overall cardpool would limit choices but really it allows a lot more cards into the realm of playability.

xopods
Oct 26, 2010

Here's a list of definitions I drew up (to send to the guy I'm debating with on Facebook). Let me know if anyone has any objections:

Absolute strategic complexity: The size of the strategy set, which is basically the size of the position space multiplied by the average number of choices available at each node.

Effective strategic complexity: Same as above, but after the choices that can easily be eliminated by a typical player have been removed, along with the game states that can never be reached without employing those choices (e.g. removing early game Go positions in which someone has played at the 1-1 point). Note that mathematical rigour is not required for these deletions... it doesn't matter if a given move isn't provably wrong, if no one ever chooses it.

Skill factor: (W - 0.5) / 0.5, where W is the win rate of a perfect player over a random player. So something with zero luck has a (1 - 0.5) / 0.5 = 100% skill factor, while something completely random, where no player can have an advantage, has a (0.5 - 0.5) / 0.5 = 0% skill factor.

Skill level: Two players are said to be "one skill level" apart if the better player's win rate is 0.5 + S / 6, where S is the skill factor. The reason for the 1/6 factor is that it means that in a 100% skill game, it means a player one skill level better wins twice as often (1/2 + 1/6 = 2/3, while the weaker player wins 1/3) , and you scale down from there based on how luck-influenced the game is. If a perfect player wins 75% of the time against a random player, then the skill factor is 50%, and one skill level difference means the better player wins 58.3% of the time.

Learning curve: The function E(s) where E is effort and s is skill level. These functions are monotonically increasing up until the effort level required to solve the game, undefined thereafter (as there are no skill levels beyond solution, by definition). The first derivative is likewise typically increasing, but needn't be monotonically so; you can have steep bits where great effort is required to overcome a given mental block, followed by shallower areas where progress is rapid once the lightbulb has turned on.

Absolute strategic depth: The skill level at which the learning curve becomes undefined, i.e. the game has been solved and no further progress is possible.

Effective strategic depth: Same as above, but rather than starting from the solved game, start from the best player who exists or has ever existed. Note that this depends on the game's popularity.

------

Regarding the "Magic metagame," drawing the kingdom in Dominion, random chess setups, and so forth... these are ways of linearly increasing the absolute strategic complexity of the game, in that they increase the size of the position space, without changing the number of options at each node.

By contrast, doing something like making the chessboard bigger, or having a 12-card kingdom in Dominion, etc. increase complexity in a non-linear way, because they both expand the position space AND increase your options at each node. You have to be careful about this, because it can rapidly make a game unmanageable.

For example:

Without accounting for symmetry and deleting illegal positions, Go on a 3x3 board has a position space of about 3^9 = 19.5k and something like an average of something like 3-4 legal moves in most of those positions. 19.5k x 3.5 is about 68k possible strategy sets.

Go on a 4x4 board has a position space of about 4^9 = 262k and something like an average of 6-7 legal moves in most of those positions. That's around 1.7M possible strategy sets.

So by making the board a bit bigger, you've increased the absolute strategic complexity by a factor of 25. (It's actually a bigger increase than this because we're ignoring illegal positions, and they make up a bigger fraction of the total positions on smaller boards).

In the real world, Go used to be played on 17x17 boards, more than a thousand years ago in Tibet. Eventually they moved it up to 19x19 and that's kept us busy ever since. Computers still have quite a ways to go before being able to beat the best humans, but I've heard it suggested that when they do, we'll just move up to 21x21 and that'll set them back another few hundred years. (I'm not convinced of this, but it depends on how clever the AI techniques are... certainly if it's largely by brute force, a slightly bigger board makes it much harder for them than for human players).

xopods fucked around with this message at 17:19 on Nov 23, 2012

jmzero
Jul 24, 2007

quote:

Skill factor: (W - 0.5) / 0.5, where W is the win rate of a perfect player over a random player. So something with zero luck has a (1 - 0.5) / 0.5 = 100% skill factor, while something completely random, where no player can have an advantage, has a (0.5 - 0.5) / 0.5 = 0% skill factor.

It seems hard to define a "random opponent" for a lot of games. Obviously games without cleanly discrete states and choices are a problem (eg. what does a random basketball player look like?). But even for clean discrete games there's going to be a lot of times where a generalized random opponent (ie. chooses from legal moves with equal probability) doesn't seem to work well as a measure.

Take "Maze Race", where players can move one orthogonal space each turn and attempt to get to the end of a maze first. For a sufficiently large board, a random player has close to zero chance of beating a real player, so the game is going to have a stratospheric skill rating. At the same time, for that same large board (and assuming time limits are in force for turns or something), this is going to be a very luck intensive game for human players.

For other games, a random player won't do nearly so bad; in 7 Wonders, it's very possible a random player could win. It seems like a random player is either hard to define (if their behavior is more complicated), or is measuring something else about a game (often it'll be something like "how bad does the worst legal move tend to be"). For example, having the option of offering a Doubling Cube makes Backgammon more skill intensive for human players - but just gives a random player a way to shoot himself in the foot every turn.

jmzero fucked around with this message at 18:05 on Nov 23, 2012

xopods
Oct 26, 2010

jmzero posted:

It seems hard to define a "random opponent" for a lot of games. [...]
Take "Maze Race", where players can move one orthogonal space each turn and attempt to get to the end of a maze first. For a sufficiently large board, a random player has close to zero chance of beating a real player, so the game is going to have a stratospheric skill rating. At the same time, for that same large board (and assuming time limits are in force for turns or something), this is going to be a very luck intensive game for human players.

You're mixing up "skill factor" with what we discussed re: effective depth.

