Talking about Skills again

[Christoffer Lernö]

I'm (as usual) running into trouble because I can't accept the common way of solving a certain mechanic. Actually, this is the mother of all mechanics. I'm talking about skill resolution of course.

Let's leave the issue of skills vs stats aside for a while and just look at the resolution part of it. Let's assume we have a rating of some sorts which says something approximate about your ability to perform something.

Resolution is about using that information and other input to figure out what's happening.

Of my primary interest is the traditional situation with little directorial input but I guess some of these observations might be possible to extend to other situations.

Anyway, I want to start out by looking at a few situations.

Situation A: The Character is wants to jump across a chasm. This is probably the most simple scenarios of them all, because there is no extended contest, there is simply a single try and that's it. A "yes" or "no".

Situation B: The character wants to do the long jump at the olympics. How far does he/she jump? Situation A can be considered to be a special situation of B.

Situation C: The character wants to pick a lock. As far as success goes, it's the same as situation A. You either pick it or don't. On the other hand it's similar to B in that aside from the success/failure there is also a gradual performance scale going from quick to slow. Many games force this skill into an A type of resolution by considering the roll telling not so much if it's possible as if the character managed to do it within a specified time or by simply ignoring the performance aspect of it.

Situation D: The character is walking into an ambush. Possible parameters are: how early does he/she see it and how much of the situation is clear to the character (is it only a feeling or does he/she actually spot some people and if so how many of them?)

Situation E: The character tries his/her skill against another in a friendly arm wrestling contest.

Situation F: The character attacks someone in combat.

Clearly(?) all of these situations are different.

There are a few parameters which pop up:

* Randomness: how random is the situation? Does it depend a lot of luck or a lot on skill. Depending on the rest of the system, simulating situational modifiers might be heavily built into the system.

* Effect on success/failure: How much does the success depend on the skill rating?

* Effect on quality: How much does the quality of the success depend on the skill rating? Is it only possible to do well with high rating?

Let's look at the situations again. I do some rough estimates which you may or may not agree with. Anyway, the estimates are not the point. Anyway, I try to give the dependency as high, medium or low

Code Select Expand


  Random  Success/Fail Quality
A    low        high      n/a
B    low        n/a       high
C    low        high     medium/low
D    high       medium    high
E    low        high      high
F    low/medium high      high


Arguments could be made for other estimates, I know, but it's besides the point. The point is that most games run a "one size fits all" approach to skills.

Let's look at the estimated behaviour of a few methods:

Code Select Expand


Method      Random Success/Fail Quality
BRP         medium    medium      n/a
Shadowrun   medium    medium    med/high
Pendragon   medium    medium      low
opposed     med/low  high/med     n/a
test BRP  

   
And so on. The trend is quite clear. Keep the roll in the middle to be a little applicable to all cases but not perfect for a single one.

It might sound like a good idea, unless of course we consider what actually happens in the situations where they don't work.

Also remember that the difference from working from soliloquy to game system resolution is much bigger if the resolution is strongly random. Add to that that very few system correctly creates the wanted quality/skill dependence for the simple reason that this changes from situation to situation.

The point where most games go wrong is when they try to use a single system for more than one thing.

This is especially the case with heavily fortune based simulationist games.

In those games, rolling the dice might be motivated by a lot of different things.

On the surface we might be doing a Perception Roll to see those bandits to simulate that we might or may not see them, but in reality it's done to differentiate between the characters who discover the bandits and those who don't. It's a way to reward characters who has invested in Perception and punish those who hasn't. And the roll itself is a way for the GM to avoid deciding exactly what the individual circumstances are.

A very common improvised rule I've run and I'm sure other have as well is the simple: "roll a die". From the die roll and the skill you get a feeling of what happened. Sometimes the die is very important, sometimes the skill is, as a GM you decide on the fly.

Usually these things work just as good and often better than any written down rule, because you know yourself how much you want to value the dice and how much you want to value the skill.

Formalized, that improvisation can be written like this.

1. Create a minimum skill need to be allowed to make a roll, x (can be skipped)
2. Decide on a skill level where success, on an average roll will succeed fairly well.
2. If the character has this minimum skill, roll a die (can be anything from D4 and up)
3. Modify the performed skill level up or down depending on roll. As a guideline, the highest roll corresponds to a bare success for a chacter with minimum skill level.

If desired one can create a graded die scale, like this:

Code Select Expand


D20 roll of:
1     -3
2-4   -2
5-8   -1
9-12  0
13-16 +1
17-19 +2
20    +3

or whatever and then increase or decrease the modifier depending on the type of test. So for another those modifiers might be much higher up and down.

Still, it seems like roundabout way of dealing with the problem.

From my own design notes I want something which "seamlessly can go from karma to fortune resolution". I think I remember someone mentioning a game which had this? Does it ring any bell?

Anyway, what I was thinking about was something like this:

Bob the Barbarian (Str 7) wants to armwrestle Eggo the Elf (Str 4). In the real life noone would think there's a FRIGGIN CHANCE IN HELL that Eggo could beat Bob. Bobs bulging muscles are simply too much for Eggo. (Yet many games give a good chance between 1-10% if not more to Eggo to win)

Basically when the difference is too big, just give it to the best.

When we get to a 50-50 situation, like Bob vs Grog the Ork, both having Str 7. Here chances are obviously 50-50. So just let the die resolve it like that.

Questions arise when Tokk the Troll (Str 8) steps up to armwrestle Grog. Does Tokk win automatically? Well in some situations, sure why not? If one wants it a little more random, give it a 25% chance to Grog. (2+ is win for Tokk on a D4, 1 is a win for Grog if Tokk rolls).

In more random situations, say Tokk and Grog play poker. They use their Int (2 and 3 respectively), the chance of Grogg winning might be calculated on a D6. So Grogg having a +1 advantage on Tokk gives him a win on 3-6, or 2 chances out of 3 (if it had been a D4 the chances for Grogg would have been 3 out of 4 instead). Basically a larger die but the same bonuses (the difference between the stats) are choosen to simulate more random situations. For situations with very little difference depending on stat you could roll D20s or something.

However this is only in the opposed situation. You could cover situation A this way by assigning a difficulty to the static value and then doing the same as the oppposed test.

What about situation B? It gets a lot less intuitive. Here you have to get into the guts of the die roll. Basically you have to have to see how much the roll differs from the average value on the die and then figure the stat as being performed that much better or that much worse. For example for the D6.... If Dex 5 represents being able to jump 5+m 50% of the time and Dex 6 is 6+m 50% time, then the modification to Dex is D6-3. So a roll of 4 means you jump as Dex 6, which  should then be 6 meters.

Then look at the situation of finding the vital hint hidden in the sofa. It's a BIG dose of luck. It could be solve by rolling a D12 or something big like that against a difficulty as detailed above, but it obscures what's going on. Here it's not only about performance but about luck as well.

A solution would be let the GM actually roll a difficulty for every player and then let them do a less random test.

Why? Because that way you can resolve the skill check with karma while retaining randomness.

