When does combat resolution become too slow?

[Callan S.]

Zero abstraction is reality.

You can't do zero abstraction.

Some things will be abstracted more than others.

What should be abstracted more and what should be abstracted less?

These discussions always go wobbly when people turn to reality for the answers of what to abstract in reality, when really they are turning to their own personal preferences. Yes, that's what you'd be doing. Because there's nothing else you can turn to except personal preference.

"I want to make a realistic system"
"By who's opinion?"
"Not by any particular persons opinion, I just want it to be realistic!"
"How about your own opinion?"
"No, my opinion is only that realism is good and that's what it's going to be based on!"

Problematic.

[Bill Cook]

I'd like to make an unprofound comment. Even if your design matches handling time to aesthetic, to be played, it must be internalized. So is d20 faster than the Hero system? Yes. Unless it's that guy named Ed I met at the meetup last month, running Hero. (He's like, a 20 year vet.)

Dauntless:

I think what you're after (and what Mike broached) is elegance. (I enjoyed that quadrant analogy, BTW.) Since I've been reading Burning Wheel lately, I'll use that for examples.

[*]To contrast with D&D, instead of rolling d8+2 (or what have you) to determine damage for a sword, you consult its Power (e.g. 8) and roll a d6 as the Die of Fate; ostensibly, to fine grade location a la TROS. Then, (no, we're not done yet,) for 1-2, you reduce Power by half, and for 5-6, you increase Power by half; for 3-4, use base Power. (e.g. 4, 8, 12.)

So you can scale with a d6 regardless of the base. But you have to employ division and rounding. Do the input requirements justify the output? It's a stretch, IMO.

[*]Another example is armor failure. Anytime you roll one or more results of one with your protection dice pool, remove a die to reflect weapon damage, causing your armor to lose function. When I read this, I thought, Gak! More overhead . . . so I can lose function! But when you playtest it, it's just thrilling (to me). I don't know why.
[/list:u]

I'm sure you've processed the "design for the play you want" comments. I think it's at least as valuable to qualify the cost/pay trade-off as not being a play barrier.

** ** **

Hey, I just had the thought: maybe you should ask for examples from designers of their lightbulb moments in moving closer to the low cost/high pay corner.

Precious Villain:

I disagree in one respect; consider the attraction of the sub-optimal character. Back your way, some: if you're shooting for a Gam thing, and you want a kit, some rock-paper-scissors balancing may be required.

Even so, wraiths in Starcraft: Broodwar will always suck:)

[Edited: fixed poster names and tail end of my comment to Dauntless.]

[Doug Ruff]

Hey, I just had the thought: maybe you should ask for examples from designers of their lightbulb moments in moving closer to the low cost/high pay corner.

May I offer my own "lightbulb moment" as a designer-in-training?

Detail in combat is normally there for two reasons:

(1) To provide an engaging tactical system.

(2) To mediate the description of what actually happens.

For example, critical hit rules; under (1) these provide an extra element of risk to the proceedings, and usually give the lesser combatant an additional chance to prevail through a lucky strike.  Under (2) these may also provide a list of charts which describe where the critical blow landed.

My theory is that (1) gives much more value than (2). Wherever possible, combat mechanics should be geared towards generating mechanical effects, and the descriptive elements should be left to the GM and/or players.

Back to the previous example; critical hits that generate a penalty to future actions, or additional damage dice add something to the mechanics. A critical hit location table which lists different locations and their effects also adds something to the mechanics, but delivers less 'bang for buck' in that it adds handling time and limits the creativity of the participants in describing their own effects.

My subsequent -2 to actions could be because I tweaked a hamstring, or because I am scared by the prowess of my opponents, or because I just tripped over a duck - does it matter?

Of course, this is a personal revelation, and it may be that for most people this is either trivial, or not what you want. But it was a big deal to me, and it's cured me of my previous hangup about combat mechanics, which was that they should deliver 'realism' on a plate, unmediated by any conscious thought by myself or by my gaming group.

I think this takes me back to addressing the original question: combat resolution becomes too slow when players are relying on the combat mechanics to do the thinking for them, instead of thinking about how to best use those mechanics. And that is my personal "lightbulb moment."

[Dauntless]

Let me broach the subject of realism from another viewpoint to try to help explain why I think it's a good idea to model things based off reality.