Skill factor has nothing to do with human ability, only the intrinsic luck factor in the game. A perfect player WILL beat a random opponent 100% of the time in the limit of an infinitely large maze, and that's the result we want, because "Maze Race" is a perfect information, zero chance game (once the game is underway, that is: there is randomness in the generation of the maze, but since this happens prior to any player making a move and the results are known to all players, it doesn't introduce any luck into the game itself). In 7 Wonders, the perfect player will occasionally lose to the random player, and that's also as we want it; the skill factor will be less than 100% (though still very high), because the random player will occasionally luck into better draws than the perfect player, and if he then gets further lucky to make all the right choices, he'll win no matter what the perfect opponent does.

Keep in mind the hierarchy of the definitions:

"Skill factor" is based only on the effect of a game's intrinsic randomness on the game's outcome. It reflects what the hypothetical maximum win rate is (assuming the opponent isn't actively trying to lose). 100% for chess, 83.3% for the "Chess+Luck" game, where you can win the main game 100% of the time and still lose the die roll 1/6th of the time.

"Skill levels" reflect players' abilities to beat one another, and have to be defined relative to the skill factor of a game - otherwise Chess+Luck wouldn't have as many skill levels as Chess, but obviously the strategic considerations are the same in both, so the number of skill levels should be equal.

"Effective strategic depth" is where the limits of human ability come into play - the range of skill levels achievable by humans.

-----

We could, however, perhaps refine our definition of effective strategic depth by chopping off the low end of the scale, since for most games, even a complete beginner will do better than random play:

Absolute strategic depth = skill levels between perfect play and random play
Effective strategic depth = skill levels between best player ever and average skill level of a first-time player

-----

I understand that you want our definitions to reflect how the game is experienced by actual players, but to get there, we have to start with intrinsic properties of the game itself, and bring in the human aspect later. We need to define skill factor in order to talk about skill levels, and we need the definition of skill levels in order to talk about learning curves, and learning curves to talk about effective depth... if you go back and try to make the definition of skill factor depend on the game's effective strategic depth, you've created a circular definition which is of no use to anyone.

What's actually happening with the "VERY VERY BIG Maze Game" is that you've expanded the position space to the point that the learning curve has become very steep... eventually steep enough that not even one skill level of improvement is possible in a human lifetime. That doesn't make it a game of chance, it just makes the effective strategic depth very small.

Essentially, effective strategic depth is limited by one of two things; either the point at which the game becomes solved, for simple games, or the point at which the learning curve E(s) hits E = Emax, for more complicated games, where Emax is the maximum effort you expect anyone to ever spend on your game.

To achieve the maximum effective strategic depth, then, you basically want the shallowest learning curve possible without allowing the game to become solvable. In general, increasing the absolute strategic complexity of a game will increase the absolute strategic depth, but also steepen the learning curve. So if a game is too simple, you can increase its effective strategic depth by complicating or expanding it... but if the issue is that the strategy gets beyond human grasp at a certain point, you can actually increase the effective depth by reducing the complexity, allowing people to get to those higher levels of skill in a more reasonable timeframe.

-----

On further thought, it may be worth inserting a definition for "effective skill factor," meaning the different between a perfect and a random player after the game space has been pruned in the same way we use to get from absolute to effective complexity. I.e. the advantage of a perfect player over one who merely rejects the moves which are "obviously" wrong (by some definition of obviousness) and chooses randomly from the rest. For instance, never visiting any space in the Maze Game a number of times greater than its number of exits.

But the question then is at which point we make the step from absolute to effective when we're establishing the subordinate definitions... i.e. do we calculate skill levels based on absolute skill factor and then just chop off the learning curve below E(s) = Eobv (where Eobv is "the level of effort required to recognize the obvious")? Or do we go straight to effective complexity and base everything else off of that, forgetting that the "obviously wrong" part of the game space exists at all?

xopods fucked around with this message at 19:17 on Nov 23, 2012

jmzero
Jul 24, 2007

quote:

Skill factor has nothing to do with human ability, only the intrinsic luck factor in the game.

Sorry - I do see what you're getting at, and I did kind of smush a bunch of your stuff together in my comment there. Let me start again, maybe.

Consider 7 Wonders. Your calculated skill factor is going to be lower than most comparable games because of actual randomness in the game - but also because the game kind of forces you to towards viable moves. Say we add in a horrible move that would almost never be taken by a real player. Our skill factor is going to go through the roof, because our random player is going to take it 3 times a game.

And then we use that factor as a starting point to calculate our other numbers. We're going to end up with very different results when we talk about the learning curves of our variants - that despite the fact that for non-random players the games are almost exactly the same.

quote:

...we have to start with intrinsic properties of the game itself...

What I'm saying is that comparing a random player and a perfect player will not allow us to separate the effects of randomness from other properties of the game; for one, we'll also be measuring the number and value variance of options available in the game (and we won't be able to restrict ourselves easily to any kind of reasonable subset of those options without ungeneralizing our random agent). This taints the rest of our calculations, and means we will have a hard time using these other numbers to compare different games in a useful way.

quote:

For instance, never visiting any space in the Maze Game a number of times greater than its number of exits.

Edit: Yes, sorry, that's very much what I was trying to get at. It just seems likely to be messy.

xopods
Oct 26, 2010

jmzero posted:

Consider 7 Wonders. Your calculated skill factor is going to be lower than most comparable games because of actual randomness in the game - but also because the game kind of forces you to towards viable moves. Say we add in a horrible move that would almost never be taken by a real player. Our skill factor is going to go through the roof, because our random player is going to take it 3 times a game.

And then we use that factor as a starting point to calculate our other numbers. We're going to end up with very different results when we talk about the learning curves of our variants - that despite the fact that for non-random players the games are almost exactly the same..

Yep, if you look at the stuff I just edited in while you were responding (down at the end of my post), you'll see I was starting to come to the same conclusion.