For example, you ride into an ambush. Eggo has Perception of 7 and Grog has Perception 5. The GM rolls difficulty. Eggo gets a difficulty of 8 to and Grog gets a difficulty of 4. What's happening here? Well maybe Eggo was busy paying "attention" to the beautiful princess they saved,  or maybe he just was riding last. Or he was thinking about other things. Grog on the other hand might be high on the a magic potion he likes to imbibe, or his horse seemed unrurly or whatever. He's paying more attention. With a D4 roll we have Grog having 75% chance and Eggo 25% chance despite Eggo would discover more if the circumstances were equal. Also if Eggo do discovers something he probably see more than Grog would.

Well it still has problems, but at least it addresses some of the problems. Anyone out there with some better ideas?

[Le Joueur]

I'm (as usual) running into trouble...

Situation A: The Character is wants to jump across a chasm. This is probably the most simple scenarios of them all, because there is no extended contest, there is simply a single try and that's it. A "yes" or "no".

Situation B: The character wants to do the long jump at the olympics. How far does he/she jump? Situation A can be considered to be a special situation of B.

Situation C: The character wants to pick a lock. As far as success goes, it's the same as situation A. You either pick it or don't. On the other hand it's similar to B in that aside from the success/failure there is also a gradual performance scale going from quick to slow. Many games force this skill into an A type of resolution by considering the roll telling not so much if it's possible as if the character managed to do it within a specified time or by simply ignoring the performance aspect of it.

Situation D: The character is walking into an ambush. Possible parameters are: how early does he/she see it and how much of the situation is clear to the character (is it only a feeling or does he/she actually spot some people and if so how many of them?)

Situation E: The character tries his/her skill against another in a friendly arm wrestling contest.

Situation F: The character attacks someone in combat.

...There are a few parameters which pop up:

* Randomness: how random is the situation? Does it depend a lot of luck or a lot on skill. Depending on the rest of the system, simulating situational modifiers might be heavily built into the system.

* Effect on success/failure: How much does the success depend on the skill rating?

* Effect on quality: How much does the quality of the success depend on the skill rating? Is it only possible to do well with high rating?

...From my own design notes I want something which "seamlessly can go from karma to fortune resolution". I think I remember someone mentioning a game which had this? Does it ring any bell?

...Then look at the situation of finding the vital hint hidden in the sofa. It's a BIG dose of luck. It could be solve by rolling a D12 or something big like that against a difficulty as detailed above, but it obscures what's going on. Here it's not only about performance but about luck as well.

...Well it still has problems, but at least it addresses some of the problems. Anyone out there with some better ideas?

Well, yeah, but I don't think you'd have the same subjective opinion (mine is better for me, mine is not better for you: subjective).  And I think that's one of the two problems with your article.  The whole thing reads as though you're searching not for an answer, but for the answer.  Well...there isn't one.

The second main problem is that you are attempting to do the impossible.  You want to "seamlessly" go from "karma" to "fortune" resolution; reading closely it sounds more like you want to shift from non-mechanical play to resolution-system play.  Switching between karma and fortune isn't really at issue with the problems you describe (the way I read them).  It'll probably fall under the common, 'karma unless someone cares,' type of mechanic.  Going from non-mechanical to resolution-system as presented is completely impossible.

Basically what you've sketched out is every way you can think of using rules for situations that don't use rules.  You can't have rules for when to switch from rule-less play to rule-full play because then rule-less play isn't rule-less.

Let's take your Situations.  Situation A is presented like any common pass/fail mechanic; the example is classic 'whiff factor' syndrome in action.  Let me ask you a question; what purpose is served making a die roll at this point?  It seems to suffer from a problem inherent in many games; you miss the die roll and you completely out of the game.  Do you think you game will be better if one die roll can 'kick someone out?'  That's a question only you can answer.

Fortune-in-the-Middle (FitM) die mechanics were created for just such a contingency.  With FitM, you do make the roll, but it doesn't kick-you/keep-you, it just makes things harder if you 'fail.'  (You don't really fail unless you want to, you just have to expend system-dependant resources to continue.)  'Whiff factor' syndrome goes farther; it includes situations like arm wrestling contests.  As you point out, some systems allow the wrestler with the overwhelming advantage to fail without giving much indication of why.  (I could see a whole other level of battle not covered by rules - perhaps psychological - that is 'hidden' simply as a chance of failure.)  Again, what is the point of using any kind of resolution system karmic, fortunate or otherwise?

Because "someone cares."  If no one else cares, skip the resolution systems; the match happens and it's over.  There are a number of different reasons someone could care about a resolution of action, the outcome could be important to them (vested interests), how it happens might be important, how much happens may be at issue, how long something took may matter, the kinds of opportunities offered could also be at odds, or a number of other things.  You'll note that amongst that list, only the first item is served by a simple pass/fail mechanic.

That brings us back to your Situations.  Situation B requires finely graded results between competitors to support the detail required in the example.  Situation C's example has, in my opinion, always underscored the zaniness of using exclusively a pass/fail mechanic.  (I mean, I've picked locks; it's not a matter of whether you succeed or not, but how long it takes as what ends you'll resort to.  I mean a favorite character concept of mine that's never fit any system I've tried, is a fantasy cat-burglar who uses a six-foot pry-bar as his main 'pick and tools' item.)  Situation D is confusing because it isn't really a matter of sensory deprivation, but the opportunity to act deliberately.  Situation E suits the first of what I listed; players care because of vested interests otherwise it's not worth resolving.

Finally, Situation F is probably where the real problem lay.  Because of the history of role-playing games evolving from war games and the tradition of duplicating success in the industry, most role-playing game resolution systems are based on how a game resolves a combat action.  Mike Holmes has made a very good (and very big) point about this in his Standard Rant #3.  For the reasons he lists, basing a whole game's resolution system on how it handles a simple combat action basically means that that's all it's good for, combat.

Your parameters are a really good set of 'dials' that you could build into a resolution scheme (as well as those implied by the Situations).  Let the players decide if they want a situation go random; let them 'take their chances,' otherwise go 'karma.'  How much is the finesse/power balance is tilted?  What about the quality measures of success?  (Time, craftsmanship, impact/side-effect, and opportunity, to name a few.)  What tools, materials, and time are needed/available?  If you want a broadly applicable resolution system, you'll need to address how the players can affect these 'dials.'

As far as a game system that "seamlessly can go from karma to fortune resolution," I think that was me crowing about Scattershot.  Except, I was talking about seamlessly going from General play (people talking, no mechanics involved) to Specific or Mechanical play (using resolutions systems, conversion of relevant information).  Karma resolution is frequently confused with General play because of the "he's way bigger so he wins" comparison.  Karmic resolution has little to do with a character 'going about their daily business;' most systems don't even talk about these kinds of events, they are simply assumed successful.  I've always felt this was a flaw in the many systems that give mechanics that are supposedly capable of resolving any action.  The 'whiff factor' looms large when a player mistakenly thinks they need to roll dice to successfully make a peanut butter and jelly sandwich (or when a gadgeteer decides to create a fork); some systems are very vague about when and how to apply their resolution systems.

One classic mistake I think you make is 'in the sofa.'  It always becomes a problem when a vital hint is 'hidden' behind a successful resolution invocation.  I'll only say this once, if it's vital, just give it to them!  Never, ever, make the mistake of thinking that vital elements of play should be left to chance.  Don't make them roll life-or-death to jump the chasm, don't make them roll for a vital clue, don't hold the game or a player's participation hostage to a single resolution event.  It just doesn't make any sense; why torture yourself, there are much better ways of creating suspense and intrigue (and none of them require this kind of potential sacrifice).

