Skip to Content
 

Settlers vs Ra

13 replies [Last post]
zaiga
Offline
Joined: 12/31/1969

A discussion of Spielfrieks about whether Ra contains more interesting decisions that Ra made me write the following article. However, I think it is more appropriate on this forum, as it delves so deep into gamedesign and gametheory and I don't know if such a "deep" article would be fully appreciated on that discussiongroup.

Anyway, here it is:
----------------------

Alan Kwan's original post on this subject contained a lot of bold statements, such as that Settlers is more *fun*, but that Ra requires more *skill* and that there is a roughly same amount of luck in both games.

I am not going to make such statements, because they are very subjective. However, it is interesting to compare Ra and Settlers, because, although they both contain a significant amount of luck, there is a fundamental difference in the place where the luck manifests itself.

In Settlers the resources you acquire are determined by a random event: the dice rolls and by your opponents (if they are willing to trade, although you do need some resources to start with). You do have some say over which resources you might get by placing villages on "good" hexes during the setup phase, which is an important phase and requires a lot of skill to do right, but after that it is mostly up to the dice.

In Settlers you try to convert your resources (which you get through a random process) as efficiently as possible into more resources, until you have enough resources to end the game and win it. Turning resources into more resources is typical of a game with a runaway leader problem, by the way, although Settlers does contain some methods to drag down the leader somewhat such as the trading mechanism and the robber.

Ending the game by reaching or fulfilling a certain condition is what I call a binary wincondition. You have either won the game or you haven't. Another example of a game with a binary wincondition is Attika, where the first player to either connect shrines or place all his buildings wins. This is fundamentally different from a game with a victory point system, in which the player who ends the game not necessarily always wins it (ie. "Through the Desert", "Lost Cities" and... "Ra").

In Ra the resources that are at you disposal are not random, everyone gets a stack of suns that are of equal worth at the beginning of the game(well, more or less, there is a small difference in the numbers). You try to use your minimal amount of resources as efficiently as possible to turn them into as much Victory Points as possible within the give timeframe (3 rounds, each of which end when a certain number of Ra tiles has been drawn). Of course, in typical Knizia fashion, resources are also worth victory points at the end of the game, which makes sure that players are not "wasting" their resources on VP's when the end of the game nears.

In Ra the time you have to turn resources into VP's is a random thing, because you never know when that last Ra tile will come up. Also, the victory points come up in a random manner and how much resources you have to spend to get a certain amount of VP's depends, up to a certain extent, also on your opponents and what resources (suns) they have left, because of the auction mechanism.

In Ra you already see two dimensions that Settlers combines in one thing: a timeframe and a victory point system. I know that Settlers also has VP's, but in Settlers the VP's are used to determine whether a player has already won or not. So, basically the dimensions of time and VP's are intertwined in Settlers, whereas they are separate things in Ra. This typically means that Ra requires a different mindset than Settlers and that the goals and priorities in each game are of a different nature.

For example, in Settlers your goal might be to upgrade that village as soon as possible and with the minimal amount of resources into a city. Once you have accomplished that, you can use the extra resources gained from that city to build roads and a village to create a port. Of course, when the opportunity presents itself (maybe because of trades or unexpected die rolls or because certain spots get blocked or because you have 8 cards in hand and just want to get rid of some of them) it might be wiser to build a second city instead of that harbor or to buy developmentcards.

Basically, what you are constantly doing in Settlers is converting resources as efficiently as possible into other resources. This is certainly not an easy process, because the resourcesystem is complex (is an extra city better than a port, for example) and you constantly have to re-evaluate your goals based on what resources you get, what the boardposition looks like, etc. Only at the very end you are actually interested in accumulating things that do not produce extra things, but just give you enough VP's to win the game. And sometimes it is necessary to "steal" VP's from other to prevent them from winning.

