Theme and mechanics

[Thierry Michel]

In http://www.indie-rpgs.com/viewtopic.php?t=6878"> this thread , the recommendation for a budding game designer is to "imagine the scene in your head and play it out as a game but without mechanics".

Another way to do things would be to tinker with game mechanisms until one has found something that works(that is makes an interesting sub-game, for instance) and then think of a theme that fits with it.

That's apparently the way "German games" designers work. Is that really inappropriate for a RPG ?

[Mike Holmes]

I have this vague sense that it won't work. It might be predicated on an irrational association that you've pointed out with GAGs (German Abstract Games). They don't evoke their subject matter much at all, and as such, I can't see them working to create Exploration.

Basically, all RPG rules are built around what's being Explored. In selecting to have a combat system, for instance, you instantly say that the game is about combat. So you really can't avoid chosing the subject matter as you design. I think...

I mean, isn't this how you make a Generic game? Won't that be what you get?

But, heck, give it a try. What can it hurt? :-)

Mike

[John Kim]

Basically, all RPG rules are built around what's being Explored. In selecting to have a combat system, for instance, you instantly say that the game is about combat. So you really can't avoid chosing the subject matter as you design. I think...  

Well, no.  By deciding that a system is a "combat system", you are doing it wrong.  The key to this approach is that you work on only the actual mechanics, not the meaning of it.  So you might have a way to roll dice, and a stat which is lowered through rolling dice -- but this could in principle be used for negotiation, psychic reading, or an infinite variety of things.  

I mean, isn't this how you make a Generic game? Won't that be what you get?  

Again, no.  The better parallel in existing RPGs would be a thematic game using a house system, like say Tribe 8.  However, I suspect that these aren't usually designed this way -- i.e. Tribe 8 is trying to retrofit and modify the Silhouette mechanics to work a fantasy world that I suspect they came up with separately.  This approach goes a step farther in one way: you can make up whatever setting, theme, color, and other elements freely to fit the mechanics.  You can even go further and redefine any in-game meaning of the mechanics.  i.e. So you can define the attributes to anything you want, as long as they have the same mechanical effect, for example.  

I have certainly seen this to some degree -- in that I have used generic systems, and observed that they work better for some settings and genres than for others.  I'm not sure the approach would work for an RPG in its most pure form, but it could work in a hybrid form:  i.e. You tweak and adjust the mechanics some after you come up with the setting and theme.  

----------------

Incidentally, I have frequently used a similar approach in designing my PCs in the past.  That is, I would first look at the system and pick what mechanics I wanted to exploit.  Then I would design my PC personality, background, and theme around that.  Certain people find this shocking, but I have developed a lot of excellent PCs this way -- and by excellent I mean with very deep personalities and powerful themes.  Basically, arbitrary restrictions often spur creativity rather than inhibiting it.  It takes a really high level of constraint, I find, to actually inhibit creativity for me.

[Thierry Michel]

The key to this approach is that you work on only the actual mechanics, not the meaning of it.

Spot on.  The idea is that after figuring out the system, you decide what it represents.

I have this vague sense that it won't work. It might be predicated on an irrational association that you've pointed out with GAGs (German Abstract Games). They don't evoke their subject matter much at all, and as such, I can't see them working to create Exploration.

I have the same feeling, but I'm not sure why. Indeed German games systems are often loosely linked to their theme, but not everytime, and not more so than, say, the central mechanics of Herowars for instance - to me, at least.

[I just noticed it's my 99th post - now I have to take a break and think of something really intelligent to say next - could take a long time.]

[Mike Holmes]

Good points, John. I suppose if you kept all setting material out of the mechanical design you could get somewhere. But then, again, I'm doubting how well it will match anything you decide to say that it's about. Still, I'd like to see an attempt.

I have the same feeling, but I'm not sure why. Indeed German games systems are often loosely linked to their theme, but not everytime, and not more so than, say, the central mechanics of Herowars for instance - to me, at least.

I'd disargree strongly. The thing that makes the central mechanic of HW relate strongly to the subject material, is that chargen involves enumerating whatever abilities you like from that setting. In doing so, you tightly link the imagination of the setting in question to the mechanics.

OTOH, the HW system is very generic by itself (and maybe that's what you mean). I can see it being useful for just about any setting; and indeed, I'm using it currently to run the old ICE setting Shadow World. Meaning that if I'd come up with HW as a system first, that it wouldn't have evoked the Gloranthan setting at all, nor any sort of game particularly. Perhaps if you include the chargen related to Cultures. But even that I see as fairly universal.

Mike

[John Kim]

I suppose if you kept all setting material out of the mechanical design you could get somewhere. But then, again, I'm doubting how well it will match anything you decide to say that it's about. Still, I'd like to see an attempt.
...
The thing that makes the central mechanic of HW relate strongly to the subject material, is that chargen involves enumerating whatever abilities you like from that setting. In doing so, you tightly link the imagination of the setting in question to the mechanics.