So...if you can create a resolution system with all the 'dials' I listed (or more), I think you'll finally have what you're looking for.  Keep up the good work!

Fang Langford

p. s. I just wanted to toss this in; compare the effect of giving yourself more time, first to picking a lock and second to shooting a target.  Wouldn't the second be aiming?  An interesting 'dial' there, I think.

[Christoffer Lernö]

The second main problem is that you are attempting to do the impossible. You want to "seamlessly" go from "karma" to "fortune" resolution; reading closely it sounds more like you want to shift from non-mechanical play to resolution-system play.

No, I mean karma to fortune. Because I figure you can go soliloquy -> karma -> fortune.

Basically what you've sketched out is every way you can think of using rules for situations that don't use rules. You can't have rules for when to switch from rule-less play to rule-full play because then rule-less play isn't rule-less.

I'm not quite sure what you're saying here.

Let's take your Situations. Situation A is presented like any common pass/fail mechanic; the example is classic 'whiff factor' syndrome in action. Let me ask you a question; what purpose is served making a die roll at this point? It seems to suffer from a problem inherent in many games; you miss the die roll and you completely out of the game.

Well, for Ygg, since that's the perspective I'm working for, you always have Fate points. Keeping in consideration that Fate retroactively converts a failure into success this works in two ways.

In one sense the "challenge" of the Chasm reduces the hero's chance of surviving later "challenges" by possibly requiring Fate.

The other thing is to clearly show the difference between a normal mortal and a fated hero. And a hero with no fate left is no more than mortal in the end. And it's there to let the NPCs use the same rules and plunge to their death if they're not important.

'Whiff factor' syndrome goes farther; it includes situations like arm wrestling contests. As you point out, some systems allow the wrestler with the overwhelming advantage to fail without giving much indication of why. (I could see a whole other level of battle not covered by rules - perhaps psychological - that is 'hidden' simply as a chance of failure.) Again, what is the point of using any kind of resolution system karmic, fortunate or otherwise?

In a draft of Ygg I had predictable skill rolls which introduced very little randomness. It was done to the degree that you knew exactly who knew and could do what.

The problem turned out to be those situations where I as a GM didn't want to decide. For example, who gets to see X first? What do they recognize about that and that? Small irrelevant details I didn't feel like deciding. Not the major stuff like "Do they make the make over the chasm" thing, but small things like perception rolls and the like.

I simply didn't feel like deciding every single detail, not the unimportant stuff. One character goes out hunting, does he find a deer to shoot? I don't know. Both could be ok as far as the story goes. Sure you could make this work with narrative mechanics but I'm not really looking for that type. The illusion that the GM knows-it-all is one I want to maintain.

So anyway, I'm looking for a system which easily supports that. It doesn't have to be much. Just so that you don't have to roll a die and improvise from there. It would be easy enough to formalize it and I think it would be a definate improvement.

As far as a game system that "seamlessly can go from karma to fortune resolution," I think that was me crowing about Scattershot.

Wasn't that soliloquy to mechanics? Karma is a sort of mechanics after all.

I think I was thinking a little like this I guess, when I really sort out my thoughts:

Case A: Unimportant stuff, daily business
GM and players use karma to soliloquy stuff in unambiguous cases. For ambiguous cases, like equal stats, the GM resolves through drama.

Case B: Important stuff
GM and players again use karma for unambigous cases, directly reading off effects from stats/skills/whatever. For ambiguous cases the dice assist in resolving proper effects.

Case C: Details
When the GM doesn't feel like deciding, it should be simple to use the Case B mechanic even though it's not really important.

Case A->Case B is solioquy->mechanical play.. but the karma stuff... well I guess what I mean is that I want to try to read off as much as possible from the values before deciding the dice is to be used, and even then the dice only slowly input randomness into the system.

The 'whiff factor' looms large when a player mistakenly thinks they need to roll dice to successfully make a peanut butter and jelly sandwich (or when a gadgeteer decides to create a fork); some systems are very vague about when and how to apply their resolution systems.

Actually I'd say that more than "some" systems are vague. I think most are. :-/

One classic mistake I think you make is 'in the sofa.' It always becomes a problem when a vital hint is 'hidden' behind a successful resolution invocation.

I wasn't really thinking about that. More of "who gets to find it (and keep it)?". It helps defining the characters because they can be known for their ability to be the first to find out something. Stuff like that.

[Le Joueur]

The second main problem is that you are attempting to do the impossible. You want to "seamlessly" go from "karma" to "fortune" resolution; reading closely it sounds more like you want to shift from non-mechanical play to resolution-system play.

No, I mean karma to fortune. Because I figure you can go soliloquy -> karma -> fortune.

That's good; have you spelled out the "soliloquy -> karma" part yet?  It may need to be very like the "karma -> fortune" part to make the game make sense.

Basically what you've sketched out is every way you can think of using rules for situations that don't use rules. You can't have rules for when to switch from rule-less play to rule-full play because then rule-less play isn't rule-less.

I'm not quite sure what you're saying here.

Basically, since the "soliloquy ->" part is essentially rule-less, you can't have rules regarding the transition between "soliloquy -> karma" because that would mean the "soliloquy ->" then had rules (at least about ending it).  It's a choice you can't avoid making (either "soliloquy ->" is with rules or without; you can't say that "soliloquy ->" is without rules and then have "soliloquy -> karma" transition rules)

Let's take your Situations. Situation A is presented like any common pass/fail mechanic; the example is classic 'whiff factor' syndrome in action. Let me ask you a question; what purpose is served making a die roll at this point? It seems to suffer from a problem inherent in many games; you miss the die roll and you completely out of the game.

Well, for Ygg, since that's the perspective I'm working for, you always have Fate points. Keeping in consideration that Fate retroactively converts a failure into success this works in two ways.

In one sense the "challenge" of the Chasm reduces the hero's chance of surviving later "challenges" by possibly requiring Fate.

The other thing is to clearly show the difference between a normal mortal and a fated hero. And a hero with no fate left is no more than mortal in the end. And it's there to let the NPCs use the same rules and plunge to their death if they're not important.

I don't think you're getting my point.  If a player is out of Fate points, one bad roll puts them out of the game.  Harsh realm.  Who wants to play a game that punishes player misfortune?  (Say I'm having a bad night with the dice, but I am really trying to fulfill my destiny so I'm really burning up the Fate points.  I run out, jump the chasm, and the Agility roll completely destroys the destiny I was headed towards.  That's neither similar to stories of 'Fated' characters or players' attempts to fulfill prophecies or destinies.)

I had the same problem working on Scattershot.  Then it hit me; I could reverse it.  It works something like this: a character has a Destiny (if that's how the player wants it), if something other than his own choices get in its way (like an untimely death) the system yields him more Out-of-Character (OoC) resources to get him back on track.  (I take it a step farther and use the player-chosen Destiny as the measure of role-playing when it comes time to decide upon rewards, which also happen to be the same OoC resource.)

How does that work in the example?  Well, the hero is jumping the chasm (actually, I wouldn't even allow this to be a roll; again, what is the point of endangering the players only input in the game with so arbitrary a hazard?), his Destiny is to die at the hands of his father, the lich-king.  The roll goes badly spelling out the character's death.  This is not appropriate to his Destiny, so Destiny kicks in a Fate point.  It's not the player's fault that his character wasn't going to fulfill his destiny so it shouldn't be the player who pays for it.