In Ra you are constantly evaluating how much resources a certain set of tiles, that translate more or less directly in a number of victory points, is worth for you and your opponents and you have to base your decision (placing a bid or not and if yes then how much?) on that. If you could somehow find out the "magic formula" for this (which undoubtly is a very complex one) you would have a good chance of winning, were it not for the random order in which the tiles are drawn and the random endings of the rounds (although you could factor this into the equation too of course).

As you see, it is impossible to say which game has more interesting decisions and more luck as the luck and the decisions in both games appear at totally different levels in their resourcesystems.

I am not advocating one game over the other, because tastes just happen to differ. I am just trying to explain the fundamental differences in their resourcesystems, so that people hopefully can better understand *why* a certain mechanic in a certain game works a certain way.

- Rene Wiersma

jwarrend
Offline
Joined: 08/03/2008
Settlers vs Ra

Great article! I think the point that resonated with me most is the separation along the "fault line" of game end styles -- the "race" style (first player to do X wins) and the VP style (when it's over, we count 'em up). I think this is an important distinction that can be fruitfully analyzed and evaluated -- in what kinds of games is the former "better" and in what games is the latter "better"?

I would also probably make a distinction in the latter case between games like El Grande or Wallenstein where the game end is inevitable and out of the players' control, and games like Puerto Rico or Acquire where it is subject to the players' control. And the latter could be subdivided even more, I suspect, because in Puerto Rico, for instance, the game can end when the VP chits run out, creating a meta-race in which game-ending and VPs are related (though less directly, and with less of an immediate ending than a true race game, since other players can still get VPs in PR even if the chits run out). But there's the other kind of "player controlled" race game like Web of Power that's more configurational -- when the cards run out, the game ends, but players have control, based on how many cards they draw, etc, of when that's likely to happen.

So, I might say there are really 4 (at least) categories:

The "true race" -- the game ends when one player accomplishes goal X.
Examples: Settlers, Mississippi Queen, Attika, Citadels

The fixed length game -- the number of turns is fixed and does not vary from game to game.
Examples: El Grande, Carcassonne

The player-controlled, VP race -- Player actions determine the game end, and players can bring the game to an end by selecting actions that also generate VPs.
Examples: Puerto Rico, Euphrat & Tigris (?), Acquire (making all chains "safe" or one "41+" gives the owner money)

The player-controlled, non-race -- Game length varies depending on players' actions, but the actions that bring the game to an end don't in and of themselves generate VPs.
Examples: Web of Power, New England

Are there more categories than this? Have I properly demarked the categories here? Are there better names for them?

Edit: Well, there's one obvious category I omitted, the "non fixed-length non-player controlled game", where the end of the game is triggered randomly. An example would be Domaine/Lowenherz. I think this is functionally the same as the "fixed length game", but it does provide a very different kind of player experience, so it probably deserves its own category:

the variable-length, non player-controlled . A random event triggers the game end.
Examples: Lowenherz, Union Pacific

And then the question becomes a design question -- which is "best"? Obviously, not a sensible question. Maybe a better question (or rather, a discussion subject) is, what kinds of design problems can crop up in the different styles? What kinds of player experiences go well with the different styles?

As an obvious example, the "race" style can lead to a big "hit the leader" problem where everyone just brings the leader back to the pack, then the next guy tries to win and he's brought back etc. Yet, it's also atmospheric if it's indeed a "race" feel that is trying to be created. (ie, a car racing game that you won based on VP rather than time would be unconventional).

Cool article, Rene. I welcome your thoughts (or others', of course) on this further specification of your categories...

-Jeff

Brykovian
Brykovian's picture
Offline
Joined: 07/21/2008
Settlers vs Ra

Very good article, Rene ... and an excellent follow-on, Jeff ...

The only thing I'd say in response to the questions at the end of Jeff's post is this: Which "fits best" for the specific game you're making? I think this *is* a sensible question when focussed on a specific game.

