The Forge Archives

This is a post sort of inspired by Shreyas Applied Design post and others.  It's a notional mechanism offered for comment, and in this case aimed at a reinterpretation oof conventional RPG combat and IIEE concerns.

Its called the 'Sword of Damocles' because, contra all normal starting assumptions for RPG combat, the participants are assumed to be in imminent and immediate danger of death at all times.  Sudden death is the DEFAULT condition, and all character actions aim at preventing it.  In short, combatants can be visualised as having the sword of damacles hanging over their heads.

Rule:
Any failure on a combat action results in the sword falling on the acting character.

This obviates the need to determine or resolve blow by blow type combat.  I think it should also eliminate quite a lot of weapon modifier type stuff.

It seems to me this would work best associated with a sort of escalating bidding device.  So a sample resolution model would be something like this:

- players propose actions with a difficulty number
- they then roll some dice trying to get higher than this difficulty
- difficulties accumulate through multiple actions by both parties

So if Joe and Bob are fighting by means of a system in which they each have, say, 2d6 to roll, it might be something like:

Joe proposes a 4-point difficulty opening thrust, rolls, gets more than 4, stays alive.

Bob proposes a difficulty 2 beat, total difficulty is now 6, rolls more than 6, stays alive

Joe proposes a 4 point feint-and-lunge, difficulty is 10, rolls below 10, dies.

Note that there is no fixed time to any action required for this resolution.  Nor do we need to know much of the detail of the action proper.  Any given action comprises only a single difficulty number incorporating what would, in a more orthodox system, have been modifiers.

Also, because only the ACTING player is dicing with death, rolling initiative is obviated.  Players can just look each other in the eye and decide whether they wish to act or not, and thereby carry the risks.  On the other hand, seeking to raise the difficulty to a point that your opponent cannot achieve is viable, so you might deliberately carry out safe actions you will automatically succeed at in order to raise the overall difficulty.

I suspect also this removes much of the need for armour and damage mechanisms, the bulk of which would probably be represented by manipulating the dice rolled and base or starting difficulties.

Any thoughts appreciated.

Two things:

I don't see any motivation in this for anyone to act.  At all.  Just wait the other person out, and let them kill themself by rolling.  Perhaps you can clarify what they achieve other than "not dying"?

Connected... it seems a bit odd that somebody could kill themself with a bad roll, even absent the other character having done anything.  For instance, if Joe fails his first roll than his opening thrust results in his immediate death.  Yes, yes, I see how it can be narrated, but it still seems odd.

An interesting concept.   I'd suggest adding an additional layer, however.

When you fail a roll, instead of dying, you become vulnerable.

If your opponent succeeds their next roll THEN you die.  However, if they fail than THEY are vulnerable.

What this means is that if you failed your roll, in order for them to roll and kill you they have to roll at an even higher difficulty than the one you failed.  And if they fail, then they are vulnerable.  That would mean the player would have a choice, to escalate and risk death or break off.  Breaking off the combat would automatically succeed after any failed roll by the opponent.  So the choice is concede/escape or press on.

You'd need a rule to deal with situations where the number has increased to the point where neither party can succeed at it.  Perhaps if both combatants fail in a row than the TN is reset or another die added to their pools or something of that effect.

I'm going to agree with Tony.  As a player, I set the possibility of death, then immediately test to find out if my character is gone?  Has a feeling more like Russian roulette than combat or conflict.

Perhaps if the Proposal (or Bid?) were to be tested by the opponent first...
Meaning, Joe proposes a difficulty for Bob.  If Bob beats that difficulty, then Bob proposes a difficulty for Joe that stacks on top of the difficulty that Joe made for him moments ago.

Of course, this way there would have to be a limit on how much one could increase the difficulty by in a single bid, otherwise there wouldn't be any reason not to bid the maximum difficulty right away, in an effort to keep your opponent from bidding back on you.

But over all, I've gotta say it sounds kinda boring.  I'm having a hard time imagining this type of super-simple resolution model being used in anything except a game where you intend to keep the players from wanting to get involved in combat.  Of course, I'm just assuming right now that the death of the character means it's time to stop playing that character and make another.

-Eric

In which case there'd still not be a reason to act first - because you'd be the first to become vulnerable.

That's not neccesarily a bad things, though - there are plenty of schools of thought where both skirmishers try to get to other one to commit first. Place for a 'taunting' ability here, if desired.