That gets back to the point I was trying to make.  The gamemaster says how far the chasm is.  The gamemaster compels the jump (like "you're being chased by an orcish army").  The player has no Fate points left; he rolls badly and the character dies.  I see no other interpretation than the gamemaster just killed his character (not the chasm, not the orcish army, but the gamemaster).  This sets up an antagonizing relationship between the player and gamemaster that I cannot think is healthy (especially if the gamemaster is the only source of Fate points).

All I can say is the chasm roll should not happen.

I cannot see an instance where the player should be kicked out of play by one die roll (even subjecting him to the chance of it) could help him continue to play.

'Whiff factor' syndrome goes farther; it includes situations like arm wrestling contests. As you point out, some systems allow the wrestler with the overwhelming advantage to fail without giving much indication of why. (I could see a whole other level of battle not covered by rules - perhaps psychological - that is 'hidden' simply as a chance of failure.) Again, what is the point of using any kind of resolution system karmic, fortunate or otherwise?

In a draft of Ygg I had predictable skill rolls which introduced very little randomness. It was done to the degree that you knew exactly who knew and could do what.

The problem turned out to be those situations where I as a GM didn't want to decide. For example, who gets to see X first? What do they recognize about that and that? Small irrelevant details I didn't feel like deciding. Not the major stuff like "Do they make the make over the chasm" thing, but small things like perception rolls and the like.

I simply didn't feel like deciding every single detail, not the unimportant stuff. One character goes out hunting, does he find a deer to shoot? I don't know. Both could be ok as far as the story goes. Sure you could make this work with narrative mechanics but I'm not really looking for that type. The illusion that the GM knows-it-all is one I want to maintain.

Can you explain how there would be a problem to simply skip such a roll and let the player determine the outcome?  That's what I do.  Like I said in the last post, only bring it down to the resolution system when someone cares.

If you don't want to decide, let the player.  If the player doesn't want to, have a resolution system that can do it for him.  In Scattershot we call that 'creating detail.'  So the outcome of the deer hunt is not important to you, but the player doesn't want an arbitrary outcome; they roll some dice and the dice decide.  The magnitude of success determine the size of the deer, something like that.

Is the "gamemaster-knows-all" harmed by not applying it to irrelevant details?

So anyway, I'm looking for a system which easily supports that. It doesn't have to be much. Just so that you don't have to roll a die and improvise from there. It would be easy enough to formalize it and I think it would be a definite improvement.

I'm just not getting how the 'roll-and-improvise' thing is harmful to your gaming.  Let the players do it; it makes gamemastering easier.  Can you explain?

As far as a game system that "seamlessly can go from karma to fortune resolution," I think that was me crowing about Scattershot.

Wasn't that soliloquy to mechanics? Karma is a sort of mechanics after all.

I think I was thinking a little like this I guess, when I really sort out my thoughts:

Case A: Unimportant stuff, daily business
GM and players use karma to soliloquy stuff in unambiguous cases. For ambiguous cases, like equal stats, the GM resolves through drama.

Case B: Important stuff
GM and players again use karma for unambiguous cases, directly reading off effects from stats/skills/whatever. For ambiguous cases the dice assist in resolving proper effects.

Case C: Details
When the GM doesn't feel like deciding, it should be simple to use the Case B mechanic even though it's not really important.

Case A->Case B is soliloquy->mechanical play...but the karma stuff...well, I guess what I mean is that I want to try to read off as much as possible from the values before deciding the dice is to be used, and even then the dice only slowly input randomness into the system.

Sounds like you got a very clear design specification for your resolution system, a really intriguing one at that.  The only pitfalls I would warn is you need to be really clear with yourself (and how you write the final product) about what "ambiguous cases" and Drama/Karma/Dice means.  From there, the design should be relatively easy to judge the degree of function of (as opposed to the actual design being produced by the design specifications, they are used to judge whether a design is successful).

On suggestion I would make is create source material for the "read off as much as possible" stuff, because you don't sound like you want lots of improvisation.

The 'whiff factor' looms large when a player mistakenly thinks they need to roll dice to successfully make a peanut butter and jelly sandwich (or when a gadgeteer decides to create a fork); some systems are very vague about when and how to apply their resolution systems.

Actually I'd say that more than "some" systems are vague. I think most are. :-/

And that's bad right?

One classic mistake I think you make is 'in the sofa.' It always becomes a problem when a vital hint is 'hidden' behind a successful resolution invocation.

I wasn't really thinking about that. More of "who gets to find it (and keep it)?" It helps defining the characters because they can be known for their ability to be the first to find out something. Stuff like that.

Then shouldn't it be a conflicted roll rather than a sensory roll?  You're not deciding if it will be found, but who finds it.  Similarly, if the chasm decides where the player goes as opposed to whether he stops playing or not, it isn't a problem.  As a gamemaster, you're still giving the players a free victory, just letting the dice decide how it goes.  That's a lot different than what it sounded like you were describing.

This would be probably the most vital part of the instructions to using your resolution system.  Without this tidbit, your game would be every bit as vague as the peanut butter and jelly example.  With it, I think this could be one of the best games ever!

Fang Langford

[Ron Edwards]

Hi there,

Fang is dead on correct (like he needs me to confirm this). The classic "chasm" question is a non-issue at the general level of game design, much like the classic "airlock blows out" question.

It all comes down to satisfaction levels of the game, and what brings them about. If you want to appeal to people whose priority is in-game physics, and such role-players are indeed out there, then handle it strictly in that way. Plenty of game systems do this to better or worse extent; I recommend BRP (RuneQuest or Call of Cthulhu), GURPS, JAGS, and Multiverser for the most thought-provoking and/or consistent. And I'm waiting for my copy of Pocket Universe to arrive too, which is probably going to be added to that list.

If, on the other hand, you want to appeal to people whose priority is more Narrativist, then all manner of system design here at the Forge is more appropriate. I suggest comparing Otherkind, InSpectres, and Dust Devils, all of which handle narration and "competence" a bit differently from one another. Commercial games of this sort, again a bit different from one another, include Sorcerer, Orkworld, and Hero Wars.

Going back to your previous Ygg threads, though, I think your priority (and correct me if necessary) is Illusionist Sim - that is, you want the GM to be able to override pretty much anything that the dice do, without it looking too obvious. Is this correct? In that case, you are going to have to heed Fang's advice very carefully, because the "inappropriateness" of the chasm roll does apply (as it would in a BRP game), but it cannot be simply acknowledged as inappropriate and passed over (as it would in, say, Hero Wars, if the chasm posed no particular conflict in story terms).

It's tough. The players are intended to be "in" a story - "Oh, shit! I'm really risking it to jump across this chasm!" - but the role-playing process cannot actually devalue the story-priority by having them fall and die. This is the Impossible Thing, unless it's resolved by effective Illusionist design. I think most attempts at this have failed, historically, relying on ultimately-dysfunctional Balances of Power among GMs and players. I would be very interested in seeing a game design that does accomplish this goal.

Now that I think of it, Multiverser might pose some of the best current answers in this regard, due to its special use of character death as a means to continue play.

Best,
Ron

[Christoffer Lernö]

Ok, I'm gonna check the other stuff in detail later. For now just something about the chasm.