As mentioned, each of the endings has a certain flavor ... each might be the "right" one depending on the theme and/or feel of the game. For example, in a 2-player tactical battle game, having a "race" to defeat the other player makes a lot of sense.

Also, the problems that may arise from one type of ending may be different depending on what type of game it is being affixed to, and what the intended audience is.

Great discussion starter, guys ... I feel potential wiki material will be gleaned from this thread ... ;) :D

-Bryk

FastLearner
Offline
Joined: 12/31/1969
Settlers vs Ra

Kudos to Rene and Jeff. I don't have anything to add at this point, but did want to thank you for some great, thought-provoking stuff.

-- Matthew

zaiga
Offline
Joined: 12/31/1969
Settlers vs Ra

Cat... ate.... long... post.... :cry:

Oh well, at least now I know what I'll have for dinner tomorrow :twisted:

Let's try again, it is not as if I have something better to do...

In this post I'll expand my theory and show some resource schema's of different games.

Before I do that, I want to explain some different terms I'm using.

When I'm talking about resources I mean all the things that are at a player's disposal when he needs to make a decision and of which he has a limited amount. Resources can be tangible things such as cards, chips, playmoney, but it can also be more abstract things such as actions, action points, board position, even seating position and whatnot.

With time I mean the number of turns (or rather: decision moments) a player gets before the game ends. Time, in a sense, is also resource, because it is something that players have a limited amount of. However, I feel compelled to distinguish time from other resources for a number of reasons:

- Time is a universal resource that is an important feature of many games
- Time is a tool for a designer to control the length of a game
- Time introduces a story arc to a game, in the sense that at a different stages in the game players will have different goals and objectives
- Time introduces a sense of urgency, the idea of "getting things done in time" and appropriate rewards and penalties

A victory point is that which you count up at the end of the game and the player who has the most of wins the game. Victory points don't always need to be VP chips, they can be money or a position on a score track, etc. When I talk about victory points I don't mean the number of points of a single player, but rather the whole delta of all victory points of all players combined. For example, when I gain a point it is the same as everybody else losing a point.

A victory condition is the condition which a player must fulfill in order to win the game and thereby also ending it. Some games also use VP's for this (when someone reaches x vp's he wins the game), but in the context of this theory they fulfill a different purpose than "true" victory points.

In the schema's you will see some "blobs" which depict the influence that random factors (var) and opponent's (human) have on a player's resources or on the time of a game. The bigger the blob the more influence. The size of a blob is mostly a subjective thing of course, but it is useful to illustrate a point.

All right, here is my first resource schema.

You see that there is a large random factor (the dice rolls for resource cards mostly) influencing the resources a player has at his disposal. There's also a large human element influencing the resources of a player. This interaction be through the use of the robber, trading cards and blocking spaces on the map.

The looping arrow on resources means that players can turn their resources into more resources. Resources in this game are of course the resource cards, but also development cards, villages, cities, board position, etc.

The looping arrow on resources indicates that this game might have a runaway leader problem, because resources are used to gain more resources to gain more resources, etc. The game tries to migitate this by letting other players have a large influence on other player's resources. Or, to put it more bluntly, players are able to bash the leader. The fact that a player has to discard cards when he has more than seven in hand when a 7 is rolled also puts a cap on the runaway leader problem.

A player can also use his resources to try and advance towards his victory condition. In Settlers, most things that advance a player towards a winning condition are also resources themselves (villages, cities), so the decision whether to go for the victory condition or more resources is often easy, although deciding what is the most efficient way of accomplishing that is not simple. Only at the end, when a player can win the game it makes sense to obtain things that are not resources but only advance towards the victory condition. Of course, sometimes it is also necessary to obstruct another player, to keep him from winning, when you cannot win yourself.

A game that has resource schema very similar to that of Settlers is Monopoly. In Monopoly a player also tries to convert his resources into more resources, until someone wins the game by reaching the victory condition (which is the bankruptcy of everyone else). I think that in Monopoly the random factor on a player's resource is a lot bigger and the human factor smaller.