It's probably more useful (though much harder to work with) to define things in this sequence:

code:
ASC -> ASF -> ASL -> ASD
    \
     > ESC -> ESF -> ESL -> ESD
instead of what I was doing before:

code:
ASC -> ASF -> ASL -> ASD
   \      \             \
    > ESC  > ESF         > ESD
I.e. decide up front whether we're talking about the intrinsic (absolute) or extrinsic (effective) properties of the game and derive everything from there, rather than going down the chain of absolute definitions and just tacking on "effective" variants of each.

EDIT: I have to do some serious work now, but I'll think about this a bit more and try to draw up Definitions 2.0 tonight or over the weekend based on this "more useful but harder to work with" chain of contingency. It does mean, though, that everything is going to hinge on a very fuzzy notion of obviousness... and that means down the road we're going to have shouting matches about what it is and isn't reasonable to expect to be obvious to a total beginner.

xopods fucked around with this message at 19:31 on Nov 23, 2012

wins32767
Mar 16, 2007

xopods posted:

...down the road we're going to have shouting matches about what it is and isn't reasonable to expect to be obvious to a total beginner.

It gets even worse than that. I've been trying to teach someone who is a world class Magic player how to play wargames on the high end of the complexity scale. Stuff that is obvious to other people that have played hex games (what are the key hexes for a given defensive line for example) is a challenge to someone who has mostly only played Magic. Even someone with excellent abstract reasoning capability is going to be at a disadvantage in their first game against another beginner that has played a game from the same family.

xopods
Oct 26, 2010

wins32767 posted:

It gets even worse than that. I've been trying to teach someone who is a world class Magic player how to play wargames on the high end of the complexity scale. Stuff that is obvious to other people that have played hex games (what are the key hexes for a given defensive line for example) is a challenge to someone who has mostly only played Magic. Even someone with excellent abstract reasoning capability is going to be at a disadvantage in their first game against another beginner that has played a game from the same family.

Right, well, I think for our purposes we're talking about "obvious to anyone capable of understanding the rules, regardless of any prior gaming experience or lack thereof."

You can always chop off the beginning of the learning curve at a certain point and talk about "effective depth" and "learning curve" given a certain level of prior knowledge if you want... i.e. "the learning curve is steep for true beginners, but manageable if you've played other worker placement games."

...but when I say "obviously wrong" I mean things that are really, really obvious, like sacrificing pieces for no gain whatsoever, or buying something and immediately reselling it for half the price, or tapping lands and not spending the mana, just giving yourself mana-burn for no reason. Things that should clearly be omitted from the choices for a "random" opponent if that's your model for a complete beginner.

xopods fucked around with this message at 05:08 on Nov 24, 2012

xopods
Oct 26, 2010

By the way, I hope I'm not driving anyone off with all these extremely nerdy megaposts about defining things.

Please do continue to bring up more down-to-earth issues and questions! :)

As a possible lighter topic, does anyone have any mechanics they feel are underexplored? (As opposed to, say, worker placement, which seems a little overdone these days... the roll-and-move of the early 21st century?)

signalnoise
Mar 7, 2008

i was told my old av was distracting
What are the major pitfalls of creating a game from scratch? I have had this idea kicking around in my head for about 2 years now of a squad tactics strategy game that uses a cards for activation of special abilities, special effects on the board, and placement of new terrain on a grid map. I am 100% married to the idea of effect cards + miniatures. I am just unsure what I should do to bring it about other than make, playtest, repeat. My plan is to buy some blank playing cards to write on, and get a laminated map printed for a grid map, and go from there trying to get my friends to play and help develop it. I know it'll take a long time of course, but it seems like a fun thing to do. I just don't want to get started on the wrong foot out of the gate.

signalnoise fucked around with this message at 15:17 on Nov 24, 2012

Dr. Clockwork
Sep 9, 2011

I'LL PUT MY SCIENCE IN ALL OF YOU!


This is pretty much my gut reaction whenever I sit down to playtest games I design. Learning to take criticism and improve your game, especially if you playtest with other gamers, is really important.

Nemesis Of Moles
Jul 25, 2007

xopods posted:

By the way, I hope I'm not driving anyone off with all these extremely nerdy megaposts about defining things.

Please do continue to bring up more down-to-earth issues and questions! :)

As a possible lighter topic, does anyone have any mechanics they feel are underexplored? (As opposed to, say, worker placement, which seems a little overdone these days... the roll-and-move of the early 21st century?)

Don't worry about it. Defining things is important otherwise we lack the basic language to be able to build games.

I think Roll And Move could really do with some exploration. There isn't really such a thing as 'Bad Mechanics', only underdeveloped ones and rolling a big load of dice can be a lot of fun on a base level.

Deck building has been done to death by now but the Dice Building thing that Quarriors did was neat and I'd like to see more of it.

My fave mechanic style that I want to see more of is the physical kinetic style stuff of Ascending Empires and that one dungeon game where you flick your pieces around with your fingers.

Ulta
Oct 3, 2006

Snail on my head ready to go.

xopods posted:

By the way, I hope I'm not driving anyone off with all these extremely nerdy megaposts about defining things.

Please do continue to bring up more down-to-earth issues and questions! :)

As a possible lighter topic, does anyone have any mechanics they feel are underexplored? (As opposed to, say, worker placement, which seems a little overdone these days... the roll-and-move of the early 21st century?)

As one of those hippie story gamers, I really like conceptually what "Aye Dark Overlord" does, but wish it had a more vigorous mechanical structure.

The game is you are goblin minions returning from a botched heist. You make excuses, adding story elements with cards. You can "pass the buck" or interrupt other players with other cards. You lose by contradicting yourself, being unable to pass the buck, or generally displeasing the Dark Overlord, who is a player that acts as a quasi-GM, or judge.

Guildencrantz
May 1, 2012

Somewhere, something incredible is waiting to be known.

xopods posted:

As a possible lighter topic, does anyone have any mechanics they feel are underexplored? (As opposed to, say, worker placement, which seems a little overdone these days... the roll-and-move of the early 21st century?)