Actually, I don't find myself doing the chasm thing myself. But in the cases I heard it properly used I figured it goes something like this:

a) there are n number of ways to get to point A in the story (with "way" I mean a sequence of decisions by the players)

b) one of these ways end up with the chasm. If the players want to take this way out, then sure this is possible. It's just that they can fall down. It's either that or doing something else. There is never ONLY the chasm.

So in one case the players decide to run from the orkish horde. Uphill leads to the chasm, and if they make it over they can be sure they won't be followed, or they can take their chances to try to pass the orkish sentries and go downhill. Here it's a matter of where the risk is taken.

But this reminds me of another golden rule used by GMs which go for this method: there is only a potentially leathal outcome if the players come up with it themselves. Or in other words, if the players know they are heading for something potentially deadly to their characters, they get that. "Let's run and jump over that chasm to survive" is not the way out the GM had in mind, but it's allowed although it's consequences must be taken in full.

Or even simpler put: you're only at risk if you deliberately put yourself at risk. A good Illusionist GM rule I think. Maybe it should even be written into the rules. The cool thing about it is that it never disturbs the "fairness" of the game. It reinforces it.

[Le Joueur]

a) There are n number of ways to get to point A in the story (with "way" I mean a sequence of decisions by the players)

b) One of these ways end up with the chasm. If the players want to take this way out, then sure this is possible. It's just that they can fall down. It's either that or doing something else. There is never ONLY the chasm.

So in one case the players decide to run from the orkish horde. Uphill leads to the chasm, and if they make it over they can be sure they won't be followed, or they can take their chances to try to pass the orkish sentries and go downhill. Here it's a matter of where the risk is taken.

I'm still going to have to list this as dysfunctional play.  It is one of the hardest gamemastering techniques to do is the "number of ways to get to point A" style; play always gets to "point A."  That, in itself, is not the dysfunctional part.  Any use of the chasm is.  Of all the "n number of ways," the chasm presents a binary outcome dependant entirely on one die roll; either the game continues or it stops.  Even taking the risk of ending the game is a bad idea in my head.

Honestly, I have only seen this used badly.  The chasm appears because the gamemaster doesn't want the players to 'go that way.'  The players, not knowing they have other choices, go for it.  Since the chasm was meant to be the 'immovable object,' the more the gamemaster doesn't want this course of action, the harder it becomes.  It's a no-win situation; if the player fail, they can't play anymore, if the players succeed, they go where the gamemaster doesn't want them to go.

If you want to keep the players from doing something, good Illusionism requires that they don't even have the opportunity to think they have the option.  Using a Hobson's Choice lethal barrier is a no-win situation.

If the players are confronted with Orkish sentries, there are many ways to 'turn them back' without endangering the play of the game; with a chasm there is only death.  Almost every time I see this kind of gimmick applied, no one knew there was a chasm until they got to the edge of the gamemaster's map.  Since the gamemaster will not tell them that he has other plans, a chasm appears (with the gamemaster hoping that his description will dissuade the players).

Certainly there will be exceptions to the above case, but they only apply if you broaden the example with information that was not given.  Take for example the sentries; did the players know about the sentries and the chasm before they were being chased?  In that case, the chasm will have already identified as a potential choice, in which case it probably isn't a spur-of-the-moment jump, but instead an impediment.  That being the case, I again question why there is a die roll; you certainly don't want to end the game prematurely, a bad chasm encounter should cost the party, but not their lives.

Simply, if the chasm is supposed to be a disguised wall, there should be no die roll because it simply shouldn't be attempted (perhaps a better disguise could be found for the impasse); if the chasm is meant to be an impediment, then there should be no die roll, because death is not conducive to gaming (any number of other 'costs' can be assessed for 'taking the chasm' route).  No matter how I slice it, a chasm should either not appear or not be a randomized life-or-death situation.

But this reminds me of another golden rule used by GMs which go for this method: there is only a potentially lethal outcome if the players come up with it themselves. Or in other words, if the players know they are heading for something potentially deadly to their characters, they get that. "Let's run and jump over that chasm to survive" is not the way out the GM had in mind, but it's allowed although it's consequences must be taken in full.

Or even simpler put: you're only at risk if you deliberately put yourself at risk. A good Illusionist GM rule I think. Maybe it should even be written into the rules. The cool thing about it is that it never disturbs the "fairness" of the game. It reinforces it.

That golden rule would work if the player either A) created the chasm themselves (making it their choice) or B) simple failure does not result in instant death.  I think it is bad Illusionist gamemastering to provide opportunities to abruptly end the game.  If I am not mistaken the outcome of an Illusionist game is the purpose of the Illusionist gamemaster; if the player characters die, it is a result of the gamemaster's choice.

That means the gamemaster killed the characters.  Is that good gamemastering?

I think the rule should be not to offer a Hobson's Choice where death is an option unless the death serves the goals of the gamemaster's plan.  Let me put it another way; you're an Illusionist gamemaster, you want the players to get to a certain spot.  You put covert impasses into play to cause them to move the desired direction.  An impasse that offers a chance for the players to kill their characters means that they don't get to the spot.  Therefore a potentially lethal impasse is poor Illusionist gamemastering.

I know one of the common hallmarks of Illusionist gamemastering is Hobson's Choice; 'you can either go my way or you don't go anywhere.'  The problem comes when you disguise this using potentially lethal force.  Better a simple dead-end that can be underplayed rather than letting the players mistakenly believe that the lethal consequence is actually an option (being an illusion, right?).

I'm sorry, I going around on this one again and again, but I have always detested the chasm (or the air-lock) with every fiber of my being.  I am also sorry that I have taken this thread so far off-topic, but after the night I just had, I felt a need to vent.

Please return to your originally scheduled thread.  This is no longer a discussion; we're either wildly misunderstanding each other or have reached our own impasse.  (If you feel a need to further clarify the 'chasm situation' to eliminate my point, then it clearly isn't what I am talking about - which I feel is a valid concern about poor gamemastering - and therefore does not require further input from me.)

Fang Langford

[Ron Edwards]

Hey,

I think the difficulty here is that both Fang and I have a personal beef with Illusionist play, and so it's hard to distance oneself from that in discussing how to make it work.

A: What's a good way to do X?

B (struggles with X as a goal): Um, I think this would work. But wait! Is X good in the first place?
A: Well, I want to do X. I'm thinking of trying these methods. (Explains)

B (really squints): Umm, OK, that's kind of problematic. But is it problematic because it might not work for X, or because I hate X with a passion? Damn!

On the plus side, we tried really hard and I think Christoffer's weathered both of us pretty well. This is a good thread for raising all the issues necessary to consider the problem.

Christoffer, I'll be curious to see what you come up with after thinking it over.

Best,
Ron

[ks13]

I'm still going to have to list this as dysfunctional play. It is one of the hardest gamemastering techniques to do is the "number of ways to get to point A" style; play always gets to "point A." That, in itself, is not the dysfunctional part. Any use of the chasm is.

The chasm is only one example of any situation where a die roll can lead to character death. Christoffer's view is that as long as this choice is made by the player, he should deal with the consequences. According to Fang, this should not be part of the game, which to me implies the following: character death as a player choice is OK, but the player's choice to let the dice decide the characters fate is not.

The players, not knowing they have other choices, go for it.

