First, I am sorry if this has already been beaten to death, I haven't spent too much time trolling through the posts.

I have read most of Ron's essays and while I think the notion of resolving conflict rather than actions is a good idea, I don't understand how it works in practice.

It seems as though some parts of a game will always need more in-depth resolution. That is, if I get to the end of a session and the Big Bad is waiting to kick my butt, I (pretty much regardless of GNS) would be let down if that dramatic conflict were reduced to the fickle fate of a single roll.

I'm sure there is something key here that just isn't making it through my thick skull. Can anyone clear this up for me? Any clever examples of play that illustrate how to make this work?

Thanks,

Hi Jay,

The trouble is that there are two variables to consider, not just one.

#1 is conflict vs. task resolution. The best way to think about this is ... let's see, one guy is chasing another guy. The chase-ee comes to a fence and wants to vault it as he runs.

What do the mechanics address? In many games, the issue is the jump per se - can or cannot this guy jump this fence, at this speed? Quick reference to various physical parameters, jumping skills, et cetera, to set up the probabilities. The chase itself is put on hold for a minute while we focus on, think about, and carry out the jump, in terms of the character's capabilities. That's task resolution.

However, in conflict resolution, the chase is the thing. If we're gonna roll (for instance), it's going to resolve the chase, or some aspect of the chase per se. The fence and whatever relevant physical parameters for the character modify the chase, in favor of the pursuer, and are not considered as an isolated task. It might be a dice penalty to the chase-ee, or whatever, but when we swing the system into action, whatever rolls or energy-points or whatever are involved, we'll still be focused on the chase's outcome.

#2 is about scale. It's easy to get confused, because in the above example (and many similar examples), the jump exists inside the chase, so people get mixed up and think that conflicts must always be bigger than tasks. That's not the case at all.

Scale refers to how many or how much of the game-world events get covered by this particular resolution mechanic in action. The usual range is "scene resolution," for the biggest/most, to "single-action" for the smallest/least.

As I say, this is a completely independent variable from #1.

Examples:
Tunnels & Trolls combat: task resolution at the scene scale.
HeroQuest simple contest: conflict resolution at the scene scale
Cyberpunk: task resolution at the single-action scale
Sorcerer rolls in (say) combat: conflict resolution at the single-action scale

Does that help?

Best,
Ron

Well, one of the things to keep in mind is that what you would be missing is not the additional rolls. What you'd be missing is all of the tactical fiddling that goes along with those rolls.

Conflict resolution mechanics are not simply task resolution mechanics moved to a larger scale (where you correctly note you'd lose the desireable level of resolution). Conflict resolution mechanics must be structure so that all of the tactical fiddling is shifted to the roll itself.

Good examples of this would be Story Engine, Trollbabe, and Universalis where a whole bunch of ==stuff== goes on before the dice even hit the table. Its at the level of this "stuff" where all of things you wouldn't want to miss out on occur.

Ron--
Thanks for the response. I understand both of your points, but I am still not entirely satisfied.

I just don't buy "conflict resolution at the single-action scale". I reread the combat section of Sorcerer and it sure seems to be action resolution to me.

Can you give an example of how an action in Sorcerer combat is resolved as a conflict?

Thanks,

coxcomb wrote: It seems as though some parts of a game will always need more in-depth resolution. That is, if I get to the end of a session and the Big Bad is waiting to kick my butt, I (pretty much regardless of GNS) would be let down if that dramatic conflict were reduced to the fickle fate of a single roll.

A distinction from film might help here (assuming I understand what's going on). I'm going to use an example I like from The Princess Bride because most people seem to have seen it and it's such a clearly laid-out film. It also rocks.

Remember when the Dread Pirate Roberts (actually the hero in disguise) has to climb up the cliff, then at the top fight Inigo Montoya, then run off and fight Andre the Giant (I forget the character name)? Okay, working from large to small:

Sequence: This could be divided two ways. First, you could take the whole thing from cliff to running off as one sequence. Second, you could take the cliff part as one sequence and the fight with Inigo as a second. I would tend to go with the former, since much of the cliff involves a running conversation with Inigo, who is really hoping that the hero gets to the top so they can have a swordfight. Much backchat and banter happens here, all of which sets up the swordfight to come.

The next thing is the swordfight itself, which has lots of good banter, some crunch moments ("I am not left-handed," "I am not left-handed either," "I would rather cut my arm off than kill an artist like you," etc.), and a lot of great action.