Voting! I find the area of decision-making in democracies fascinating, and it's one that board games could explore very well on a small scale. Take a basic multiplayer game of, say, economics, add an element of rules changes by vote, and you have a whole open-ended dynamic of decisions and interactions for the players. And there are plenty of parameters the designer can screw around with, such as secret vs non-secret ballots or veto power, that can massively change the way players make their choices. Sadly, I can't think of any current games that really use voting.

Nemesis Of Moles
Jul 25, 2007

Plenty of party games use Voting. I'd love to see something like Risk Legacy or Chaos In the Old World or Civilisation with an added Player Government model. With rulesets for being different forms of government.

PaybackJack
May 21, 2003

You'll hit your head and say: 'Boy, how stupid could I have been. A moron could've figured this out. I must be a real dimwit. A pathetic nimnal. A wretched idiotic excuse for a human being for not having figured these simple puzzles out in the first place...As usual, you've been a real pantload!

Guildencrantz posted:

Voting! I find the area of decision-making in democracies fascinating, and it's one that board games could explore very well on a small scale. Take a basic multiplayer game of, say, economics, add an element of rules changes by vote, and you have a whole open-ended dynamic of decisions and interactions for the players. And there are plenty of parameters the designer can screw around with, such as secret vs non-secret ballots or veto power, that can massively change the way players make their choices. Sadly, I can't think of any current games that really use voting.

The problem with voting in games is that the guy winning, or in the position of the most power is the one who gets screwed. Because the game lasts a finite number of turns there is rarely a good amount of incentive to work with the winner. Thus most times voting, or diplomacy occurs it's to unkingmake someone.

Guildencrantz
May 1, 2012

Somewhere, something incredible is waiting to be known.

PaybackJack posted:

The problem with voting in games is that the guy winning, or in the position of the most power is the one who gets screwed. Because the game lasts a finite number of turns there is rarely a good amount of incentive to work with the winner. Thus most times voting, or diplomacy occurs it's to unkingmake someone.

That's a problem that can be circumvented, though. Hidden victory conditions come to mind.

Now I have a conspiracy theory-inspired game in my head.

modig
Aug 20, 2002
Role drafting is a fairly common element now, but I want to see/try something with role giving. I especially enjoy the Citadels version where there is some deduction and guesswork involved. So instead of taking your own role, you give a role to someone else.

I tried to start by coming up with a theme rather than just writing down mechanics, because I really don't have much to go on aside from that one idea. So far I've come up with the idea of a hippie commune where everybody shares work, but all they really want to do is get high. That leads me to having some element of either doing a good job at your role, or doing a crappy job and getting high instead. But thats about as far as I've gotten so far.

Admin Understudy
Apr 17, 2002

Captain Pope-tastic
I've been committed to board game design for a while now, I have a couple different notebooks or rather scrapbooks I'd call them at this point as I'm usually hit with ideas randomly throughout the day and jot them down on whatever paper I can find nearby. The process is fascinating and fun and I think the two biggest things impeding me right now are my strong desire for originality and also trying to make a game that everyone would enjoy.

So it really bugs me when I have someone try out a basic prototype and they'll innocently ask "so this deck of action cards sets the length of the game like in Brass or Dominant Species?" which is something I had never even thought of and I'll immediately want to scrap that whole game because I'm convinced that I've simply pilfered the core mechanics of other games. I have to believe this is a very unhealthy way of thinking because no game of any slight level of complexity consists of all original mechanics.

As for my second concern about trying to please everyone, this past year of 2012 I have played more games than any other year since I got into modern boardgaming about 5 years ago (played over 100 different board games this year) and it's helped me refine my personal preferences in gaming. That was a big step in being able to decide the target audience when designing a game.

xopods posted:

Skill factor: (W - 0.5) / 0.5, where W is the win rate of a perfect player over a random player. So something with zero luck has a (1 - 0.5) / 0.5 = 100% skill factor, while something completely random, where no player can have an advantage, has a (0.5 - 0.5) / 0.5 = 0% skill factor.

Skill level: Two players are said to be "one skill level" apart if the better player's win rate is 0.5 + S / 6, where S is the skill factor. The reason for the 1/6 factor is that it means that in a 100% skill game, it means a player one skill level better wins twice as often (1/2 + 1/6 = 2/3, while the weaker player wins 1/3) , and you scale down from there based on how luck-influenced the game is. If a perfect player wins 75% of the time against a random player, then the skill factor is 50%, and one skill level difference means the better player wins 58.3% of the time.

Learning curve: The function E(s) where E is effort and s is skill level. These functions are monotonically increasing up until the effort level required to solve the game, undefined thereafter (as there are no skill levels beyond solution, by definition). The first derivative is likewise typically increasing, but needn't be monotonically so; you can have steep bits where great effort is required to overcome a given mental block, followed by shallower areas where progress is rapid once the lightbulb has turned on.

Absolute strategic depth: The skill level at which the learning curve becomes undefined, i.e. the game has been solved and no further progress is possible.


I like your relationship between what you're calling "skill level" and the depth of a game. I would probably argue that the learning curve isn't a relationship between effort and "skill level" though but rather "effectiveness of tactics" and skill level. If I don't understand the consequences of my actions my effort can consistently be very low and have a wide range of positive and negative outcomes whereas my ability to be successfully tactical is more revealing of how deep a game is.

Admin Understudy
Apr 17, 2002

Captain Pope-tastic
Actually one more thing I'd like to get some insight on; designing with a range of players in mind. I get hung up on this sometimes where I initially go through simulating a couple rounds of a real basic construct of a game at 4 players and then when I try the same at 3 or 5 players it completely breaks down. I tend to be the most productive when I design something with an exact player count in mind but apart from the niche of 2 player games that seems rather unappealing if my 5 player game can't handle 4 or 3 instead.