The players always have a choice. How about just standing at the foot of the chasm and waiting to see what happens? Or looking for other options. Now I will concede that a GM might force the players to choose between "must roll dice to see if they survive" situations. In this case it is no longer a player choice. But this a problem with the GM, not the chasm. Fang's argument seems to be that in most situations it will come to this. However, the solution of eliminating all chasms, and cliffs, and enemy weapons or anything else that could kill the character with the proper die roll leaves us with what? How do the alternatives play out?

the chasm presents a binary outcome dependant entirely on one die roll; either the game continues or it stops.

I really disagree here. Who says the jump over a chasm has to be a binary outcome? If you take such a narrow view than of course the results can be very unsatisfactory.

The chasm appears because the gamemaster doesn't want the players to 'go that way.'

Than it should be clear that there is no way the characters can jump across. Again, I am putting the responsibility on the GM's shoulders, not the situation. If the players insisting on jumping a 100 foot a chasm, and the GM is not presenting alternative options, then the dysfunction is within the group.

That golden rule would work if the player either A) created the chasm themselves (making it their choice) or B) simple failure does not result in instant death. I think it is bad Illusionist gamemastering to provide opportunities to abruptly end the game. If I am not mistaken the outcome of an Illusionist game is the purpose of the Illusionist gamemaster; if the player characters die, it is a result of the gamemaster's choice.

So if B) is present, what's wrong with the chasm?

An impasse that offers a chance for the players to kill their characters means that they don't get to the spot. Therefore a potentially lethal impasse is poor Illusionist gamemastering.

Yes, this is true, but only if the system presents a binary outcome as the only possible result. Fang's point is valid. If you want the characters to go to point A, why give them the opportunity to die before they get there? But I think the trick is providing an "illusion" that they might die before they get there. I suppose it depends on what the players want; a journey to "A" frought with danger, or jumping right into action at "A".

[Ian Charvill]

The chasm is only one example of any situation where a die roll can lead to character death. Christoffer's view is that as long as this choice is made by the player, he should deal with the consequences. According to Fang, this should not be part of the game, which to me implies the following: character death as a player choice is OK, but the player's choice to let the dice decide the characters fate is not.

The factor which would make me balk at a situation as a GM, or bitch about the situation as a player, is that a single die roll decides the entire course of the adventure (or a single decision, depending on how you look at it).  It places a lot of emphasis on a decision that should not really carry that emphasis.  Designing an inconsequetial, steps along the way, scene that can lead to a player death is designing a scene which could either make a player unhappy or make them... well nothing really because there's no big reward in a scene like that.

The risks heavily outweigh the benefits.

(This may be a problem inherant in a success/failure mechanic rather than a success/complication mechanic.)

Ian

[Le Joueur]

The players, not knowing they have other choices, go for it.

The players always have a choice. How about just standing at the foot of the chasm and waiting to see what happens? Or looking for other options. Now I will concede that a GM might force the players to choose between "must roll dice to see if they survive" situations. In this case it is no longer a player choice. But this a problem with the GM, not the chasm. Fang's argument seems to be that in most situations it will come to this. However, the solution of eliminating all chasms, and cliffs, and enemy weapons or anything else that could kill the character with the proper die roll leaves us with what? How do the alternatives play out?

As Ian points out, I don't think an 'all-or-nothing,' single die isn't "the proper die roll."  The solution isn't the elimination of "all chasms, and cliffs, and enemy weapons..." it is making them obey the propriety of die rolls.  Ian makes a good point about the danger/benefit balance here.  Chasm (as listed): one roll, little benefit, heavy risk = improper roll.  Enemy weapons: lots of rolls and mitigating factors, big benefit, moderate risk = proper roll.

the chasm presents a binary outcome dependant entirely on one die roll; either the game continues or it stops.

I really disagree here. Who says the jump over a chasm has to be a binary outcome? If you take such a narrow view than of course the results can be very unsatisfactory.

And this is easily misunderstood in the heat of play.  The risk of the outcome of a misunderstanding and a bad roll is far too great.  What's so wrong with 'softening' it a little?  It's not a chasm, but a canyon?  (Many rolls, mitigating circumstances = proper roll.)  Or it's a picket line (like sentries or enemy weapons = proper roll).

The problem is when its just a bottomless chasm you either jump or die (avoid was not given as an option in the original).

The chasm appears because the gamemaster doesn't want the players to 'go that way.'

Than it should be clear that there is no way the characters can jump across. Again, I am putting the responsibility on the GM's shoulders, not the situation. If the players insisting on jumping a 100 foot a chasm, and the GM is not presenting alternative options, then the dysfunction is within the group.

That's exactly what I wanted in the 'how to use this resolution system' instructions.  Either you make it clear it can't be jumped or at least keep it from being a game breaker.  It's a gamemaster thing!  You can't say that it's "not the situation," but the gamemaster, because this is a situation entirely created by the gamemaster.  All situations in this game are the gamemaster.

That golden rule would work if the player either A) created the chasm themselves (making it their choice) or B) simple failure does not result in instant death. I think it is bad Illusionist gamemastering to provide opportunities to abruptly end the game. If I am not mistaken the outcome of an Illusionist game is the purpose of the Illusionist gamemaster; if the player characters die, it is a result of the gamemaster's choice.

So if B) is present, what's wrong with the chasm?

Nothing whatsoever.  That's why I suggested adding B) to every situation.

An impasse that offers a chance for the players to kill their characters means that they don't get to the spot. Therefore a potentially lethal impasse is poor Illusionist gamemastering.

Yes, this is true, but only if the system presents a binary outcome as the only possible result. Fang's point is valid. If you want the characters to go to point A, why give them the opportunity to die before they get there? But I think the trick is providing an "illusion" that they might die before they get there. I suppose it depends on what the players want; a journey to "A" fraught with danger, or jumping right into action at "A".

I think this is one of the most important skills for any gamer, gamemaster or otherwise.  I have always felt the major use of dice is creating suspense.  You don't know how things are going to turn out until they roll = suspense.  The real trick here is to not be fooled into thinking that by creating a situation where character death is a blatant, in-your-face possibility, you are creating good suspense.  Escalating the penalty does not equally escalate the suspense.  I think a scene of creeping past the sentries making many hidden stealth rolls tends to create a lot more suspense than one quick do-or-die chasm jump.  And the penalty is not nearly as set or bad.

Fang Langford

[M. J. Young]

Someone pointed me to this thread because Multiverser was mentioned favorably in it. It is gratifying to me that so insightful, intelligent, and informed game designer and critic as Ron Edwards can say positive things about the game, and I feel like I should thank him for his kind words if I can do so without insulting him with the implication that they were not sincerely written.

But I think I can also add something to this resolution question.

There is nearly always an unnoticed drama component to event resolution in every game, and that is that someone has to decide when the situation requires the intervention of mechanics of some sort. This is itself a mechanic, but is not treated as such. A good example lies in the fact that Multiverser includes a skill for walking. Obviously no one checks for walking every time a character attempts to cross a room; yet the skill has value in the context of the game. If a character is inebriated, can he still walk? If a character is on a ship, but unaccustomed to the roll of the deck, will he keep his feet? In such situations, it might be that the one with more experience "walking" in contrary conditions would have a better chance of staying on his feet. The drama component comes in because someone has to decide whether this particular situation warrants a roll (or whatever resolution method is to be used). Given the near infinite possibilities of the game world (not just with Multiverser, although it is certainly multiplied there) no set of rules could cover all situations; someone has to use their judgment to decide whether the circumstances rise to a level at which mechanics are implicated.