An essential point is that if you take it all as one sequence, the cliff-climbing establishes a whole lot of things about both characters. The hero is quite blase, chatty, undisturbed by the prospect of fighting Inigo. Inigo is actually a decent guy, if a little scatterbrained, and really cares a lot about swordfighting. Once they get to the swordfight, all this really comes to fruition; nothing particularly new is introduced about the characters, apart from the incidental fact that they are both right-handed. The only new thing is that both are fantastic swordsmen, and frankly you expected that.

Scene: Break up the sequence into scenes, and you're there. I would divide it into 4 scenes. First, the cliff. Second, the swordfight with the hero doing better, up until Inigo reveals that he is not left-handed. Third, the swordfight with Inigo doing better, up until the hero reveals that he too is not left-handed. Fourth, the resolution of the whole thing, in which the hero wins and refuses to kill Inigo; he knocks him out instead and runs off.

Structurally, then, a sequence is usually made up of scenes for exposition, dramatic tension, and climax or resolution. Here there are two dramatic tension points, because of the hand-shifting. The cliff is exposition -- it sets up the fight. The hero winning, knocking out Inigo, and running off is climax -- you don't want to waste time on this because it's sort of a foregone conclusion from the moment that the hero shifts hands. The fight is the dramatic tension -- it's the focus and the important part, and garners the most interest.

If you wanted to see this as straight RPG stuff, you could imagine resolving each scene mechanically with a single die-roll or equivalent, with the players filling in all the cool stuff that in effect explains each die-roll.

Action: Suppose you broke down each of those swordfight scenes into a whole bunch of smaller actions. Attack, parry, attack, parry, trip, recover, attack, parry, etc.

The point here is that the traditional RPG focuses at the action level. There's nothing inherently wrong with this, but if you think of the film the really interesting stuff doesn't happen here. The interesting stuff happens at the scene level: who wins? why? who shifts hands?

You could do the same thing with big kung fu fights in classic Hong Kong films. They have lots more scenes, of course, but the interesting stuff doesn't happen with every punch and kick; it's a kind of progression of larger blocks that are independently fun and interesting. You get new weapons, new physical settings for the fight-stage, new styles for characters to switch to, new characters to fight. Kentucky Fried Movie pillories this wonderfully by having the main kung fu guy fighting an opponent, and then the camera backs up to show a bunch of additional opponents waiting on line for their turns.

The advantage of the traditional action-based resolution is that you get lots of bang for your scene buck: each flash of the swords gets its own moment. The disadvantage is that you tend to forget the forest for the trees.

The advantage of the scene-based resolution is that you get story and narrative flow. The disadvantage is that you have to have the players fill in all those flashing blades through open description.

Hope this helps somewhat.

Chris Lehrich

Hi there,

Jay (coxcomb), you wrote,

I reread the combat section of Sorcerer and it sure seems to be action resolution to me.

Can you give an example of how an action in Sorcerer combat is resolved as a conflict?

Thanks Chris. Excellent example! Actually, I totally get what you're talking about.

It's funny that you should use the Princess Bride example. I do stage combat choreography, and I use that fight to show new fighters that "kewl" choreography doesn't make a great fight. First we watch the swordfight normally, and everybody cheers and has a great time. Then I start it again, but with the sound turned off. Thing is, the fight sucks on a technical level. Without the actors skillfully delivering their lines, it falls flat.

I think a big problem with many RPGs is that they encourage combat with the sound off. The focus is on which maneuver you should choose for maximum mechanical impact and not on what you are trying to accomplish, or how your actions relate to who your character is. (Yes, I know, I clearly don't mesh with Gamist thinking, in which finding mechanical impact is often much of the point)

My difficulty here is not in understanding breaking sequences into chunks and resolving those chunks. That part is instinctual for me and I've been doing it in my games for years.

My difficulty comes from extending the term "conflict resolution" to a system that is clearly resolving at an task level. Take a look a Sorcerer (I really don't mean to pick on Sorcerer, or Ron. I love Sorcerer, and admire Ron. I use it as an example because it is a game that talks conflict resolution--and I have a copy to refer to) Anyway, in the all things except for combat and some sorcery, Sorcerer clearly presents rules for conflict resolution. The combat is framed in rounds (each of which is of ambiguous length but confined to approximately two seconds of activity). During each round everybody states an action. Dice are rolled to determine the outcome of each action. Clearly task resolution.