As I explained early on, the job of a game is to model a system.  Models by their very nature are not perfect and can not capture in totality everything about that system (only reality can prefectly describe reality).  Some have posited that realism, and hence accuracy is not what we're truly after but rather verisimiltude....the appearance of reality.  I couldn't agree more.  Doing a good model design is doing precisely that.  However, there is a subtle difference between what we the players experience as reality through our own subjective filters, and how the real world really works.  As Mark Twain said, "the truth is not only stranger than you imagine, it is stranger than you can imagine".  The disparity between accuracy of reality and the versimilitude of reality comes from our learned and created misperceptions.  For example, I remember talking to someone who grew up on a diet of cheesy 80's shows who thought that the explosions in Saving Private Ryan were stupid because they weren't fiery explosions.

I want my game to have the appearance of truth as dicated by how closely I can get my system to approximate the truth.  Sometimes however, there's simply no way to model what happens in real life with any consistency or accuracy.  We have no choice but to abstract things and hope that the results turn out to feel right for the situation.  For some strange reason though, many people who are against having detailed rules are against this notion.  In their minds, if you're going to base your rules calculations as much as you can by reality constraints, then everything should be.  Look at it this way, humans got along fine without science for untold generations.  Science is still far from perfect, but by following what they have revealed about how things work, look at all the innovations (and nightmares) that have come from it?  The same can be applied to game design.  Following reality in a scientific approach to help develop the game rules can reveal not just what outcomes are achieved, but how and why they are achieved (and hence, you start to realize the interdependencies of things).  The what of things is trivial, you don't need realistic rules for that.  But discovering how and why things are the way they are much more vital and if you want consistency, then looking at reality, and designing things logically are crucial.

Here though we run into the problem as Noon pointed out...."My expectations (perceptions) for reality are different than yours, so achieving realism is impossible".  Ultimately, all human beings are subjective creatures and therefore reality is subjective.  In effect, one can't even prove that reality is objective because hard as you try, everything we know about reality is nothing more than a mental state.  For that matter, one can not disprove the argument that my mind is the only mind inexistence (solipsism) nore conversely prove that other minds exist.  However, we assume that there is an objective reality and that there are seperate minds because these imagined mental objects seem to repeat behavior and share behavior similar to our own.  It is this repetition of observable effects from a given cause that we create this concept called reality (and I'm talking "real" reality here...not just game reality).

I'm not afraid to let my game system produce results which go against player expectation or view of reality.  As was explained to me by M.J Young about credibility and authority, the players may not be totally convinced by the outputs my game delivers, however, I would appeal to the authority of statistics of similar results that the results are fairly accurate.

Perhaps more to the point, the realism isn't there to say, "see, I have the most realistic game system on earth...now bow down to my simulationist design...".  Reality as I mentioned earlier is subjective though it seems to follow a pattern of predictability.  From this predictability we are better able to make choices about what to do and more importantly what to expect.  And that's exactly what a game system reality should do.  It should provide an underlying mechanism of "reality" (how things happen in the shared imagined space) that helps players make choices and calculate what to expect.

Now as others have mentioned, often, games which are too tactical provide too much knowledge and predictability for the players.  This has two drawbacks.  The first is that it eats up time (my primary concern).  The second is that it grants too much knowledge about one's odds.  To answer the second problem, even if the inputs are well known and their effects on probability are well known, there are two easy methods to solve the "odds calculation problem".  Firstly, players often have no control over some of the inputs.  Secondly, it's very simple to create hidden or obfuscated variables.  Hidden variables are inputs or initial conditions the player is unaware of, obfuscated variables are variables the player is aware of, but not the precise quantity/quality.  For example, in my system, the injury of a player is always obfuscated.  Only the GM knows the exact level of damage a player has (it's life threatening seerity, the minuses to proficiency, etc.).  The GM can give word descriptions that will give the player an idea...but the actual number is hidden.

As a matter of fact, it is precisely the hidden and obfuscated nature of input variables that makes modeling an algorithm difficult.  Because the designer himself isn't sure exactly how a certain task works, this same uncertainty can be applied to the players.

Lastly, verisimilitude is the key as noted before.  All games try to create effects which fit a certain perceived outcome.  This is why some games are described as gritty, dramatic, light hearted, etc.  It's not just setting which creates this tone, but how the game mechanics support the outcomes of events.  Realism as the word is usually used denotes that the outcome in the game will try[/] to match the outcome as happens in the real world given the same inputs.  Realism can also imply how the game mechanics themselves were designed.  Descriptive design is generally speaking not a realistic way to design things as compared to procedural design.  As an example, most modern systems that allow you to build a character with a point system and choose skills however you like (a descriptive design approach) is not very realistic.  In this sense, class systems and random rolls to build things like attributes are in fact more realistic because they mirror more closely how reality works.  Descriptive design allows one to create the desired object as long as you "pay the points for it".  Procedural design can get you close to your desired output as long as A) you have the right inputs,  b) you are lucky enough or C) can afford the design requirements/prerequisites (mass, cost, social class, minimum trait score, etc).