But there is also this mistake made about fortune mechanics that suggest they must be black and white, pass/fail, when they can be just as flexible as drama mechanics if desired. Let's take the infamous chasm example. For what it's worth, in twenty-two years of gaming I have never either seen nor run a scenario in which someone tried to jump a chasm; but it's an interesting brightline test.

So we have a skill check; we're going to roll a die and decide whether the character has "succeeded" in making his jump.  But how do we determine this, and what does it mean?

In Multiverser, your skill would not be "can jump". That's too obvious.  Rather, it would be something considerably more specific. For my brother, who competed in the physical fitness team in high school, it would probably read Standing Broad Jump 9'6"; for me, it would probably say that I can do a standing broad jump of about 4'. O.K., I was never a fitness nut. The question is, can I jump this chasm? Let us suppose the chasm is 6' across (and that there's no running room, because I don't happen to have any idea how far either of us can jump on a run). For him, there hardly seems reason to check--someone who can on a good day jump 9'6" can jump 6' without hardly thinking about it. For me, maybe there's no reason to check, either--my chance of right now jumping two feet farther than I have ever jumped before seems so small that if I've any brains at all I'll turn around and find something else to do. But let's suppose that my son's skill is defined as 6'; that means for him it's a does he or doesn't he situation. Now, how do we handle the rolls?

One way we can handle it is to modify the chance of success.  For my brother, he's going to get a substantial bonus. Maybe he's got a one or two percent chance to fail in this; maybe he's got so much bonus, he's got a guaranteed success (but maybe the referee doesn't tell him, lets him roll anyway to keep the suspense). For me, penalize the roll--I'm trying to do something I've never done, jumping 50% farther than my best recorded jump. Maybe the penalty will make it impossible; maybe I've got a slim chance. For my son, he's on even odds; his chance of success is exactly for this situation. He might make it, he might not.

Another way we could handle it is by way of relative success and relative failure. This actually provides two different ways of handling it, one subjective and one objective. I'll treat the objective one first.

Relative success and relative failure maintain that your success roll is not merely whether you succeeded, but how well did you succeed. Using a percentile system where low succeeds, if I roll a 01 I made my four foot distance, as stated on my sheet; the same roll will enable my son to jump 6' and my brother to jump 9'6". But everyone thinks that when the pressure is on they can run a bit faster and jump a bit higher. If I roll better than 01, I jumped a bit further than 4'; and maybe if I rolled 50 (and that was a successful roll for me) I jumped 50% farther than 4', which would be 6', and I made it.  Maybe conversely my brother rolls too high. But technically if he fails his roll, it only means that he did not jump 9'6"; he does not have to jump 9'6" here--he only has to jump 6'. That's only 63% of his jumping distance.  If he failed with a roll of 75 against a chance of 60, that could easily be taken to mean that he jumped 15% less than the reported distance--which is still 8', so even in failure he made the distance necessary. Now, let's suppose my son needs 60 or less, and he rolls 61. Does that 1% failure mean he is plunging to his death? No, it means he jumped 5'11"--and if he can't grab onto something to try to save himself at this point, there's something wrong somewhere.

Subjectively, you can discard the math and thumbnail relative success and relative failure. A close failure would mean slipping down the far wall looking for handholds; a far failure would mean plunging down the middle needing something far more drastic to save you; a botch would suggest you tripped on take-off.  Similarly, a very low success roll says that you barely made it, while a stronger success is an easy leap.

These mechanics are all options in the Multiverser system for handling skill outcomes.

All of this fails to answer the core objection: why is the chasm there in the first place? To reply that it's a bogeyman, a strawman that no one ever really does, is to miss the point at one level. There is the monster that's too tough to tackle. There are always things characters can try which are incredibly difficult and have severe consequences. The chasm is just a convenient way of simplifying the problem. I agree that such situations lead to the possibility that the character could be killed, in most games ending play for that player (not, obviously, in Multiverser, but that's a special case, and it doesn't completely eliminate the sting of death).

But I think that the kind of play that eliminates these is geared to a particular kind of player; if you don't have that kind of player, that kind of play doesn't work.

I have a world in the center of which is a city. The city is uninhabited, but patrolled by what are known as the Giant Robot Spiders. These are killing machines. They are there to be trouble. The intelligent player will observe their conduct, and realize that they are entirely dormant at night to the point that they cannot be activated. It is only during daylight that they are a problem. Because the city is something people want to explore (and it contains its own kind of valuables of a potentially practical nature), players sometimes take risks, knowing that if they fail to leave the city by sunrise they will probably be killed. But I have one player who walked all the way to the city, saw the spiders, and decided to bum rush one with a flying kick. He was dead before he reached it. The spiders create a challenge, an intellectual problem for those who decide to tackle them as a problem to be solved; but the character who merely tries to take one on in hand-to-hand combat cannot be saved. Or can he?

The problem is that character invulnerability only works for players who will stay sane when they have it. It would have been entirely possible for me to resolve the situation in a way which allowed him to defeat the spider. But for that kind of player, success here leads to greater "risks" because there is no risk. I was in a game in which there was PC invulnerability, but the referee had not actually said so. Among some of the players there was a gradual escalation of nonsense, as they seemed to be testing the limits of how stupid and brazen their characters could be before they would suffer any consequences. They seemed to delight in the fact that they could spit in the eyes of the gods and walk away unscathed, as it were. These were not bad players; they had played well in my games, had taken intelligently considered risks and succeeded admirably. But without the constraint of the necessity to protect their characters, they threw caution to the wind.

The 6' chasm is no different from the sleeping dragon. It offers the possibility that you might throw your character to his death by your stupidity, and the alternative that you might think about possibilities and find something that will work. If you have to jump this chasm, can you at least tie a rope around your waist and get someone to secure the other end, in case you don't make it? Taking precautions, using wisdom in your decision making, is a primary technique in keeping your character alive. Some players will play a character as if he were at risk even when he is not; others will play as if the character is not at risk when he is. Most people seem to fall in the middle, assessing the actual risk to the character in making their plans, and trying to find ways to minimize this. For them, the risk must be real; and that means it must be possible to choose to leap the chasm, and to fail, and to suffer the consequences of failure.

By the way, I love the Fate Points idea; but I think it comes to the same thing eventually. If the character had a Fate Point left when he reached the chasm, he could save himself by spending it if the die roll fails. If the character is out of Fate Points, he should think twice about leaping the chasm, because he might fail. But if he has run out of Fate Points before now, it suggests that the mission was more difficult than he could manage (such points are, after all, one of his resources to be considered in determining how strong he actually is), or that he took unnecessary risks along the way (he may have been forced to spend such points because he attempted to do something dangerous where something safer was readily available). The point of limiting Fate Points would seem to be the same as the point of having the rolls in the first place: to prevent the character from being invincible, to steer the player away from those insane high-risk choices from a desire to conserve this resource. Unlimited Fate Points would mean that the character was invulnerable; invulnerability is bad if and only if it means the player will abuse it by regularly doing the impossible; if the player will not abuse it, Fate Points are unnecessary. The limitation on the fate points is precisely so that the player who takes too many risks will eventually deplete them, and so face the chasm on the basis of his die roll alone. I understand the problems of bad dice luck; I have it constantly. The lesson I learn from it is to be conservative in my plans and efforts, and to be certain there is a safety net whenever I do anything dangerous. If I were given fate points, you can bet I'd still have some when I got to that chasm, if only because I would never put myself in a situation in which a single roll would be fatal where I could find a safer way to do something.