So it would seem that Sorcerer (along with many other games) is actually a mix of conflict resolution (with variable scaling) for general actions and task resolution for combat. What Sorcerer does that a traditional task resolution system (e.g. D&D) doesn't do, is mechanically encourage playing with the sound on. That is, instead of encouraging careful selection of static advantages (feats, maneuvers, etc.), Sorcerer has dynamic advantages in the form of bonuses for creative description, clever tactics and so on. The bonuses you get for a roll are not described on your sheet, they are dictated by your ability to add appropriate desription to the scene. Success depends more on character and player investment and context than on character engineering.

I think many people confuse "task resolution" with playing with the sound off (sorry, this metaphor is really working for me this morning).

Am I making sense?

Hi Jay,

The way I see it, Sorcerer (and I accept that it's just an example, not "against" it or me) is based on conflict resolution that in addition provides some structure for describing the tasks within it.

Dust Devils, for instance, provides some structure for the tasks within the conflicts it resolves, but not quite as much as Sorcerer (i.e., open to more interpretation). And The Pool provides none at all, just says "goal fails" or "goal succeeds" and leaves narration of anything within that up to the designated person.

I think all of these are a lot different from task resolution alone, even when each action is accompanied by lots of descriptive color. Playing Feng Shui is a lot like that, in my experience. Conflicts are only resolved in Feng Shui, if at all, through the accumulation of many separately-resolved tasks through many "shots." Each shot has a lot of flash and narration associated with it, but that doesn't change the esssential structure of resolution.

Best,
Ron

Not to do a pure "me too" post or anything, but the "swordfight with the sound off" thing seems to me an amazingly good analogy. If y'all don't mind, I'm going to steal it for explaining this same issue in my own current game design. Nails and heads have been hit!

Actually, come to think of it, I went to a screening of the Errol Flynn Robin Hood and the sound on one reel was incredibly screwed up; it sounded like the movie was underwater. When it came to the big fight scene up and down the stairs with Basil Rathbone, the screeners came out, apologized, and asked us if we'd rather see it with the sound off or screwed up. We asked for "sound off," and believe me, it was a pretty mediocre experience at best. A little better than it might have been, since we'd actually asked for the sound off, but still mediocre.

Chris Lehrich

That makes sense, Ron. I still don't quite agree with the terminology, but I think we're down to semantics, so I'll stop arguing. ;-)

Thanks,

Steal away Chris! I'm actually rather pleased with myself about the analogy. Every time I play D20 (with my predominantly Gamist group) I feel like the sound is off.

Attack, Feat, Repeat. Ugh! Works for some, but not for me. :(

Cheers,

A reply to Chris and Jay --

First of all, my Threefold-Simulationist side feels compelled to point out the obvious caveat: that "The Princess Bride" is not an RPG, and what works for a film may or may not work for an RPG. It's worth talking about, but one should keep that in mind.

clehrich wrote: The point here is that the traditional RPG focuses at the action level. There's nothing inherently wrong with this, but if you think of the film the really interesting stuff doesn't happen here. The interesting stuff happens at the scene level: who wins? why? who shifts hands?

You could do the same thing with big kung fu fights in classic Hong Kong films. They have lots more scenes, of course, but the interesting stuff doesn't happen with every punch and kick; it's a kind of progression of larger blocks that are independently fun and interesting.

John Kim wrote: I think that depends on what you find interesting. Sure, if you rip each individual action out of its context, then its not interesting -- but that doesn't mean that the actions themselves aren't interesting within their context.

To be clear: I do not intend any value judgements about style of play. I know lots of folks who think the character engineering of a game like D&D (what feats and class abilities work well together to do what I want) is great. It seems to work for Gamist and some Simulationist play styles quite well.

I was trying to differentiate this style of resolution from that used in Sorcerer combat, where the tactics and strategy come more from adding to the narrative than from setting up your pre-planned perfect combo.

Of course there is lots of middle ground here too. As a long time Chamions/Hero player, I hear what John is saying about its style of task resolution. If you want to get your tactical groove on in character context, you can't get much better. But Hero does fall between the extremes. Even though there are specific maneuvers, the system encourages flavorful description, and having strong metamechanics enables the GM to roll with the unusual tasks without fudging the system too much.