Any tips for scaling games? Is it generally easier to do it from the offset or to first get a completed game and then find a solution for accommodating more or less players?

Guildencrantz
May 1, 2012

Somewhere, something incredible is waiting to be known.

modig posted:

Role drafting is a fairly common element now, but I want to see/try something with role giving. I especially enjoy the Citadels version where there is some deduction and guesswork involved. So instead of taking your own role, you give a role to someone else.

I tried to start by coming up with a theme rather than just writing down mechanics, because I really don't have much to go on aside from that one idea. So far I've come up with the idea of a hippie commune where everybody shares work, but all they really want to do is get high. That leads me to having some element of either doing a good job at your role, or doing a crappy job and getting high instead. But thats about as far as I've gotten so far.

A hippie commune doesn't strike me as a very competitive environment, is there a victory condition other than chilling out? Because so far I'm envisioning a meta-boardgame where you draw a few cards, then get high, forget about the game and say everybody wins. And I think we've all played that one a few times.

modig
Aug 20, 2002

Guildencrantz posted:

A hippie commune doesn't strike me as a very competitive environment, is there a victory condition other than chilling out? Because so far I'm envisioning a meta-boardgame where you draw a few cards, then get high, forget about the game and say everybody wins. And I think we've all played that one a few times.

The victory condition would probably be like smoke X amount of pot... it could work but its not super compelling.

OXBALLS DOT COM
Sep 11, 2005

by FactsAreUseless
Young Orc

Guildencrantz posted:

A hippie commune doesn't strike me as a very competitive environment, is there a victory condition other than chilling out? Because so far I'm envisioning a meta-boardgame where you draw a few cards, then get high, forget about the game and say everybody wins. And I think we've all played that one a few times.

Dunno, any small community gets competitive when you have limited resources yet competing interests at hand. Especially since people who run off and join communes tend to be 'characters' with weird quirks and motivations.

Or just give out secret victory conditions. Everyone is supposed to be a hippie looking for utopia, but instead (by hidden draw of the card), you have undercover FBI, lechers trying to assemble a harem to run off with them, con-men, sell-outs, power-freaks, religious nuts looking to start a cult, Manson types, etc. Maybe make it so that the hippie itself is a unit of value, and when you get enough hippies on your side instead of another player's, you win.

You do a good job in your role to try and keep it, or you goof off and smoke pot to try and get assigned a different role that will let you achieve your goals. If everyone goofs off and does nothing long enough, the community falls apart, the hippies wander off, and everyone loses. Or else it's a bluff, and you're actually smoking pot so people think you don't want to keep the role you actually need to collect your victory. Etc.

OXBALLS DOT COM fucked around with this message at 21:00 on Nov 24, 2012

Nemesis Of Moles
Jul 25, 2007

Admin Understudy posted:

So it really bugs me when I have someone try out a basic prototype and they'll innocently ask "so this deck of action cards sets the length of the game like in Brass or Dominant Species?" which is something I had never even thought of and I'll immediately want to scrap that whole game because I'm convinced that I've simply pilfered the core mechanics of other games. I have to believe this is a very unhealthy way of thinking because no game of any slight level of complexity consists of all original mechanics.

Quit this. This is like, the silver rule next to 'Accept criticism' and 'Learn How to Type' an poo poo. No game is unique, ever. Dominion is just Magic is just Cosmic Encounter is just Blackjack. What you do with the idea is what counts.

Rudy Riot
Nov 18, 2007

I'll catch you Bran! Hmm... nevermind.

signalnoise posted:

What are the major pitfalls of creating a game from scratch? I have had this idea kicking around in my head for about 2 years now of a squad tactics strategy game that uses a cards for activation of special abilities, special effects on the board, and placement of new terrain on a grid map. I am 100% married to the idea of effect cards + miniatures. I am just unsure what I should do to bring it about other than make, playtest, repeat. My plan is to buy some blank playing cards to write on, and get a laminated map printed for a grid map, and go from there trying to get my friends to play and help develop it. I know it'll take a long time of course, but it seems like a fun thing to do. I just don't want to get started on the wrong foot out of the gate.

A cheap and easy method for making cards for testing that was mentioned earlier in the thread is to sleeve some cheap Magic or regular playing cards, then slip your printed out/ hand written custom card in front of it. You can use regular paper and go nuts changing stuff as often as needed. Much cheaper than blank playing cards I'd imagine.

Loving this thread by the way! A friend and I had pitched a game to Travis at Indie Boards & Cards and he was super cool and responsive. Nothing came of it, but the process was really fun and inspiring. I've got a creative design I'm working on that I will post here soon once I get past some initial hurdles I'm struggling with. I'm lost on a lot if the deep game theory stuff that's been posted about so far, but its still interesting to read.

iceyman
Jul 11, 2001

Nemesis Of Moles posted:

I think Roll And Move could really do with some exploration. There isn't really such a thing as 'Bad Mechanics', only underdeveloped ones and rolling a big load of dice can be a lot of fun on a base level.

I agree. Roll and Move gets a bad rap because it conjures images of simple and bad designs like Monopoly and Sorry where this mechanic is very strict and unmerciful. But it can be used better and has. A Touch of Evil uses it to randomize limited options each turn. Depending on board layout, it's really no better or worse than other "start of your turn" randomizers like card drawing or dice pool rolling.

modig
Aug 20, 2002

Rudy Riot posted:

A cheap and easy method for making cards for testing that was mentioned earlier in the thread is to sleeve some cheap Magic or regular playing cards, then slip your printed out/ hand written custom card in front of it. You can use regular paper and go nuts changing stuff as often as needed. Much cheaper than blank playing cards I'd imagine.