Of course, time pressure might also be a reason to commit - can't very well wait out the lone infantryman while his cavalry has started the relieving charge... :)

edit: crossposted with eric - responding to valamir

I like the general idea, and I like the specific mechanism. But as described it doesn't seem to match the characters' actions. It doesn't seem right that a change in the situation that causes a permanent increase in the danger level of the combat whould be caused by a single momentary action such as a thrust. Instead, we should narrate an escalation in the intensity of the combat overall. Something like:

"I (we?) dance around and feint, probing for weaknesses." (Danger 3)

"I (we?) switch to more aggressive strokes, swiping at limbs." (Danger now 4)

"I begin to parry and lunge in earnest." (Danger now 6)

I think it might work even better for monster combat, where the narration can progressively bring the monster's threats into play. The cinematic sequence would arise naturally, from "the monsters snarls and snaps while the swordsman keeps it at bay with threatening swordplay" to "the monster spins trying to bring its poison stinger to bear while the swordsman looks for an opening to its belly" to "the swordsman has leapt onto the monsters back and is trying to stab its skull while it tries to crush the swordsman against the wall."

There are several recent conflict systems (Trollbabe and Dogs in the Vineyard, for instance) based on raising the stakes. What you seem to be shooting for is a bit different, because the stakes start out high (someone's going to die, if neither backs down) from the beginning. That's cool. But it's also a problem, since raising the difficulty level affects only how quickly the combat will be over (and only at one's own expense) so the only basis for deciding whether and how much to raise appears to be how it would shift the odds. As currently described, the safest course short of backing down is to never "raise" (or, if the action is reversed as other posters have suggested so that when you raise, your opponent faces the higher risk first, to always raise as much as you can). One possibility for a refinement is that if you raise, your opponent faces the higher risk first, but the act of raising also forbids you from backing down on the next action. Another interesting possibility is that the players can only back down in stages the same way they escalated, which means that seeking short-term advantages can inadvertently (or advertently) lock you into a death struggle. Which would make this even more conceptually similar to the "gain extra dice by raising the stakes" systems already out there.

The probabilisitic behavior of Russian Roulette might be useful to know here. Assuming a single bullet in a six-round revolver, and the chamber (or whatever you call it, the revolving thing in a revolver) is spun (randomized) before every shot, the "player" who goes first has a 6/11 (about 55%) chance of losing, a significant advantage over the second player whose chance of losing is of course 5/11 (about 45%). The game has a 52% chance of being over within the first four pulls of the trigger, and a 77% chance within the first eight pulls.

If after the first spin and shot the mechanism is simply allowed to rotate to the next chamber, which is a bit more like how your combat system behaves if the difficulty is steadily increased each turn, then the Russian Roulette game becomes fair, with each "player" having a 50-50 chance of survival. Of course, the game is then guaranteed to end within six pulls. With randomization at each turn, the game will be fair if the oddds of dying on each successive turn are any steady progression of reciprocals (such as 1/12, then 1/11, then 1/10, and so on). However, designing dicing mechanisms that generate these probabilities is difficult unless you allow for a lot of rerolling.

- Walt

Firstly, yes I fully agree there is no reason at the moment to take any action whatsoever.  If you do nothing, your opponent will eventually kill themselves.  so this is definitely not a whole mechanism and I certainly posted it too early.

That said I'd like to take up the idea that there is no benefit to going first.  Firstly, this is not necessarily a problem, becuase the Riddle of Steel operates quite well with the Counter to an opponents attack being pretty much the optimum manouevre.  So that in itself does not bother me too much.  

Secondly, making an attack first does let you set the tone, and if there is a significant imbalance of force, you could conceivably outbid your opponent in the first move.  If your ability was such that your opening bid was so high that the opponent could not even bid 1 and succeed, you've won.

Tony remarks:

Quote
Connected... it seems a bit odd that somebody could kill themself with a bad roll, even absent the other character having done anything. For instance, if Joe fails his first roll than his opening thrust results in his immediate death. Yes, yes, I see how it can be narrated, but it still seems odd.

Indeed, it does seem rather odd.  What I was aiming for was that each exchange would be a flurry of activity rather than discrete blows.  That is, in the opening bid the "defender" is not a passive recipient but is actively trying to stab you back - that is why you can still be killed.  I would expect the narration to go something like: "I close in behind my shield, alternating blows over and sideways for a few strikes, and then duck under and cut to the legs"  Now the other combatant is still active, blocking, cutting back etc, they merely are not in control of the narration because they are not bidding.

This is where IIEE gets changed - intent, initiation and execution all happen with a single statement followed by an immediate roll.  Effect is the only step that requires a separate determination.