So...
Task = tactical
Conflict = strategic

I mean, you can boil it down to that, can't you? Whether your testing the tactic, or testing whether the strategy works.

Of course, testing the strategy involves squishing down the specifics of the tactical.

And once you do this you realise that you can squish down each strategy and instead resolve a greater strategy.

Of course that has to ceiling out. You can't get to the point where your testing such a grand strategy that RP is also being simulated rather than done (not saying that's what's being here, just postulating a ceiling).

Just thinking...

Seems to me you're all still getting the two variables mixed up. Check this out: Task Resolution tells you whether you succeed or fail. Conflict Resolution tells you whether you win or lose.

Player: "I crack the supervillian's safe!"
Task Resolution: do you open the supervillian's safe?
Roll: Failure!
GM: "Nope, the safe's too good. You can't crack it."

Player: "I crack the supervillian's safe!"
Conflict Resolution: do you find out who is the supervillain's next target?
Roll: Loss!
GM: "Okay, you spin the dials and pop the safe open. There are a bunch of papers in there, but they don't say who the target is."

Both of these are at the single action scale, but the former's succeed-fail and the latter's win-lose. You can succeed but lose, fail but win.

In conventional rpgs, success=winning and failure=losing only provided the GM constantly maintains that relationship - by (eg) making the safe contain the relevant piece of information after you've cracked it. It's possible and common for a GM to break the relationship instead, turning a string of successes into a loss, or a failure at a key moment into a win anyway.

Whether you roll for each flash of the blade or only for the whole fight is scale, not Task vs. Conflict. You can Conflict-Resolve the whole fight in one roll or Task-Resolve the whole fight in one roll:

Player: "I fight him!"
Task Resolution: do you win the fight (that is, do you fight him successfully)?
Roll: Success!
GM: "You beat him! You disarm him and kick his butt!"

Player: "I fight him!"
Conflict Resolution: do you make it to the ship before it sails away?
Roll: Victory!
GM: "You disarm him, kick his butt, and then run down the dock just in time! They let you on board and then pull up the gangplank!"

-Vincent

I'm really sorry.... but I've read the post trhough twice now..... and I still don't get it......why does it have to be different?

Win / Lose
Succeed / Fail

Not alot of difference... infact, I can only see it as being personal and impersonal, yet the mechanics are still the same, and the scale can be as well.

The only thing that is different is whether its a general check or a specific check.

the idea of having a fight, running to the docks, then trying to get to the ship.....
3 different things! The first is combat... you need to beat the opponent... failure results in death etc. Success means you can move on.... yet time may be important!
The second is the running, so you need to check if you are running fast enough, possibly avoiding obstacles, so a few more checks, much like the combat situation, you need to succeed to win, failure results in overal failure, and time is still important!
The third is leaping for the ship. This may be a Static thing, or may be altered by the speed at which you achieved the first two things. You check to see if you make the jump. Success means you made it, failure means you hit the water!


So you have the big picture, which could be resolved with one check,......
so, do I beat the guy, make it to the docks in time and leap aboard.....result of success = yes, you win! Failure = No, you lose!
Or you could handle it the three seperate parts, with failure at any point resulting in failure throughout!
Basically, it seems like the difference between a single resolve or an extened resolve!

So, if I have got it wrong, or round the wrong way, or I'm not only holding the wrong end of the stick - but a completely different stick, pease explain it to me in simple english, (apparently its the language I'm meant to be able to speak, yet I think they lied!)

The fight on the dock example doesn't work, okay. How's the safe example for you?

The key to the two kinds of resolution is: what's at stake?

In task resolution, what's at stake is the task itself. "I crack the safe!" "Why?" "Hopefully to get the dirt on the supervillain!" What's at stake is: do you crack the safe?

In conflict resolution, what's at stake is why you're doing the task. "I crack the safe!" "Why?" "Hopefully to get the dirt on the supervillain!" What's at stake is: do you get the dirt on the supervillain?

Win / Lose
Succeed / Fail

Not alot of difference... infact, I can only see it as being personal and impersonal, yet the mechanics are still the same, and the scale can be as well.