A game that is often considered the polar opposite of Monopoly (and perhaps Settlers too) is Chess, but its resource schema is remarkably similar:

You see that in Chess players also try to use their resources (pieces, boardposition, number of possible moves) to gain more resources (less pieces for opponent or trade a lesser piece for an opponents better piece, better boardposition, more moves). Chess also has a runway leader, small mistakes often lead to a loss, which is the reason why players often concede long before the game is technically over. If no player is able to run away with the game it ends in a draw.

There is no randomness in Chess (except for who plays white and black, which I omitted for clarity), but the influence a player has on the resources of the opponent is huge.

------

So far, I have shown some examples of games that have binary winning conditions. In my next post (I better post this before my cat eats it again) I'll show some more schema's of games that have time as a prevalent resource and I'll explain how that affects play.

- Rene Wiersma

zaiga
Offline
Joined: 12/31/1969
Settlers vs Ra

A victory point based game which makes clever use of time is Ra.

The most resources important resource in Ra are the sun tiles, of which you have a limited amount (three or four, depening on number of players). Other resources are obtained god tiles, the option to start an auction when it's your turn OR to draw a new tile, the tiles and sun available on the track and perhaps some other less obvious and less tangible resources as well. Obtained tiles more or less directly translate into victory points.

What resources you have at your disposal is influenced by the random draws of the tiles and what other players do, especially how they bid and whether they start auctions or draw new tiles.

Time is also very important. You only have a few suns to bid with and you want to squeeze the most out of them. However, how much time you have in each round depends on the random drawing of Ra tiles. When the last Ra tile is drawn the round is over and any unused suns are basically wasted. So, you want to time it so that the last Ra tile will be drawn right after you have used up your last sun, your most important resource. This makes for some great push-your-luck scenarios, but it also makes the game a bit random.

Although time is largely determined by a random factor in each round, it somewhat evens out over the course of three rounds. If in one round the Ra tiles come up quickly, there a large chance that in a different round they come up more slowly, but not always of course.

A last schema that I want to show is that of the great Puerto Rico:

At has in lot of common with the Settlers schema, but it is also quite different.

First of all, the random factors in Puerto Rico are very small. There's only a bit of randomness in plantation draws. I think the influence players have on each other are rather big, although you have more influence on your neighbour directly to the left than the other players.

Like Settlers, Puerto Rico has a looping resource arrow. This points at a runaway leader. However, unlike Settlers Puerto Rico has the element of time. This means that a player may runaway with the game resource wise, but he still has to convert those resources into victory points before the game ends. In that sense Settlers lacks a dimension when compared to Puerto Rico.

Players have some influence over how long the game will take, but because there are 3 different end game triggers one of them will trigger in a reasonable amount of time, depending on what strategy players are following.

----

I hope this was all helpful. It might be useful to draw such a resource schema of your own game and see how it works.

I agree with Jeff that it is always good to know how a game works, but that it is actually more interesting to talk about why some schemas better than others and what the advantages and disadvantages are of the different schemas.

- Rene Wiersma

Torrent
Torrent's picture
Offline
Joined: 08/03/2008
Settlers vs Ra

What might be interesting is to do a schema for something like Monopoly. It's a game often referenced around here as an example of almost what not to do. It might be really neat to see how different (or perhaps how similar it is to other games). I admit I don't exactly understand where those schemas are coming from.
My wonder is if a schema is merely an analytical tool, or a specific schema can lead to a specific feel of a game. Thus things with the same schema will feel similar.

FastLearner
Offline
Joined: 12/31/1969
Settlers vs Ra

Excellent analysis, Rene, great insight. I have some thoughts on expanding on it that I hope to post here when I have a bit more time.

Torrent, a Monopoly schema (as Rene mentions in the first post) would look just like Settlers but the VAR blob would be something like twice as big and the HUMAN blob would be smaller... just how small depends a lot on how you and the other players conduct the game: some people trade with each other a lot and others do no trading at all.