IME with martial arts, limited as it is, it is quite easy to attack stupidly, to walk into a blow, to attack and find your opponent side-steps and suddenly they are behind you.  Its the old martial arts dictum about not defeating yourself that I was working towards.

Valamir wrote:

QuoteWhat this means is that if you failed your roll, in order for them to roll and kill you they have to roll at an even higher difficulty than the one you failed. And if they fail, then they are vulnerable. That would mean the player would have a choice, to escalate and risk death or break off. Breaking off the combat would automatically succeed after any failed roll by the opponent. So the choice is concede/escape or press on.

Yes quite right.  "Killed" should stand as an abbreviation for "defeated/wounded/suffering" or something.  This was only meant as a test of concept; as I see it a more developed version would determine the extent of effect by the degree of failure, that is, number of points by which the difficulty is missed, or something orthodox like that.

As for disengaging, that too was ommitted.  I did not want  to introduce a two step process though so I thought an attempt to disengage could be treated as a bid of zero.  This means the would-be escapist does not accumulate any additional risk trying to escape, but must still act and roll, and thus can get it in the back if they fail.  

Technicrat 13 wrote:

QuoteOf course, this way there would have to be a limit on how much one could increase the difficulty by in a single bid, otherwise there wouldn't be any reason not to bid the maximum difficulty right away, in an effort to keep your opponent from bidding back on you.

Thats a feature, not a bug.  It allows mook-swatting to be smoothly integrated into a mechanism that handles conflict between equally matched opponents too.  If you have reason to believe you can instantly outbid your opponent and put them in an impossible position, you should.

Also, yes it does feel rather like russian roulette, thats an apt analogy.  But not a bad thing IMO.  I'm not sure it should really serve to discourage combat, but it is also true that I am not much interested in games that treat combat as their basic modus operandi.  So this is specifically a combat system rather than a resolution system in the broader sense.

But anyway, thank you all for your comments.  Tha main idea was to articulate a combat system that operated from a very different principle than task/blow resolution.  The individual exchanges are intended to be more like exchanges in a HW extended contest than a To Hit roll.

Idea: All the characters are in sudden, immediate danger, but it is not necessarily danger of death.

This opens up the possibility of having more than one sword at once, each of which will inflict a different injury, or to run conflicts where, as one sword falls, you replace it with another. You might have a whole economy of danger management where you can make one sword less likely to fall, but this will unbalance another.

I like this system!  What you just described could be a conflict resolution system for Matrix Games -- http://www.io.com/~hamster/rules.html

1) Why not simply have the winner a Pool style Monologue of Victory?  

2) A 2D6 is an awfully steep curve.  I would look into using a flat curve for purposes of this.  Perhaps a D12 or even a D20.

3) Thinking about Matrix Games above, I could see this system also being used as the basis for GM-less style play when combined with MoV.  You would need to have some other sort of resource/mechanic to be manipulating.  But it definitely has possibilities.

I seriously like this idea.  I think there is actually a very strong reason for going first: the first actor sets the opening bid, and thus the minimum chance that his opponent will die (it could be higher, depending on your opponent's bid).  You could definitely want that control in a game like this.  For example, if facing a mook, that mook might decide to bid the minimum opening up.  That means he's got a slim chance of getting clobbered.  While your chance is slimmer, the fact is that as a player you know that you will be making rolls like this until you blow one and die.  So you'd rather make fewer rolls.  

This all reminds me very much of the hand to hand combat system for "Conspiracy X."  Although that game was much more conventional, of course.  An attacker basically launched a combo of pre-selected moves (Like punch, punch and jump kick).  But you could also have moves that started with a block or a dodge (like block, punch, kick).  If you succeeded at a block, you went right into your counter attack.  Each kind of move had a different difficulty and different damage, of course.

What I want to know is what kind of behavior is statistically optimal?  Clearly, if you have any idea of your opponent's ability (and it is less than your own) you should bid an amount that is fairly safe for you but fairly dangerous for them.  If the opponent is more powerful, however, perhaps you should try to open with the highest bid that you can stand: maximizing your chance of knocking your foe out in one round.  It seems like if there is a great disparity in ability, then the stronger will virtually always win.

Have you considered letting multiple characters pool their dice together?  I.e. I have 2d6 and my sidekick has 1d10, so we gang up on the boss villain's 3d6 with a decent chance of winning?  Or would multiple players just force the opponent to get into more than one bidding war at once?  That would certainly make things easy on the guy with the best stats.  Do you think it should be more like poker?  With every player upping the wager?

This system is screaming for some kind of fate points to make up for a botched roll or bid down a difficulty.  That would seriously alleviate some of the harsher consequences.