For a while now I've been trying to discuss "alternative" role playing with a few gamers I really suspect would be into it (by alternative I mean less gamist in simulationist clothing, probably with a focus on narrative play, but mostly just getting away from the game style we have right now which frankly no one seems to enjoy that much) for a while now. I've tried to get people interested in some games, I even got a couple started to the point where characters were made and a first session was played. But nothing much has been going on. I suspect a large part of that is because there is another player who is VERY strongly against (most games he plays are in his homebrew system and he has a tendency to lord his authorness of the rules to give him rights over the other players) who is very vocal against any such movement.

I've been trying to get people to read the forge for a while, because I'm really convinced that given

I a copy of this discussion and brought attention to the examples lumpley gave. I got responses with the email equivalent of lights going off in people's heads. Thank you thank you thank you. That was one of the better explanations I've seen here.

lumpley wrote: In task resolution, what's at stake is the task itself. "I crack the safe!" "Why?" "Hopefully to get the dirt on the supervillain!" What's at stake is: do you crack the safe?

In conflict resolution, what's at stake is why you're doing the task. "I crack the safe!" "Why?" "Hopefully to get the dirt on the supervillain!" What's at stake is: do you get the dirt on the supervillain?

John? The safe example is short and easy. It illustrates the principle. It does not describe the bounds of the approach.

You want an example of conflict resolution that a) includes many subtle shifts in intent, b) has moment-to-moment emotional choices, where the characters c) have multiple reasons for doing things, d) have some reasons that give them pause, and e) are trying to accomplish multiple goals?

The safe example ain't it. I am not startled to learn this.

You wrote: Consider "I punch him" instead, where "why" is "To try to incapacitate him". There isn't much difference between the methods here.

John Kim wrote: For example, in the last Buffy game I was in, at first Dot (my PC) was concentrating on killing a vampire she was facing. But then her friend Max was hit, and she shifted her priority more to protecting Max. But then Max cast a spell of protection for herself, and Dot was able to concentrate more on slaying.

Conflict resolution mechanics are not simply task resolution mechanics moved to a larger scale (where you correctly note you'd lose the desireable level of resolution). Conflict resolution mechanics must be structured so that all of the tactical fiddling is shifted to the roll itself.

Good examples of this would be Story Engine, Trollbabe, and Universalis where a whole bunch of ==stuff== goes on before the dice even hit the table. Its at the level of this "stuff" where all of things you wouldn't want to miss out on occur.

lumpley wrote: You want an example of conflict resolution that a) includes many subtle shifts in intent, b) has moment-to-moment emotional choices, where the characters c) have multiple reasons for doing things, d) have some reasons that give them pause, and e) are trying to accomplish multiple goals?

The safe example ain't it.

lumpley wrote:

John Kim wrote: Consider "I punch him" instead, where "why" is "To try to incapacitate him". There isn't much difference between the methods here.

Sure there is. Imagine, if you will, a mechanic where we roll a bunch of dice to determine whether you incapacitate him, heavily modified by traits on your character sheet and your expenditure of resource, and flip one coin to determine whether your punch lands. Here are the four possible outcomes: punch lands+incapacitated; punch lands+not incapacitated; punch misses+incapacitated; punch misses+not incapacitated.

John, what might be more interesting than adding "punch misses+incapacitated," is doing away with "punch misses" entirely as a mechanical outcome. Just don't flip the coin, as it were. Punches miss only when the player decides to have the punch miss, as a way (one of many possible) to explain the "not incapacitated" outcome.

-Vincent

lumpley wrote: Seems to me you're all still getting the two variables mixed up. Check this out: Task Resolution tells you whether you succeed or fail. Conflict Resolution tells you whether you win or lose.

Player: "I crack the supervillian's safe!"
Task Resolution: do you open the supervillian's safe?
Roll: Failure!
GM: "Nope, the safe's too good. You can't crack it."

Player: "I crack the supervillian's safe!"
Conflict Resolution: do you find out who is the supervillain's next target?
Roll: Loss!
GM: "Okay, you spin the dials and pop the safe open. There are a bunch of papers in there, but they don't say who the target is."

Both of these are at the single action scale, but the former's succeed-fail and the latter's win-lose. You can succeed but lose, fail but win.

*snip*

You know, I think there are probably some D&D players who treat the combat rules a lot like conflict resolution. It all depends on how you interpret what hit points mean.

Use of skills is pretty much "do I make the skill roll or not," but hit points are a little less defined than, say, Arrowflight's wounds. Does a loss of 6 hp mean a small wound, or a glancing blow, or that your luck and intuition spared you completely? Likewise, a "hit" implies that the strike delivers damage. That means a "miss" could also be a glancing blow.