Rene, I suggest that Ra does have a resource loop ( though it's not nearly as strong as most games) because sun tiles buy more sun tiles, and there can be some runaway leader as a result, though there's enough randomness that if you just try to buy high sun tiles with your high sun tiles you'll likely lose the game.

-- Matthew

hpox
Offline
Joined: 12/31/1969
Settlers vs Ra

Thanks for the great posts, they are appreciated! I also agree that certain (how do you call it?) "game flow" lend themselves better for certain mechanics and themes but they are mostly usable with any kinds of themes and mechanics.

Mixed game length could exist too. A game that end when one player wins the race OR a certain number of turns have passed. A game where players can influence the probability a random event ending the game happens.

FastLearner
Offline
Joined: 12/31/1969
Settlers vs Ra

hpox wrote:
Mixed game length could exist too. A game that end when one player wins the race OR a certain number of turns have passed.

Aye, my Everest game works precisely that way: someone reaches the top of the mountain (and therefore wins) or the climbing season ends, in which case the person with the most points (based on reaching certain plateaus first) wins. The game is balanced such that in playtests if the game ends by someone reaching the top he's never been the guy with the most points, and that winning by reaching the top occurs roughly 1/3 to 1/2 the time. Both ways to end the game occur with reasonable frequency.

zaiga
Offline
Joined: 12/31/1969
Settlers vs Ra

FastLearner wrote:
Rene, I suggest that Ra does have a resource loop (though it's not nearly as strong as most games) because sun tiles buy more sun tiles, and there can be some runaway leader as a result

Matthew, I thought about that, but the point is that you are not turning resources into more resources. Rather you turn them into different resources perhaps even better resources and into victory points, but not more resources. Ra doesn't have the snowball effect that Settlers and Monopoly have, for example, that's why I didn't draw the resource loop, although you could make a case that tiles that you keep between rounds and score every round (such as Pharaos and Niles) could count as "more resources". It's gray area.

Quote:

though there's enough randomness that if you just try to buy high sun tiles with your high sun tiles you'll likely lose the game.

True, because you are not converting your resources into victory points in an optimal way within the timeframe of the game.

- Rene Wiersma

jwarrend
Offline
Joined: 08/03/2008
Settlers vs Ra

FastLearner wrote:
hpox wrote:
Mixed game length could exist too. A game that end when one player wins the race OR a certain number of turns have passed.

Aye, my Everest game works precisely that way: someone reaches the top of the mountain (and therefore wins) or the climbing season ends, in which case the person with the most points (based on reaching certain plateaus first) wins. The game is balanced such that in playtests if the game ends by someone reaching the top he's never been the guy with the most points, and that winning by reaching the top occurs roughly 1/3 to 1/2 the time. Both ways to end the game occur with reasonable frequency.

Does it lead to zaiga's beef with Mystery of the Abbey, that a player will only end the game if by doing so he can win? That, to me, would seem catastrophic, if you have players stalling at the top rather than trying to complete the climb, because it would be so athematic. Have you thought about your game in those terms? (It doesn't seem to benefit from the Mystery of the Abbey defense, that "the game is meant to be light and chaotic, and the scoring system somewhat superfluous anyway...")

Game of Thrones also works this way -- the game ends either after 10 turns, OR when one player captures 7 strongholds. I like games that have an "upper bound" like this, as it creates the "go for it!" effect without the endless loop of "take down the leader".

The game I'll be workshopping in January has what is, as far as I know, a unique hybrid of what I called "player-controlled, non-race" and "player-controlled, VP". To tip my hand a bit (which I think I said in my journal anyway), the game is about the 12 disciples, and can end when the actions of Jesus and the disciples have so caught Rome's attention that they are forced to crucify Jesus in the interest of keeping peace and order. But, it can also end when one player, the Traitor, decides that it's in his interest to take the payoff and betray Jesus. So, in a sense, all players' actions can precipitate the game's end, but one player in particular has personal (but not absolute) control over when the game will end.