Loving this thread by the way! A friend and I had pitched a game to Travis at Indie Boards & Cards and he was super cool and responsive. Nothing came of it, but the process was really fun and inspiring. I've got a creative design I'm working on that I will post here soon once I get past some initial hurdles I'm struggling with. I'm lost on a lot if the deep game theory stuff that's been posted about so far, but its still interesting to read.

We bought some card sleeves with art on the back, I think they are a bit stiffer than just straight clear plastic. With those we find just printing paper and sleeving them is enough, no need to even pad them with magic cards. Like these: http://www.miniaturemarket.com/max7050hgw.html

xopods
Oct 26, 2010

Wow, a whole pile of posts that I'd love to respond to individually at length, but since I should be going to bed, I'll just quote en masse and respond to more briefly.

Nemesis Of Moles posted:

I think Roll And Move could really do with some exploration. There isn't really such a thing as 'Bad Mechanics', only underdeveloped ones and rolling a big load of dice can be a lot of fun on a base level.

I agree. Actually, one of my games coming out in the next year is vaguely in the same spectrum as backgammon, which I guess makes it a roll-and-move (though the movement is two-dimensional... it involves monkeys climbing and jumping through a bamboo grove). There's nothing inherently wrong with tracks, and semi-random numbers to move pieces along them... the two main historical problems with roll-and-move is that it was used in just about every unoriginal game during a certain dark period in history, and that its typical implementation involves absolutely no control on the part of the players, which makes it not so different from a more cumbersome equivalent of just drawing a card from a deck or a chip from a bag, which is what we tend to do now instead.

Guildencrantz posted:

Voting! I find the area of decision-making in democracies fascinating, and it's one that board games could explore very well on a small scale.
[...]
[Gang-up-on-the-leader is] a problem that can be circumvented, though. Hidden victory conditions come to mind.

A while back I had an extremely vague idea for a political game in which players are politicians with public agendas according to their party platforms and hidden agendas, and propose and vote on legislation. Way more complicated and less abstract than my usual stuff... but I love the idea of voting as a mechanic if you can get around that intrinsic gang-up-on-the-leader thing.

I was pleased to see in the design contest that there were a couple of games along those lines - the abstract voting one and the semi-cooperative Rome one where there's a winner in the end but you have to work together or everyone loses. (I'd have to go and look at the thread to remind myself of the names of the games and who made them, but I hope we see more of them in this thread).

Admin Understudy posted:

I would probably argue that the learning curve isn't a relationship between effort and "skill level" though but rather "effectiveness of tactics" and skill level. If I don't understand the consequences of my actions my effort can consistently be very low and have a wide range of positive and negative outcomes whereas my ability to be successfully tactical is more revealing of how deep a game is.

We're not talking about the effort made by the player during the current instance of the game, but the total effort made to learn the game up to that point - whether simply by playing repeatedly or deliberately studying the strategy.

I.e. if you've played 20 times at 2 hours each (40 hours), spent another 5 hours total discussing the game with opponents after having played with them, 10 hours lying in bed thinking about the game, and another 5 hours reading strategy-related articles about the game online, you've put 40 + 5 + 10 + 5 = 60 hours of effort into learning that game.

If at that point, you're beating a total beginner 98% of the time, the game has a shallow learning curve, because your skill has obviously improved in leaps and bounds. If you're still only beating a beginner 55% of the time (and it's not a game with a huge amount of luck), then it must have a very steep learning curve, because you're still not understanding much more of the strategy than when you started.

(Obviously the next step is to say that not all effort is equivalent... for players at a certain level, doing Go problems might be better than playing Go, while at other levels, playing is better, etc. But straight-up hours devoted is a good enough first approximation).

Nemesis Of Moles posted:

Quit [worrying too much about being original]. This is like, the silver rule next to 'Accept criticism' and 'Learn How to Type' an poo poo. No game is unique, ever.

Absolutely. There's an analogy to be made with music... unless you're going to the crazy, extreme, experimental end of things, you're using the same notes and chords as everyone else, and a lot of the same progressions. But the neat thing is that there are enough possible variations on and combinations of those basic building blocks that new and interesting songs - and games - keep coming out that make use of all those common patterns without exactly duplicating anything that came before.

Personally, I think the surest recipe for a great game is to take awesome thing A from one game, and awesome thing B from another, and then come up with original thing C that makes them work together. Sure, there are a few gamers out there who will hate on anything if they can tell where the inspiration came from, but gently caress'em. The games they like were also inspired by something, they just haven't played it themselves, so they don't recognize it.

Admin Understudy posted:

Actually one more thing I'd like to get some insight on; designing with a range of players in mind. Any tips for scaling games?

I don't know if they'll answer any of your questions, but I wrote a series of blog posts about how your design challenges change with the number of players you're targeting. They start here and the other ones are linked from the first one:

http://www.benefactum.ca/?p=120

I guess read those and if there's anything you're wondering about that's not covered, bring it up and we'll talk about it!

-----

Okay! Enough for one night. I'm glad this thread is getting traffic though, and some interesting things being discussed.

xopods fucked around with this message at 07:27 on Nov 25, 2012

xopods
Oct 26, 2010

Oh, just one other thing about learning curve.

In the example I give, I know some people are going to say that it sounds like an easier game on beginners where the more experienced player is only winning 55% than 98%. But you have to turn things around and look at things from the beginner's perspective, vs. someone who beats them a certain % of the time, and how long it's going to take the beginner to catch up in skill.

If, after 60 hours, you're only beating a beginner 55% of the time, then that probably means another guy who is beating the beginner 98% of the time has been playing thousands or tens-of-thousands of hours to become that good.