OTOH, the HW system is very generic by itself (and maybe that's what you mean). I can see it being useful for just about any setting; and indeed, I'm using it currently to run the old ICE setting Shadow World. Meaning that if I'd come up with HW as a system first, that it wouldn't have evoked the Gloranthan setting at all, nor any sort of game particularly. Perhaps if you include the chargen related to Cultures. But even that I see as fairly universal.  

Your statements here seem blatantly contradictory, so I think you're still not getting something.  Your use of the Hero Wars "system" for Shadow World is I think a direct example of what we are talking about.  

I think the problem might be in the use of the word "system", which is clearly not well defined in this system.  In a broad sense, system can definitely include setting.  Here the idea is to start with pure mechanics: what the player physically does and calculates.  This cannot include any game-world interpretation.  For example, you can specify that there are six attributes rated 3-18, but technically it goes beyond pure mechanics to say that one of them refers to the physical strength of the character.  For Hero Wars, pure mechanics would include things like the 1d20 roll under target number, along with all the rules for AP loss, bumps, resistance, and so forth.  It would also include things like the number of words to use in the character description (100), what is supposed to be included in that, and the use of keywords for culture and profession.  However, none of the keywords themselves are included in pure mechanics.  Pure mechanics could also include the character sheet design, since that is a physical thing which the player holds.  However, any descriptive labels (i.e. what the stats mean) aren't part of the pure mechanics.  

Now, the question is, could we start with a set of mechanics like the above without having a world in mind -- and then design a game concept, theme, and setting which it works for?  Your Shadow World game, if it is successful, sounds like a positive example that this is possible.  

It certainly sounds like an interesting exercise at the very least, because it questions the role of pure mechanics.  i.e. What difference does it make to roll 1d20 under target number vs rolling a pool of d10's and looking at the highest number, say?

[damion]

I think it is very difficult to seperate mechanics and setting.  Basicly, as soon as you have defined the mechanics, you have partially defined the setting, as the mechanics define the range of possible results. By saying that there are six attributes you restrict the granularity of the setting, i.e. you arn't going to have 3 mental attributes, 3 attributes that define bodily motion, and 3 that describe physical charitaristics.  This influences the setting, because the mechanic must be able to model what you want.  I.e. if you wanted, say detailed magic, with many possible outcomes, you would not want a mechanic with only 4 possible results.  There are way's around this, by making a tree of results, but this implies that this part of the system is important, becasue it has this involved mechanic. (i.e. Mikes Combat Systems Rant).
As soon as you start defineing ranges of possible outcomes, you are influencing the setting.


My point is a mechanic doesn't really suggest a setting, it just makes some settings harder to do.  
If you started with a mechanic and madeup a setting, vs starting with a setting and making up a mechanic, I think you'd get the same setting either way (What the game designer wanted)  but a different mechanic.

[John Kim]

I think it is very difficult to seperate mechanics and setting.  Basicly, as soon as you have defined the mechanics, you have partially defined the setting, as the mechanics define the range of possible results. By saying that there are six attributes you restrict the granularity of the setting, i.e. you arn't going to have 3 mental attributes, 3 attributes that define bodily motion, and 3 that describe physical charitaristics.  

Exactly!!  The point is that you cannot take a set of mechanics and expect it to work equally well for any arbitrary theme, setting, conflict, and so forth.  Thus, for a given set of mechanics, there is a type of game which it works better for.  This approach suggests optimizing that: actively working to make the theme, setting, and conflict fit with the mechanics rather than expecting it to just work.  

My point is a mechanic doesn't really suggest a setting, it just makes some settings harder to do.  If you started with a mechanic and madeup a setting, vs starting with a setting and making up a mechanic, I think you'd get the same setting either way (What the game designer wanted)  but a different mechanic.  

I'm not following this.  You seem to be pre-supposing that the game designer wants a particular setting -- which is exactly the opposite of the suggested approach.  You are right, a mechanic doesn't define what the setting is, but it makes some settings easier to do.  So the game designer creates his setting to be easy for a given mechanic.

[Mike Holmes]

Where some of the confusion lies, John, is that I think that James is assuming (and I definitely am) that the idea is that the mechanics will suggest an entirely new setting and other details. That is, you get done with the "pure mechanics" and then look at them, and see a world that will fit them.

With Hero Wars, had I created them, I'd have looked at them and seen that they support any world, not one specific world. This seems generic to me.

Thierry said:

Another way to do things would be to tinker with game mechanisms until one has found something that works(that is makes an interesting sub-game, for instance) and then think of a theme that fits with it.

That implied to me that the mechanics would suggest the theme.