So in the end, "reality" is nothing more than a design goal that allows the game to have a certain feel and flavor.  Furthermore, nothing can be 100% accurate and hence realistic.  This is not a realistic game's goal (to be 100% accurate).  A realistic game's goal is instead to give an idea of the expected outcomes and hence, to let the players know, "this is the game's definition of perceived reality....if yours matches, then the game will be credible".

[TonyLB]

Okay, I think I understand your point of view.

But you don't seem to have any more questions.  You've stated your personal answers on the questions that have been raised.  And now you're working hard to make sure that people don't misunderstand you.

Does that mean that, if I do understand you, there is nothing more for me to contribute?  I feel like we've learned a lot more about your approach than you've learned about any of the alternatives that might benefit you, but maybe you're just a naturally selfless and outgoing person and don't mind that.

[Dauntless]

TonyLB-
Not at all.  I'm trying to figure out a good balance in my tradeoffs between playability and accuracy.

As I gave in an example earlier, what seems a better tradeoff:

1) A resolution system that takes 30 seconds and is within 15% of the true score 90% of the time.

2) A resolution system that takes 2 minutes, and is within 5% of the true score 95% of the time.

It's these sort of considerations that I'm not really sure about how to go forward.  Is option #1 close enough to reality and with a good amount of playability?  Or because #2 is close to the real expected value even though it takes longer (because it has more tactical choices) a better approach?  It may seem like #1 is superior because it's pretty close to the real value, and it's relatively quick.  But just because #2 is longer doesn't necessarily mean its less fun.  Maybe having more tactical choices as inputs is fun in and of itself (for some people).  These are the questions that I haven't determined how to resolve.

I haven't included stuff like deviation from the mean or other statistical analysis for the moment (mostly because I'm better at stochastic analysis than I am at statistical analysis).  But they are nonetheless factors to consider.  As an example, 10% of the time in option 1, the error is more than 15%....but how much more than 15%?

[LordSmerf]

As has been mentioned i believe that this is entirely a case of audience.  Who plays your game?  If the players are hard-core tactical thinkers then your second option is probably the way to go, especially if the system is interesting in and of itself to resolve.  If your players are not really interested in those tactical options then your first option is the way to go, or possibly an even simpler option.

I guess what i am saying is that there is no right answer.  You tell me who your audience is, and then perhaps i will be able to tell you how much is "too much".

Thomas

[Dauntless]

My intended audience will be players who are interested in having lots of tactical options and other crunchy tidbits.  Players who like wargame/roleplaying games like Heavy Gear or Battletech would probably be who I'm aiming for.  Players that like the Riddle of Steel or the Hero system would be another example.  Wargamers who are interested in doing roleplaying are another potential category (though I'm talking more about historical wargamers as opposed to fantasy/sci-fi wargamers, the former being more concerned with accuracy, the latter being more concerned with play balance).

However, I'm also going to try to get rid of the min-max, power munchkin and rules lawyers by introducing the hidden/obfuscation system.  The rules lawyers can never exactly calculate whether their attempt is successful or not with a given die roll because they will never know the exact value of every input that goes into the calculation (only the GM will know).  Min-maxers will be minimized because probability outcomes are always contextually bound.  What may work really well within one set of circumstances will be useless in another.  Power munchkins will not have fun with my system because even though my system isn't really a point-cost based system, it's not really a random class based system either.  It is possible to play extremely unbalanced characters however....for example, you can play combat androids which all around are better than most other types of characters, but do have a few significant flaws (other than Quandroids, quantum androids, they are not truly free-willed and sapient...though they are sentient, hence they must obey certain directives and orders utterly).

[LordSmerf]

However, I'm also going to try to get rid of the min-max, power munchkin and rules lawyers by introducing the hidden/obfuscation system.  The rules lawyers can never exactly calculate whether their attempt is successful or not with a given die roll because they will never know the exact value of every input that goes into the calculation (only the GM will know).

A word of warning, take it as you will: Any system in which the GM has exclusive knowledge regarding some aspect of the mechanics gives rise to some serious potential conflict.  How do the players know that the GM is not just changing the values behind the scenes to screw them?  Maybe the GM is simply making the obfuscation "what i rolled +1" so that i can never succeed, or at least never succeed with out the GM's explicit approval.  Additionally, without knowing your plans in detail, obfuscation rarely has a significant impact on the "rules-lawyers" (who is generally Calvinballing) because they can simply argue that, whatever the GM's obfuscation choices were, that they were wrong since statistically "i should be able to do this".