It's funny, because when I was thinking of those categories yesterday, I was thinking "is there any room for new categories?" (which is what I always think about when anything is categorizable -- have we exhausted the set of what is possible, or is there room for outside the box thinking?) Then I remembered my own game and said, "hey, I guess I've done just that!" I think that the game I'll be workshopping, and Matthew's game, are both examples of hybrids of the categories I outlined; I don't think they are categories in their own right, but I think they are good implementations of more than one ending scenario as a direct way to affect the player experience in an interesting way that inherently creates tension in the game's "timing mechanisms".

-Jeff

FastLearner
Offline
Joined: 12/31/1969
Settlers vs Ra

jwarrend wrote:
Does it lead to zaiga's beef with Mystery of the Abbey, that a player will only end the game if by doing so he can win? That, to me, would seem catastrophic, if you have players stalling at the top rather than trying to complete the climb, because it would be so athematic. Have you thought about your game in those terms? (It doesn't seem to benefit from the Mystery of the Abbey defense, that "the game is meant to be light and chaotic, and the scoring system somewhat superfluous anyway...")

No, it doesnt' work like that. If you make it to the top then you win, period: the points are only meaningful if no one makes it to the top, so there's no point in waiting. Stalling only gives the other players a chance to catch up and push by (which if they're prepared then they can do all in one turn). It does give the player with the most point an incentive to make it more difficult for the players trying for the summit (through either putting a camp on a key tile, making it more difficult for them to put one of theirs in a good spot, or by backing down the mountain a bit and encouraging the weather to be as harsh as possible).

Note that interestingly and stunning (to me, as I'm not a math whiz and am glad I figured this out), the game is designed with some very tight math. In every playtest so far -- and frankly this simply amazes me because I only estimated the math and it came out perfectly beginning with the very first play -- by the time anyone makes it to the next-to-top tier (one step from the top) there are at the very most 2 turns left, and most commonly only 1 turn left. If you weren't prepared to scale it then you'll fail because time will run out. It all works out very tightly and just the way I wanted it.

The major issue with the endgame, though, is that if there are several players clustered near the top (which is also common) and you don't have the most points already, you know that you can only win by reaching the summit, and if it's clear by looking at the situation that, say, you will need at least 3 turns to get everything together and work your way to the peak but the player next to you will clearly only need 2 then there's no point at in in continuing to play because you have a 0% chance of winning. You end up playing out the last couple of turns knowing you've lost. Mind you that's not terribly uncommon in games, but it does make that last 10 minutes or so pretty dull.

But that's another topic altogether.

DarkDream
Offline
Joined: 12/31/1969
Settlers vs Ra

Zaiga,

Just a brief suggestion with the schemes you created. There are games where there are degrees of spending resources to get more resources. You could perhaps have a bigger arrow or bolder arrow where this mechanism is a central part of the game, and have a smaller and less thick arrow for where this mechanism is not in the forefront.

I think the essential idea of your schemes is a good one. To develop it further, I would go ahead and try to catalog some of the more popular games and post them. You may find you may need to add more symbols to the schemes, or introduce new elements.

Also, if you do a lot of games and find that a lot of the schemes look similar yet the games are fundamentally different, you may want to reassess certain features.

I'm a software developer, and when building software (which interestingly there are certain elements that are quite similar to game design) UML (Unified Modeling Language) diagrams are sometimes drawn. The diagrams use a standard set of symbols in a certain format to communicate and clarify certain elements and features of software. Maybe you can go ahead and create a UML for game design which would be extremely useful at a glance to understand some of the fundamental underpinnings of the game.

Obviously you spent a lot of time doing it. Great job, and thanks for contributing your work.

DarkDream

Syndicate content


forum | by Dr. Radut