But if all we're doing is making a generic system, and then tacking whatever theme on it, then, yes, of course this is possible. Universalis is just the first part without the second. We let the players chose the theme in play.

Mike

[John Kim]

Where some of the confusion lies, John, is that I think that James is assuming (and I definitely am) that the idea is that the mechanics will suggest an entirely new setting and other details. That is, you get done with the "pure mechanics" and then look at them, and see a world that will fit them.  

OK, no, I agree that's silly.  Obviously a fictional world is not going to magically spring into being from looking at some pure mechanics.  You have to create and imagine the world, theme, conflict and color.  It's work.  

However, I would argue that this work can be guided.  That is, it is possible to create a world, theme, etc.  which doesn't fit with the mechanics you have chosen.  Conversely, you can come up with a world and theme which fit especially well with the mechanics.  

With Hero Wars, had I created them, I'd have looked at them and seen that they support any world, not one specific world. This seems generic to me.  

Are you saying that system doesn't matter?  i.e. Suppose I wanted to play modern-day sorcerers who make bargains with demons at the risk losing their humanity.  OK, so I could use the Hero Wars mechanics for this.  However, I have a sense that actually it isn't the best fit.  I get the feeling that maybe a different set of mechanics would be better for that.  So I try something else.  Maybe I try a gritty western game, but again I'm not sure it is an ideal fit.  

As a specific example, I had considered Hero Wars for my current campaign which is set in a mythic alternate history where the Icelandic Vinland colony flourished.  It is set in 1392 in the Hudson river region.  Here I was using the more common approach: I had a specific setting and theme in mind, but was searching for a system.  However, there were a number of things about Hero Wars which I thought didn't fit well with what I wanted.  So I decided to go with a different system (we can discuss this in more detail if you want).

[Mike Holmes]

Are you saying that system doesn't matter?

Hardly. Just that Generic systems tend to by their nature not be better for one setting than another. That's their nature, and are in fact designed that way. A generic system will do a better or worse job at being generic.

Certainly there are more generic games than Hero Wars. It's a spectrum that ranges from a game like FUDGE as a toolkit to play in most settings, to a game like My Life With Master, which couldn't be used for anything else. Note that it's commonly cited that GURPS isn't nearly as generic as some other games. It just tends to one end.

My point here has been that one could come up with FUDGE using the "create the mechanics first" approach. But to get to MLWM, as soon as you've created the love mechanics, you're on the road to a very specific game, and no longer creating "pure" mechanics. Think about Call of Cthulhu. As soon as you create the Sanity mechanic, which is not a special application of the general rules but has details all it's own, you've gotten away from "pure" mechanics.

It's hard to imagine a designer making the BRP rules, and then saying, "Hey, what if we had some ablative stat that you also roll under to see if some condition occurs that causes it to ablate. Then you roll ablation based on some magnitude of the affecting condition, and if the ablation loss is large at any one time, then the recieving party incurs a temporary effect similar to the permenant effect that they will recieve if the overall stat is even reduced to zero." It's a very detailed mechanic, and since it isn't inspired by anything, it doesn't seem likely that it would occur with random development. If it did, the designer coud look at what had happened and say, "Hey, that would be cool for Sanity." I just think that'd be rare. Most likely the rules that would get developed would be the sorts of general rules that could apply to large areas that exist in most games like resolution.

So it seems to me that we're just talking about designing generic games, here. Which is fine, but hardly new.

As I've said, however, it may turn out that someone could throw together a purely mechanics system, and that they can cobble together some setting that fits it particularly well. Hard to say without seeing it done. So are you goint to try it out, Thierry? :-)

Mike

[Thierry Michel]

So are you goint to try it out, Thierry? :-)

Actually, when I was talking about mechanics I was thinking of something more general. If you say "there are six attributes" you're already pretty specific - no matter what name you give to the attributes.

For instance:
starting with the basic process  "Roll die and compare to value, low wins -high fails".  Now, it strikes me that the primary characteristic of the system is the independence of the results (in the statistical sense), that is the probability of failing/winning is the same whatever the order of the rolls. Tinkering with the system, I might to try something where a failed roll makes further rolls more difficult, for instance. So now I have a spiralling system.

Now, what kind of story would fit such a system ? I know - descent into madness!  Et voilà, Call of Chtulhu San system.


And yes, I might try something like that but right now I only have only the very general principles of the type of resolution system I'd like to experiment with.

[contracycle]

I've approached this thought rather differently.  It seems to me that a mechanic has "feel" to it that goes beyond what specific outcomes are or how likely they are.  We have a tendency to be a bit phobic of breakpoints but I think that where the breakpoints (or other limits that arise purely as artifacts of the sysytem) fall is part of the, umm, experience of the mechanic.  

Therefore I think it is possible to identify in a setting a locus of tension (where was that thread on tension in the game space) to which a mechanical limit or breakpoint is to be located for setting-evocative effect; or conversely develop a mechanic which has an array of breaks and limits and then layer a setting on top of that in whch points of tension correspond with the mechanical architecture.