I do not know that the above is really helpful, but i just wanted to draw attention to the dangers of GM determined obfuscation in a game geared toward players who want to "win".

Thomas

[Ben Lehman]

However, I'm also going to try to get rid of the min-max, power munchkin and rules lawyers by introducing the hidden/obfuscation system.

BL>  This is impossible.

A sophisticated tactical game requires sophisticated players and consistent rules.  Do we agree on this?

What are mini maxers other than players making sophisticated strategic choices?  (And "power munchkins," as far as I can tell, is just a slur for minimaxers.)

What are rules lawyers other than those that demand consistent rules?

You should stop trying to solve a dysfunctional social contract by giving the GM total authority.  It doesn't work for a tactical game.  Never has.  Never will.

Instead, thing about how you can maximize meaningful tactical and strategic decisions at every point in play.  Further, figure out how you can make those decision points move quickly.

yrs--
--Ben

[Dauntless]

Well, for a game which is strictly limited, constrained or defined by rules, it will always be in danger of abuse by rules lawyers.  If the GM has the ability to override the rules, the consistency and possibly the accuracy of the system comes into question.  This is in direct contrast to what the system is trying to achieve.

But I tend to look at rules the way judges look at precedents.  The rules are there to help him in his arbitration, but do not automatically confer what happens in the imagined space.  I believe this is what seperates wargames from roleplaying games.  If rules are the final arbiter, then you have a wargame.  Referees in wargames exist solely to judge on ambiguous issues, such as whether a target is in line of sight for example.  Once in a blue moon, a referee may decide on the outcome of an issue not explained in the rules.  But wargames generally only allow actions which are clearly spelled out in the game rules itself, so this problem crops up infrequently.

RPG's on the other hand are not wholly constrained and allow for open-endedness.  This open-ended aspect is what requires the GM to arbitrate certain things because the issues may not be covered in the rules, or the cases may be ambiguous.

I would rather the game allow for potential abuse of the GM in manipulating events than allowing the rules-lawyers to always calculate to the nth place what his odds are for doing something.  Players should have a good idea, but they should not be privy to information they as characters would not be privy to.

Powermaxers are more easily countered by making sure that there is a logical chain of events that creates the character, as well as ensuring a broad variety of situations that encourage a level of non-speciality.  Powermaxers almost always tend to be one-dimensional...throwing them into another dimension makes them ineffective, and will (hopefully) teach the player not to design characters like that in the future.  

Besides, who really wants to play with those kinds of folks in the first place?  Hopefully the social contract will weed them out in the beginning.

[LordSmerf]

Powermaxers are more easily countered by making sure that there is a logical chain of events that creates the character, as well as ensuring a broad variety of situations that encourage a level of non-speciality.  Powermaxers almost always tend to be one-dimensional...throwing them into another dimension makes them ineffective, and will (hopefully) teach the player not to design characters like that in the future.  

Besides, who really wants to play with those kinds of folks in the first place?  Hopefully the social contract will weed them out in the beginning.

This statement is a horribly, horribly narrow minded one.  In answer to your question: I like to play games with min-maxing and powergaming.  Not exclusively mind you, and not all that often, but every once in a while i get an urge to generate a horribly broken character and play a couple of sessions with other horribly broken characters and do all we can to illustrate just how broken we are.  It is important to note that nothing is wrong with this inherently.  This is a valid form of role playing.

It seems to me that you might be developing a system that caters to your interests while trying to invalidate all other role playing interests.  There is nothing wrong with that, in fact it can make your game more coherent and playable.  That said, i highly recomend that instead of trying to eliminate one type of play through your rules that you try to encourage the type of play you do want.  Do not design the rules so that "such-and-such" will not happen, but instead design the rules so this "other such-and-such" will happen.

I know that Ron addresses the very issue that you seem to be aiming at in either GNS and other matters of role playing theory or in Simulationism: the right to dream.  So i highly recommend that you spend a couple of hours reviewing those two essays.

Thomas

[Ben Lehman]

RPG's on the other hand are not wholly constrained and allow for open-endedness.  This open-ended aspect is what requires the GM to arbitrate certain things because the issues may not be covered in the rules, or the cases may be ambiguous.

BL>  I figure we disagree at a rather fundamental level here.  By your definition, most of the RPGs I play would be wargames.  Even the non-tactical ones.  If following the rules means a wargame -- well, hand me some minis and call me a wargamer.