So, from the beginner's perspective, given that they're facing an opponent who can beat them 98% of the time, they're asking how long it'll take them to catch up. If the other guy only took 60 hours to get that far ahead, it'll likewise only take the beginner 60 hours to be able to take that guy on, assuming he doesn't improve much more in the meantime. On the other hand, if 60 hours = 55%, then to catch up with the guy who is winning 98% might take a lifetime.

xopods fucked around with this message at 07:44 on Nov 25, 2012

Nemesis Of Moles
Jul 25, 2007

Funny that we started talking about Roll and Move the week before Rab did a review on Escape, a game that does Roll and Move in a pretty neat and fun way. http://www.rockpapershotgun.com/2012/11/25/cardboard-children-escape/

modig
Aug 20, 2002
Speaking of roll and move, check out this online board game simulator that came out of gently caress this jam. The computer doesn't enforce any rules, it just provides a 3d model of the board and components. The game they implemented is described as "chutes and ladders plus diablo".
http://www.wolfire.com/desperate-gods
They definitely went somewhere that my brain wouldn't have gotten to. I found it by following some links in the comments on Escape, which looks pretty fun. I guess this is the design thread, so maybe fun isn't what I should say. I like that it has a different take on how to make a co-op game not be single player, a bit like space alert.

xopods
Oct 26, 2010

Chutes & Ladders + Diablo = Talisman :negative:

As for this thing itself, it's kind of an interesting project... an attempt to make a better version of Vassal or whatever. But it's funny; this kind of thing is very typical of the programmer mindset. Prior to teaching myself Cocoa, I briefly tried to partner up with a programmer to do an iOS version of my game Picnic Blitz (by far my least well-received game on BGG, but as a dice game, I thought it might be good for a casual iOS audience). I gave up after he spent the first couple of weeks fussing with physics for the dice and deciding on the most efficient data structure to represent the board position (which consists of at most 24 pieces on six tracks of a maximum of 10 spaces... i.e. not a situation where efficiency is remotely an issue) and still had barely done any actual game logic. Programmers... :rolleyes:

----

When I think about "improving roll-and-move," I'm thinking about increasing the players' options so they're not just at the mercy of the dice.

For instance, this as-of-yet-untitled game I'm putting out next year (the backgammon-like one with monkeys climbing and jumping in the bamboo) has the players rolling two dice and either moving one monkey with each, or the same monkey twice, like backgammon... but both players have a third die set aside, that they can swap with either of their rolled dice. So for instance, if you have a 3 set aside and roll 5-4, you could take your 5-4, or take 5-3 instead and leave yourself a 4 for next turn, or 4-3 and leave yourself a 5 for next turn.

That same mechanic could be dropped more or less directly into regular backgammon, or Monopoly, or any other roll-and-move game with more than one die being rolled per turn. Aside from giving the player more options so they're less likely to be screwed by a bad roll, it also provides more opportunities to think ahead, in that you're not only thinking about which two dice are best for the current move, but which one is most likely to be useful next turn. For instance, in the monkey game, you can often make a move where you leave yourself the right number to hit the opponent's monkey on your next move, thereby forcing him to reposition that monkey. It gives the game a more positional, chess-like strategy than backgammon, in that, although you still have to worry about the rolls your opponent could get, you have to focus even more on the number your opponent can certainly have if he wants it.

Another example, this one hypothetical (though I wouldn't be surprised if such a game existed), would be a game where the board is composed of tiles, and players have the ability to spend cards or resources to rotate, flip or swap tiles after rolling; if you can't control the dice, at least you can control the track itself.

A game with railroad-style junctions and switches, and with the option to reverse the direction of movement, would also work, with players fighting to control the flow of traffic and get themselves to their destinations while routing opponents into detours, hazards, or futile loops.

xopods fucked around with this message at 18:59 on Nov 25, 2012

Nemesis Of Moles
Jul 25, 2007

So, thanks to SomethingAwfuls board game contests, and this thread, I've finished the first game I consider to be fairly viable. I'm going to wait a week or two to get back some more playtest stuff before I begin, but I'd really love some input on how to shop games to publishers. My degree is all about Business and stuff so I know a lil about proposal writing, but I figured, we have this thread for a reason.

Railing Kill
Nov 14, 2008

You are the first crack in the sheer face of god. From you it will spread.
A game that does roll-and-move not only well but also in a different way is Formula De. It's strictly roll-and-move, and there's little else to the game, but the randomness of "which space is the game going to let me land on" is replaced by careful risk/reward calculation on the part of the player with literally every roll. Every roll has to be weighed against clear odds, with huge payoff or pounishment for beating or being beat by those odds. I love it and I use it to show folks how roll-and-move doesn't have to be terrible once you wrestle it out of Milton Bradley's hands.

On the topic of "unkingmaking" problems in games, that's an issue I've had with a few different designs. Simple free-for-all games like Munchkin suffer worst from this because it seems like the mechanics aren't robust enough to give the leading player enough options to get the gently caress out of the way. Munchkin can be endless with four or more players. I haven't played it in years, but I wouldn't even bother with more than four, or even three. But adding complexity to a game, in the form of contingencies or endgame mechanical shifts like in Talisman, seems like a rough way to combat "unkingmaking." There ought to be a way to unseat the leader, but the game also has to end. It's a huge pain in the rear end, and especially for already complex, long games where you don't want two or more hours of player strategy to boil down to a "musical chairs" style scramble at the very end. Turn/time limits, secret agendas, and endgame shifts are a few ways to handle the issue, but what are some others (for simple or complex games)?

Admin Understudy
Apr 17, 2002

Captain Pope-tastic

xopods posted:



We're not talking about the effort made by the player during the current instance of the game, but the total effort made to learn the game up to that point - whether simply by playing repeatedly or deliberately studying the strategy.

I.e. if you've played 20 times at 2 hours each (40 hours), spent another 5 hours total discussing the game with opponents after having played with them, 10 hours lying in bed thinking about the game, and another 5 hours reading strategy-related articles about the game online, you've put 40 + 5 + 10 + 5 = 60 hours of effort into learning that game.