Hmm.  I'm struggling to say what I mean.

[John Kim]

Just that Generic systems tend to by their nature not be better for one setting than another. That's their nature, and are in fact designed that way. A generic system will do a better or worse job at being generic.

Exactly.  That's what makes this approach totally different from a generic system.  A generic system designer will look at a set of mechanics and asks "Can these be used for all settings?"  If they can't, then the generic system designer considers it a failure and changes things.  This approach is the opposite.  You look at the mechanics, and you consider it a success if they fail at being generic.  You then say "This doesn't work well for all settings.  So what settings does it work best for."  

My point here has been that one could come up with FUDGE using the "create the mechanics first" approach. But to get to MLWM, as soon as you've created the love mechanics, you're on the road to a very specific game, and no longer creating "pure" mechanics. Think about Call of Cthulhu. As soon as you create the Sanity mechanic, which is not a special application of the general rules but has details all it's own, you've gotten away from "pure" mechanics.  

OK, so I'm not familiar with MLWM, but CoC seems like a quite possible application of this.  I guess I'll illustrate with a hypothetical talk between two designers (Anne and Bob) --

Anne: Hey, Bob.  You know, we've been doing these monster fights with BRP for a while, and I'm dissatisfied with the damage rules.

Bob: What's your beef?

Anne: Basically, damage never really feels threatening.  I mean, sure you can have a big monster which does a lot of damage -- but that just seems like a killer GM.  It wears down your hit points, but as long as it doesn't go over your threshold, then there is no effect.  If you survive then you can just have your PC rest a while to heal up.  I guess I'd like a mechanic where even a small bit of damage seems more palpable.  

Bob:  OK, I can see that.  So maybe we can introduce wound penalties for being partly damaged?  

Anne:  I don't know.  That doesn't sound like what I'm looking for.  OK, so maybe you have a stat -- let's call it Integrity for the moment -- which can be damaged, but damage is increased if you fail an Integrity roll.  So it's like a slippery slope.  You really want to keep every point, because every point lost makes it worse later on.  

Bob: I see.  That's kind of cool.  So maybe double damage if you fail your Integrity roll.

Anne:  Maybe a monster would have two damages, a low one if you make your roll, and a high one if you fail it -- maybe 1d4 and 1d12.  That allows more variation.  So you could have a monster which is threatening by having normal low damage, but really dangerous high damage.  

Bob:  Got it.  So instead of just hit points we have an Integrity stat.

Anne:  Hold on.  That's only half of my problem.  You still have the problem that the players can say halfway through the session "Our PCs sit and rest for a few days."  Then all the damage gets healed away.  Because of the feedback, they're just encouraged to do this even more.  

Bob:  OK, we'll hold off on what it would be for a bit.  So you're picturing something which can't just be recovered by simple action.  That means Integrity loss is really important.  Maybe you get it as a reward at the end of an adventure, but not just from PCs resting.  

Anne:  Hey, that's good.  Then it's really nasty and threatening and scary, but the players are encouraged to push on because the only way they're going to get rid of their damage is by finishing the adventure.  That's the opposite of hit points where you're rewarded for going away and resting.  

Bob:  Interesting.  But this is kind of depressing.  I mean, if it really is nasty then you're going to have a high body count.  Plus even if you survive, you might lose more than you are awarded at the end of the adventure (assuming it is a fixed reward).  

Anne:  Darn.  So you don't like it?  

Bob:  Actually, I do like it.  But you're right, this isn't just a way to make scarier monsters.  I think this is something for a real horror genre game.  Maybe you could be fighting poisonous aliens, and the only cure is hidden deep in their nests.  

Anne:  That sounds kind of trite to me.  What about something like you're losing your humanity, and only by victory over evil can restore it.  Or maybe something similar?

etc.  etc.

[Mike Holmes]

You then say "This doesn't work well for all settings. So what settings does it work best for."

It's an interesting idea, I'll give it that. And you're example seems sorta plausible if still in a context which I think Thierry won't have. Hmmm.

Here's a mechanic as an experiment. Each player has a hand of cards. They have different numbers on them. Each "round", each player passes a card to his left. What the player's are looking for is matches in the numbers. Each set of cards is worth the square of the number of cards in the set, times the number on the matching cards. So three fives is worh 3^2*5 or 45 points. The cards are numbered one to ten, and there are ten of each in the deck. Each player has eight cards to start. Something happens when a player gets to 100 points.

OK, there's a pure mechanic. To be fair, I stole it from actual games in parts, not wanting to try to make a mechanic that was impossible to associate with something (I think I could make such a mechanic if I tried).

So, what does this mechanic suggest in terms of incorporation into a RPG?

Mike