I would rather the game allow for potential abuse of the GM in manipulating events than allowing the rules-lawyers to always calculate to the nth place what his odds are for doing something.  Players should have a good idea, but they should not be privy to information they as characters would not be privy to.

Is there any reason not to just hand over all the arbitrative power to the GM, then?  I'm thinking here about something like http://home.freeuk.net/henridecat/">Free Kriegspiel  I'm actually involved in one of these right now.  It's a lot of fun and a great game.

But I wouldn't say it has much in the way of tactical depth.  We'll see, as the game progresses.

I'll say it again: strategic and tactical depth requires consistent rules.  If you dismiss any player who wants consistent rules as a rules lawyer -- well, that's fine, but you won't be designing anything with strategic or tactical depth.

Besides, who really wants to play with those kinds of folks in the first place?  Hopefully the social contract will weed them out in the beginning.

Why have you designed a game to counter people that you won't be playing with anyway?

yrs--
--Ben

[Dauntless]

Unfortunately in the wargaming world, there are quite a few min-maxers and rules-lawyers.  Many wargames aren't so much about recreating what-if situations as about calculating every little advantage to win.  Historical wargames tend to be better about this as a) the gamers are usually interested in recreating authentic real world battles which were often lopsidedly unfair and b) the rules they use often try to account for the human side of things (command and control, morale and unit integrity for example).  In the strategy game I am also developing, I'm also doing things to weed out this behavior as well.  In fact, in the over-arching storyline, the only way to win....is to lose.  This is anathema to most rules-lawyers and min-maxers, whose main motto is to have power to achieve victory.  So for once, you'll see a proponent of wargames who discourages the victory balance gospel that pervades wargame design.

If your social contract has many rules-lawyers and power maxers, then more power to you.  What's important is that everyone is on the same page in the same book.  I personally don't want that, and I want to create rules which diminish the possibility of such behavior.  The essence of what I want to create is being able to put yourself into another's shoes and find out what the possibilities are given the world setting and character abilities.  Power for power's sake (min-maxers) is not welcome, nor are rules-lawyers who try to use artificially gained (OOC) knowledge to enhance their abilities in the game.  A character will not know his exact chances to do something but unfortunately any player who wishes to be OOC can easily deduce the odds of something or argue a point to increase his chances (or influence what action to take).

As for realistic rules requiring consistency, this consistency is based upon the judgment of the GM.  By adhering absolutely to the rules, you run into several problems.  Firstly, no rules system can account for every possibility, hence there will always be a certain amount of fudging.  Secondly, the rules are consistent....the GM is merely hiding the variables from the players so that they are unable to calculate their precise odds.  Just because there is information which is unavailable to the players doesn't mean that the GM has necessarily broken all the rules.  That possibility should always exist for the GM, but it's there only when necessary.  Now you might say that because either the information is totally hidden (the player doesn't even know it's a variable to be counted) or obfuscated (the player knows it's a factor, but he doesn't know the exact value of the variable) then in effect, the credibility of the system is broken because what the player perceives as his chances, and what the actual results are differ.

But I posit that that is what makes the world interesting.  If the world followed our predictions all the time, it'd be pretty boring.  The player must then learn to associate indirectly why his predictions were wrong.  Perhaps his wound was greater than he thought.  Maybe the character was more affected by fear than he thought.  And perhaps, just perhaps....the character isn't as skilled as he thought he was (I'm playing with the idea of having non-numerically quantified attributes).  Afterall, how would you really roleplay someone who was overconfident?  You could of course just act it out.  But wouldn't it be far more interesting if you the player saw the values of your skills and abilities as greater than they actually were?  Eventually, you might piece 2 and 2 together and realize that your abilities were not as great as you thought....exactly like you would in real life.  Afterall, do you know exactly how good your skills and attributes are?

This uncertainty factor is very important to me.  The underlying system however is quite consistent in its own internalized logic.   Absence of knowledge does not mean absence of consistency.

[Dauntless]

It seems to me that you might be developing a system that caters to your interests while trying to invalidate all other role playing interests.

I'm not so much trying to invalidate other approaches as validate why I'm designing my game the way I am.  If anything, I've noticed that if you use the word realism, it seems you'll get a number of people trying to dissuade you that such a thing is either possible or desirable.  I agree on the possible part, I disagree on the desirable part.  I didn't mean to come across that my way of doing things is the only to do things.  I'm simply explaining that I'm creating the rules the way I am for the reasons I have given.

Actually, I think it's a very good idea for designers to state the reasons why they designed the game they did because it helps the players understand the games possible strengths and weaknesses, while also helping the player decide if he wants to buy into the system (although my rules are actually going to be free...so I could care less about that).