With all the interruptions I've had around here, I hope this is coherent and does add something to the discussion. I'll look forward to reading further comments.

--M. J. Young

[Christoffer Lernö]

I think the chasm discussion here is very rewarding although it might be a little off the original discussion. In fact it's so interesting I didn't post a skill system example to this thread but made it seperate.

In that example, which might be flawed as hell mind you - I only just wrote it up, I tried to do something which deals with randomness and lets people in general know their chances pretty well as opposed to those games where it doesn't really matter whether it's easy or hard, you have a solid chance of failure anyway.

But back to the chasm example.

I, maybe not surprisingly, agree with M.J. Young on this one.

Let me take an alternative "chasm" from an actual game I played rather than GMed.

In a Southern Cross (Robotech) campaign me and another guy was playing two mecha pilots, him being the senior officer. We had ran into some hostiles which definately wasn't supposed to be there. After kicking their asses, not without some difficulties, we tried to radio back but it didn't work. We were told to go to point B, but now we were picking up about three times the force which we had only beaten with difficulty. We had no back-up, no-one knew were we where and it was important that people knew there were enemies here (or that was my impression here).

The GM doesn't really go into that of course, we players both know that very well. I'm saying: "Let's get the hell out of here! Retreat now!"

My senior officer: "No, we're standing here and fight it!". Obviously he was under the delusion that there was some way of winning.

"No let's go!" I insisted.

"Stay here, that's an order" was the reply.

"Screw you, I'm leaving!"

The last thing he said was "I'm gonna report this when I get back!"

Well, famous last words. Naturally he got totally blown to bits and my character was decorated for bravery and stuff for providing vital information about the enemy offensive. Of course I never did tell them I left before my senior officer died, but I think they wouldn't have minded too much. The other player was pissed at me(!) for not staying and fighting, blaming his character's death on me.

The GM however had the same idea as me. We weren't supposed to fight these boogies, we were supposed to run. If it weren't for me running away I guess that other player would have blamed the GM for killing him. And yet it was very much he who made the decision to die there. The odds were obvious.

Actually, we could have had a chance.. maybe one in 10 or so. On a good day anyway. However, it was obviously not a risk we were supposed to take, so why take it. The same with the chasm. Does the GM put the chasm there for players to JUMP or to tell them "heck, you can take this way if you want (that's why he makes it jumpable), but it's risking your life. Try to think about some other way, ok?"

It's more subtle than just making it end with a cliff and no way out. It's more player freedom this way.

On the other hand, there are GMs who rape these ideas totally. Like a game where we were attacked by extremely dangerous monsters. So I try to run away and for that I get totally obliterated by the GM. However, the players who simply stay and fight miraculously see the monsters fight like they had no idea of what they were doing and end up not only killing these virtually unkillable creatures, but also helping me survive. In this case it was more of a "JUMP THAT FRIGGING CHASM DAMMIT OR I'M GONNA SPANK YOU", because the chasm is really the highway, it's just that it looks like the chasm, and the GM says it's a chasm, but when you roll anything is ok.

So anyway, I guess this stuff should be pointed out, and actually even better, made into some meta rule. How that metarule should look I'm not really sure of though.

About the chasm, I get reminded about AHQ. There's a chasm in AHQ. There is also no set dungeon in AHQ. If you come to a chasm, you can choose to jump over with a heroic leap. Fall down and you have to spend a fate point or die. There is also a "sensible leap" you can attempt if you have a rope and friends to hold on to it. Or you can simply decide it's not worth crossing. After all, in AHQ, there is not necessarily anything interesting on the other side, you can just go and open another door and find another room and the potential rewards are the same as everything is random.

So even in the board game AHQ the chasm is only a risk if you choose it to be.

But maybe I'm missing the point again Fang?

[Le Joueur]

But maybe I'm missing the point again Fang?

I don't think so....

I, maybe not surprisingly, agree with M.J. Young on this one.

So do I, especially with this part:

...a challenge, an intellectual problem for those who decide to tackle them as a problem to be solved; but the character who merely tries to take [a cybernetic death-spider] on in hand-to-hand combat cannot be saved. Or can he?

The problem is that character invulnerability only works for players who will stay sane when they have it....

...because there is no risk. I was in a game in which there was PC invulnerability, but the referee had not actually said so. Among some of the players there was a gradual escalation of nonsense,

I think this is at the heart of the 'handle it with probability' discussion.  And M. J. is right about the idea that different kinds of players are suited to different kinds of games.

That means this idea is of only secondary interest to the 'chasm problem.'  We aren't discussing the player-to-rules relationship.  As far as I can tell we're talking about covert impasses in Illusionism.

Let me take an alternative "chasm" from an actual game I played rather than GMed.

In a Southern Cross (Robotech) campaign me and...we were picking up about three times the force which we had only beaten with difficulty. We had no back-up, no-one knew were we where....

[The other player said,] "...No, we're standing here and fight it!". Obviously he was under the delusion that there was some way of winning.

Well, famous last words. Naturally he got totally blown to bits...

...However, it was obviously not a risk we were supposed to take...

No, the flaw I have been pointing out is that in choosing a covert impasse, it was certainly not obvious.  The example spells that out in stark detail.

What I'm saying is 'avoid covert impasses like the plague.'  If they're not covert, then it's blatant (and I understand this style avoids blatancy).  The whole point of them being covert is that the players are meant to be fooled.  You can always succeed too well; that results in player death.

What do you do?  Use a covert impasse where the deterrent isn't death.  Simple.

The same with the chasm. Does the GM put the chasm there for players to JUMP or to tell them "heck, you can take this way if you want (that's why he makes it jumpable), but it's risking your life. Try to think about some other way, ok?"

So anyway, I guess this stuff should be pointed out, and actually even better, made into some meta-rule. How that meta-rule should look I'm not really sure of though.

So, yes you have the point exactly.  And perhaps the meta-rule is 'do not use death as a covert impasse.'  Simple instant death is okay for an overt impasse, but because of the unpredictibility of covert impasses, use something else.

Fang Langford

[Christoffer Lernö]

Ok, we pretty much weeded out the chasm problem. Can we get back to the issue about skills now?

I was thinking about the problem some more. Actually, because the skill resolution has such wide applications, I'm wondering if there aren't better ways to deal with this problem.

There are narrative ways of dealing with it and stuff, but the problem comes in the interfacing part.

As my game stands for example, there is a solid, balanced combat mechanic. Any special abilities of impact on combat should interface with the combat system in a continuous manner obviously, as should the magic system.

What this means is that we can only gradually alter the manner of which skills are resolved or we might run into contradictions.

What are the alternative ways of resolving skills and are any of them compatible with a rather rigidly defined combat system? Look at it from a gamist view. How do we balance a more narrative/directorial style skill system with other parts of the game which are strictly mechanical? Has it been done successfully? If so, where?