John Kim wrote: Well, I agree there is a difference. But how much of one? Three of these are frequent results of traditional combat mechanics. The unique result to conflict resolution is "punch misses+incapacitated". Now, this is definitely a difference, but how important is it? How will my combats be different by including this result (miss but the opponent is incapacitated) in practice? I'm not sure about the answers here -- in practice for me conflict resolution has always been at a more strategic level. But my instinct is that the difference seems minor.

cruciel wrote: The key difference in my experience is whether you are focusing on preserving the players intent (conflict) or focusing on preserving causality driven tension (task) - if that makes any sense at all. By going conflict resolution you're making an active effort to preserve the players intent, and hence avoid frustating moments like:

• Player, "I punch him in the head"
(The player's intent is to incapacitate the target, but is not vocalized and resolution is tasked based)
• Roll, Success!
(GM checks the target's sheet and finds the trait 'Cannot be incapacitated with punch in the head')
• GM, "You nail him square in the jaw, and he seems unfazed - no effect"
• Player while making cranky face, "Grrr..."

Of course, task resolution adds that bit of unknown, adds that bit of tension:
• Player, "Holy shit! He's immune to face punchy! Crap, what now? 'Sock 'em in the kisser' is the only trait I took!"

My Slayer PC, Dot, staked a vampire through the heart but he turned out to be immune to staking. Sudden plot twist, and Dot thinks "What the hell?" So she had to grapple with him, and in the meantime a minion of his hit her friend Max which she felt really bad about.

Using conflict resolution, in a game like Trollbabe, this sequence might instead read:

cruciel wrote:
• Player, "I punch him in the head"
(The player's intent is to incapacitate the target, but is not vocalized. Resolution is conflict based)
• Roll, Failure!
• GM, "You describe how you failed."
• Player "I clock him right on the mouth, but he just grins at me. He's got some magic sympathy with the rock. I grimace and say 'Oh crap.'"
• Other Player "Yeah, that's cool."

I have another example to offer, using the HW mechanic, to demonstrate a scene level resolution.

Lets say we have a mighty broght low and sent to the Roman arena. The SCENE we want to run, rather like a movie montage, is one in which this character cust there way through sundry opponents and lives or dies by the end of ther day.

So, in the HW mechanic, you could run the whole thing as a single extended contest. each EXCHANGE in the contest represents a single opponent. It can easily be presumed that all opponents bar the last are going to die; so each exchange resolves how they die and what effect this has on the viewpoint characters AP total.

So after 4 exchanges, the viewpoint characters AP will have changed and 4 opponents will have been dragged from the sand. Each exchange resolves what is really a whole duel in a single action; but this permits the narraytion to move swiftly on and maintain seom dramatic momentum, without getting bogged down in exchanges of precise blows.

This is much more cinematic - in a literal sense - than most action resolution methods, because the tasks which comprise the overall challenge are reasonably abstracted while the challenge as a whole is mechanically rigorous.

John Kim wrote: Ahhh!!! Thanks, that's a great example. That resonates with me particularly since this almost exactly happened in the last Buffy episode. My Slayer PC, Dot, staked a vampire through the heart but he turned out to be immune to staking. Sudden plot twist, and Dot thinks "What the hell?" So she had to grapple with him, and in the meantime a minion of his hit her friend Max which she felt really bad about. They eventually destroyed the vamps, and they are now in the process of figuring out why he was immune and what to do about it.

I liked it, but I agree with you that people's preferences can and will differ.

coxcomb wrote: It seems as though some parts of a game will always need more in-depth resolution. That is, if I get to the end of a session and the Big Bad is waiting to kick my butt, I (pretty much regardless of GNS) would be let down if that dramatic conflict were reduced to the fickle fate of a single roll.

I'm sure there is something key here that just isn't making it through my thick skull. Can anyone clear this up for me? Any clever examples of play that illustrate how to make this work?

coxcomb wrote: It seems as though some parts of a game will always need more in-depth resolution. That is, if I get to the end of a session and the Big Bad is waiting to kick my butt, I (pretty much regardless of GNS) would be let down if that dramatic conflict were reduced to the fickle fate of a single roll.

I'm sure there is something key here that just isn't making it through my thick skull. Can anyone clear this up for me? Any clever examples of play that illustrate how to make this work?