If at that point, you're beating a total beginner 98% of the time, the game has a shallow learning curve, because your skill has obviously improved in leaps and bounds. If you're still only beating a beginner 55% of the time (and it's not a game with a huge amount of luck), then it must have a very steep learning curve, because you're still not understanding much more of the strategy than when you started.

(Obviously the next step is to say that not all effort is equivalent... for players at a certain level, doing Go problems might be better than playing Go, while at other levels, playing is better, etc. But straight-up hours devoted is a good enough first approximation).

Ok effort as time spent does make more sense. I think you should make it come full circle with effort being a function of complexity, although that's simply my desire for a perfect collection of functions.

I guess I still don't fully see the relationship of effort to skill level. If we are purely defining skill level as tiers of winning percentage and then correlating that to effort spent getting to some discrete skill level I feel like it doesn't account for a set of strategies with a sort of RPS relationship where more basic strategies can have higher winning rates over more complex strategies but lose to middle-of-the-pack strategies. Even when I generalize it down to saying that more effort boosts a percentage chance of being able to play a specific strategy as opposed to "unlocking" a strategy, I feel like I could create a set where effort spent has an inverse relationship to winning percentage?

Sorry if that came across as convoluted, I'm posting on my phone at work cause I'm bored and love thinking about this stuff! Also I hope this doesn't come across as antagonizing you, merely going for playing devil's advocate.

xopods
Oct 26, 2010

Nemesis Of Moles posted:

So, thanks to SomethingAwfuls board game contests, and this thread, I've finished the first game I consider to be fairly viable. I'm going to wait a week or two to get back some more playtest stuff before I begin, but I'd really love some input on how to shop games to publishers. My degree is all about Business and stuff so I know a lil about proposal writing, but I figured, we have this thread for a reason.

It really depends on the publisher. They all have their own priorities and submission processes and policies.

Usually, though, you're going to want to send them a sell sheet, which should include, as concisely as possible, sections describing:

  • A nutshell explanation of the game's theme and genre
  • What kind of players you expect to enjoy it
  • The core mechanics (probably in bullet point form... don't go into detail)
  • A description of the experience of a typical turn
  • A list of components

xopods
Oct 26, 2010

Admin Understudy posted:

Ok effort as time spent does make more sense. I think you should make it come full circle with effort being a function of complexity, although that's simply my desire for a perfect collection of functions.

Circularity is exactly what we want to avoid if we're trying to come to a mutually agreed-upon definition of something. I'm trying to take this nebulous notion of "depth" and tie it to objective quantities like win rates and position spaces. If all our fuzzy concepts are only defined in terms of other fuzzy concepts, we'll forever run into situations where two people seem to be disagreeing about something but in fact they just using the same word to mean two different things.

quote:

I guess I still don't fully see the relationship of effort to skill level. If we are purely defining skill level as tiers of winning percentage and then correlating that to effort spent getting to some discrete skill level I feel like it doesn't account for a set of strategies with a sort of RPS relationship where more basic strategies can have higher winning rates over more complex strategies but lose to middle-of-the-pack strategies. Even when I generalize it down to saying that more effort boosts a percentage chance of being able to play a specific strategy as opposed to "unlocking" a strategy, I feel like I could create a set where effort spent has an inverse relationship to winning percentage?

A truly good player doesn't always employ the same strategy, though. Part of the skill of a game is identifying the strategies your opponents are employing and choosing a good counter. You're right, however, that one-dimensional players who are very good at one strategy but never do anything else can throw a monkey-wrench in the definition, as it might be possible to find players A, B, C such that player A beats player B 60% of the time, player B beats player C 60% of the time, and player C beats player A 60% of the time.

I suppose what we have to do then is define skill levels in terms of player groups, rather than individuals:

A group of players is said to be of the same skill level if every player in the group has a total expected win rate of 50% when playing an equal number of games against each other player in the group.

Thus, the group {A,B,C} above is of the same level, in that each of them beats one guy 60% but only wins 40% against the other, for an average of 50% for each of them.

We'd then define relative skill level between groups the same way - Group X is one skill level ahead of Group Y if the total win rate is Q after a series of games pitting each player in Group X against each player in Group Y. (Where Q is the previously-defined win rate: 0.5 + S/6, with S being the Effective Skill Factor)

EDIT: Of course, this doesn't actually solve our problem, because you could have a whole group of players who all play like player A above, and another group who all play like B and another group who all play like C. So I guess you have to go a little further and say that to establish a skill level, you have to find the largest possible group of players who all have a 50% win rate when playing equal games against everyone else in the group. Although this sounds impractical to the point of impossibility to do in reality, it's pretty close to what will happen naturally if you have a large pool of players and give them ELO ratings, and then study win rates in order to map ranges of ELO rating to our definition of "skill level." (i.e. maybe a typical player in the 1600-1750 range wins 66.7% against a typical player in the 1500-1600 range, so those are two skill levels, etc.)

xopods fucked around with this message at 19:06 on Nov 26, 2012

Adbot
ADBOT LOVES YOU

xopods
Oct 26, 2010

I played a guy's prototype this weekend. It was a realtime negotiation game, but for me the neat thing was that it was actually a pair of simultaneous, linked games. The players were divided into two groups and, although players in the two groups interacted with one another, in the end, there would be two winners, one from each group. It strikes me as an interesting and almost completely unexplored niche.

In the game in question, the two groups were landlords and merchants. Each was competing to become the richest within their group. The landlords made money by buying properties and renting them to the merchants, and the merchants made money by renting valuable properties and earning more money by using them than they paid in rent. But there are all sorts of other themes that could work...

For instance, you could have a game in the evolution genre, but with two classes of players, herbivores and carnivores, with one winner from each group... the winning carnivore player would be the one who successfully ate the most herbivores, while the winning herbivore would be the one who got eaten the least.

Anyone care to brainstorm additional ideas that would fit this model? Anyone know of any published games using such a system?

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply