News:

Forum changes: Editing of posts has been turned off until further notice.

Main Menu

need to reduce handling time of my combat system

Started by David Berg, December 18, 2007, 06:11:54 AM

Previous topic - Next topic

David Berg

Too many calculations, die rolls, and chart-lookups!  Help!

Desired Inputs:
Player's preference of where to move to (if applicable) + character's ability to maneuver as dictated by surroundings.
Player's preference of where to hit opponent (only relevant if hitting).
Player's preference of offense vs defense.
Character's skill in weapon.
Character's Agility attribute.
Character's Toughness attribute (only relevant if damaged).
Character's current level of Shock (only relevant if damaged).
Character's current physical wounds (only relevant if damaged).
What kind of armor the character is wearing (including shields) on what body parts (only relevant if hit).
Character's weapon type (only relevant if hitting).
All of the above for the opponent.

Desired Outputs:
It hit you or missed you.
It hit you in location A (only relevant if hitting).
It did/didn't get through your armor (only relevant if hitting).
It did B damage to you (only relevant if dealing damage).
The damage has/hasn't destroyed location A (only relevant if dealing damage).
You take C amount of Shock (only relevant if dealing damage).
The damage has/hasn't dropped you (only relevant if taking shock).
It now is/isn't where it wanted to move to.
All of the above for your attack & movement.

Current Mechanics:

PRE-COMBAT

GM says, "Next round."
PCs declare actions
   lowest AGL declares first
Non-combat movement is resolved and considered to be in progress
   highest AGL acts first
Missile attacks are resolved
Combat maneuvers are resolved, dictating what positions will be at end of round; change of positions is considered to be in progress


COMBAT

(GM figures out action order)
   longest weapons go first, ties act simultaneously

players specify where "location 7" hits will land by placing 1d8 (color 1)
players specify where "location 8" hits will land by placing 1d8 (color 2)

(GM calculates PCs' and enemies' target numbers)
   these are based on weapon skill, AGL, offense/defense preference, shields, + any positional/circumstantial modifiers for each attacker and defender per attack

players roll to hit on 2d6 (color 1)
players with adds* to hit, roll 1d6 (color 2)
players with adds to defend, roll 1d6 (color 3)
players roll for hit location on 1d8 (color 3)

(GM rolls for enemies to hit
if necessary, GM:
   rolls enemy adds
   rolls enemy location dice
   chooses "7" and "8" locations
GM determines which enemies hit, where, and for how much damage
GM looks at players' dice and determines who hit, where, and for how much damage)

GM announces who hit and who got hit, and where

players tells GM whether they were hit on armored or unarmored locations**

GM tells players hit on unarmored locations how much damage they take

GM tells players hit on armored locations their enemies' weapon type

players hit on armored locations consult the "weapon type vs armor type" chart and roll armor on the appropriate number of d6

(GM rolls armor for enemies who got hit on armored locations
GM determines how much damage enemies receive & whether this destroys any body parts
GM looks at player armor rolls and determines how much damage PCs receive)

GM announces whose hits on enemies got past armor, and roughly how deeply

GM announces how much damage armored players take, players determine if this destroys any body parts

players make TUF rolls to soak SHK, then determine if this renders them unconscious or dead

(GM makes enemies' TUF rolls to soak SHK, then determine if this renders them unconscious or dead)


POST-COMBAT

Non-combat non-movement actions are resolved
   highest AGL goes first

Locations of moving and maneuvering characters are resolved
   highest AGL goes first

*minute differences in effectiveness are represented by extra dice called "adds"
**sometimes rolling an extra die to determine this, in odd cases like where "Left Leg" includes an armored shin and unarmored thigh.
here's my blog, discussing Delve, my game in development

Narf the Mouse

First, what is the combat in your game about?

Second, drop anything that isn't important to and doesn't relate to that.

D&D, for example, is only interested in wether you hit and how much damage you do. So, Attack + d20 versus AC, then NdX damage roll. (I'm ignoring feats. They are kinda badly designed, anyway)

David Berg

Combat in my game is about two things:
1) resolving "what happens" in a way that takes into accout all the relevant factors from real-world combat, including those things a combatant can control.  Thus my Desired Inputs.
2) resolving "what happens" in a way that describes what the outcome looks like within the gameworld in terms of all the relevant factors.  Thus my Desired Outputs.

Oh, and combat in my game is also about not dying quite as quickly as you would in the real world, thus the Shock mechanic.

I don't intend to re-evaluate my inputs and outputs here, I just want a better method of getting from one to the other.
here's my blog, discussing Delve, my game in development

Callan S.

Hi David,

I think handling time is an issue of how much you deal with something that's fun. Lets say a game was about furry fighters. I don't like furries and you probably don't either - that's why I'm using it as an example, so we can look at fun from an outside perspective.

Lets say there's two resolution methods:
A: a single attack takes three minutes to resolve, but every chart look up and dice roll is soaked in furry stuff - all that furry crap is just all over the charts.
B: a single attack takes twenty seconds to resolve, but has no furry fighter references at all - it's just cold, plain numbers.

That twenty seconds, although it's faster, sucks. It's like an ratio - a fun over time ratio. B has a very poor fun over time ratio. While A has a very good ratio.

Really you can't improve the ratio until you know what the fun thing about your game is. Clearly it's not furries - it might be one main idea, or several related ideas. I think that's what Narf was asking - what's the fun thing about your game? Is the fun finding out "what happens"? If it is, you might have a very good ratio already!
Philosopher Gamer
<meaning></meaning>

David Berg

Ah, okay, I see how this looks nonsensical.  "My combat system is about determining what happens, and it does this perfectly, and the perfection is not enjoyable!" 

In isolation, indeed, I am quite happy with the combat system.  What I am not happy with is what happens when I use this combat system in my game which strives for a sense of "feeling like you're really there".  Too much mechanics-consulting per unit of established imaginary content breaks that feeling for me. 

"His club is a bashing weapon, my chest is covered with plate armor.  Plate armor vs bashing weapon equals 2d6.  Roll!  Okay, my armor took 3 points of the blow's damage," is just too much mental energy directed at categories, charts and dice for something that occupies a millisecond of in-game time.
here's my blog, discussing Delve, my game in development

David Berg

To clarify: if I had to choose between "factors in armor & weapon types" and "requires one less chart consultation and die roll", I'd choose the former.  Which is why my combat system is the way it is right now.  What I'm hoping, however, is to have my cake and eat it too.  I want some analog of "throwing my desired inputs into a supercomputer and having it instantly spit back the desired outputs!" to replace my "roll, roll, roll again".
here's my blog, discussing Delve, my game in development

jag

Quote from: David Berg on December 18, 2007, 08:25:25 AM
Combat in my game is about two things:
1) resolving "what happens" in a way that takes into accout all the relevant factors from real-world combat, including those things a combatant can control.  Thus my Desired Inputs.
2) resolving "what happens" in a way that describes what the outcome looks like within the gameworld in terms of all the relevant factors.  Thus my Desired Outputs.

I don't intend to re-evaluate my inputs and outputs here, I just want a better method of getting from one to the other.

I think you are setting yourself an impossible task.  Your thing 1 says you want to take into account all the relevant factors from real-world combat.  That's impossible.  Your list of 10 things, while extensive, is only a tiny tiny fraction of 'all the relevant factors'.  To name a couple things you've missed off the top of my head: fatigue (aerobic, muscular (differentiated by muscle group, of course), energy level), morale, motivation, the 'momentum' of battle, footing, etc).  All rpg combat is an incredibly stylized abstraction, and the only question is which factors do you consider important.

With that in mind, are all those factors you wrote down really important?  If all the inputs and outputs are important, you are asking for a quick way to convert 10*(number of participants) inputs into 8*(number of participants) outputs -- to which i can only respond a computer.

Lacking a computer (which may be an unsatisfying mechanic anyway), you need to decide if accounting for everything is worth the search-and-compute time.  My guess from the title of this thread it isn't.  So the question becomes which factors can you combine or drop.

One particular bugbear of mine is "player's pref of where to hit opponent".  It see this a lot in very crunchy mechanics.  Now, everyone's version of 'realism' is different, but in my version you rarely get to choose where to hit the opponent (unless you completely outclass them).  You take what openings you get.  At best you can feint/make a primary attack in one place, with the expectation/hope that the opponent will drop their guard where you want them to.  If hit locations are important to you, you could still streamline the system by making them semi-random (perhaps influenceable by expending some resource).

Similary, the following 5 results:

It did/didn't get through your armor (only relevant if hitting).
It did B damage to you (only relevant if dealing damage).
The damage has/hasn't destroyed location A (only relevant if dealing damage).
You take C amount of Shock (only relevant if dealing damage).
The damage has/hasn't dropped you (only relevant if taking shock).

are all highly correlated, to say the least.  They also deal with damage, which imo is one of the greatest abstractions in a combat system.  Thus maybe you could find a smaller subset that still captures the essence of what you're looking for.

I know you said you don't intend to re-evaluate your inputs or outputs, but that's exactly what i'm suggesting.  I do think that's the root of the problem -- converting 10n inputs into 8n outputs is going to be very crunchy, no matter how you slice it.  Of course, maybe all of those factors are worth the time spent, especially if the time spent is fun.

james



J Tolson

Quote from: David Berg on December 19, 2007, 02:39:00 AM
Ah, okay, I see how this looks nonsensical.  "My combat system is about determining what happens, and it does this perfectly, and the perfection is not enjoyable!"

The problem is that your system is only partially realist but you want it to be fully realistic. Realism in an RPG is a sliding scale between mechanical realism and time realism; it isn't possible for a single game to be at the opposite sides of the scale at one time.

To extrapolate: On the Mechanical Realism side, throwing a stone involves gravity, wind speed and direction, the shape of the stone, the height of the individual throwing, the force of the throw itself, the point of release, the method in which it was thrown, obstacles, etc. On the Time Realism side, throwing a stone takes less than a second. The more mechanics that have to be resolved, the longer it takes and the less realistic the game is in terms of Time Realism. Conversely, the quicker the action is resolved the less mechanically realistic it is.

Your game is heavily on the Mechanical Realism side of the scale; if that is what you want, that is fine, but the only way to make it quicker is to simplify the mechanics, thus making it less mechanically real.

Still, there are a few things you can do to help things go quicker. Roll fewer dice against fewer numbers, for one. Player 1 rolls to attack and the damage, then, is how much that roll exceeds the defensive number, for example.

dindenver

Hi!
  I had a similar snafu in my game. I decided to remove a lot of details (like hit location) and combine as much as possible into one roll. You could do something similar, like this maybe:
Roll + Agility + Skill + + Weapon Modifier + Move Modifier + Hit Location Modifier + Offensive Effort Modifier - Attacker Shock/Wound Modifier - Attacker's Armor Penalty - Target Agility - Target Skill - Target Toughness - Target Movement Modifier + Target's Shock/Wound Modifier - Target's Armor Defense Bonus +/- Conditional Modifiers vs Target Number. Where anything over the TN is Damage that got through to the hit location specified...

Or You could have both the attacker and defender roll something like:
Roll + Agility + Skill + Weapon Modifier + Move Modifier + Hit Location Modifier + Offensive Effort Modifier - Attacker Shock/Wound Modifier - Attacker's Armor Penalty +/- Conditional Modifiers
vs
Roll + Target Agility - Target Skill - Target Toughness - Target Movement Modifier + Target's Shock/Wound Modifier - Target's Armor Defense Bonus +/- Conditional Modifiers vs Target Number.
  Where every point the attacker beats the defender by equals damage that got through to the hit location.

  If you want to differentiate between wounds/shock/special effects, have the attacker divide the points into those three categories based on a location or player preference. Like Shock cost one success point (points that you made the roll by), Wounds cost 3 and special effects cost depending on a table, maybe temporary blinding costs 6 permanent blinding costs 12, etc... Or you could just have the success number subtract directly from the target's Agility and/or toughness. That way Wounds, shock and hit points are cleverly combined into one score...

  The advantage of a system like this is that a lot of info can be pre-recorded on the char sheet and there are no wasted die rolls (you could put a slot next to each line for their weapons where it could have their attack total with and without armor, and a similar slot on the armor line of the char sheet for their defense roll).
  Also, it defeats the classic "d20 gotcha" where you roll a 19 to hit (a real good roll) but a 1 for damage (a real bad roll) that creates a sort of anti climactic result...

  Well, that is a long post. A lot of info. And I think I might have used some short hand for some of my ideas. So, if you have any questions or ideas how to tweak it, lemme know because I did a good solid year and a half of playtesting on a similar system (my own) and can give you some pointers...
  Good luck man!
Dave M
Author of Legends of Lanasia RPG (Still in beta)
My blog
Free Demo

contracycle

I think you can cut the damage roll and inflict fixed damage numbers, as you have a soak roll that comes later to provide some variety.  In fact the randomness of the damage and the soak together might cancel each other and produce fairly constricted results.
Impeach the bomber boys:
www.impeachblair.org
www.impeachbush.org

"He who loves practice without theory is like the sailor who boards ship without a rudder and compass and never knows where he may cast."
- Leonardo da Vinci

David Berg

Everyone, thanks for your comments!  I think we have a pretty well-shared idea of what's going on here.  Sadly, I cannot implement some of the suggestions I've received so far, for the following reasons:

I already do have damage determined by the to-hit roll.  My "current mechanics" list says "roll" wherever there's a roll.  For damage, it says, "determine" -- this reflects that the determination is based on a prior roll.  This may seem like a weird way to write it out, but I've chopped the list into the tiniest possible pieces, in order to illustrate not just the physical processes involved, but also the mental ones.

I do prize Mechanical Realism over Time Realism.  But like I said, I'm trying to have my cake and eat it too.  Anything I can do to improve Time Realism (even if I never achieve a super-high degree of it) would be great.

As for deciding what parts of "realism" are relevant, I want to model every aspect of the gameworld reality that reflects player decisions and achievements. 
  • Achievements: if I stole a left greave off a dead goblin, there should be a chance that at some point I'll get hit in it and it'll save my leg.
  • Decisions: if I try to circle around my opponent so that his back is to the pit of lava, I should be able to resolve whether or not this works.
(As for specifically "where I hit", I largely agree with James -- usually, you take what opening you get.  Thus, a d8 location roll has locations 1-6 predetermined, regardless of player preference.  Only a 7 or an 8 hits the locations you specifically targeted.  In practice, I find that getting to pick where you want to hit enemies is fun, even if you can't always actually do it.)

On to some specifics:
James, I see your point with the 5 parts you singled out being highly related, but I'm not sure how to use a smaller subset.  Getting your chest armor hit is not the same thing as getting your chest hit is not the same thing as having your chest destroyed (i.e., you dead) is not the same thing as dropping from accumulated injury (which is slightly unrealistic, but fun: you can sometimes lose a fight without losing a character).  Can you think of a more efficient mechanic that doesn't ditch any of these distinctions?

Dave, good point about pre-recording.  I'd forgotten my earlier intent to mark on character sheets how each piece of their armor behaves against all weapon types.  Opponents' weapon and armor types and characters' weapon types would go on the GM's stat list alongside the derived O & D values for every combatant.

As for letting attackers pick which of their damage points go where, that's simply too meta for me -- it doesn't correlate with anything I could be thinking if I'm standing here swinging a sword at you.
here's my blog, discussing Delve, my game in development

Callan S.

Quote from: David Berg on December 19, 2007, 02:39:00 AM
In isolation, indeed, I am quite happy with the combat system.  What I am not happy with is what happens when I use this combat system in my game which strives for a sense of "feeling like you're really there".  Too much mechanics-consulting per unit of established imaginary content breaks that feeling for me.
Your gunna hate me for asking lateral questions, but never mind what breaks "feeling like you're really there". What actually establishes "feeling like you're really there" to begin with? What sets that up? How'd you get that feeling in the first place?

I think what sets that up should also be the resolution process to use here.
Philosopher Gamer
<meaning></meaning>

David Berg

Quote from: Callan S. on December 20, 2007, 01:00:03 AM
What actually establishes "feeling like you're really there" to begin with?
No, that's a great question.  I have a feeling, though, that the answers that come to mind are of the wrong class -- i.e., not covering it in a way that can be used for resolution.  Can you whittle down the "what" in "what actually establishes" to anything more specific?

First answers that come to mind:

  • GM describing my character's sensory environment in detail
  • having something interesting to do in the imaginary environment; this encourages me to investigate the environment further
  • the GM answering all questions as if the answers are gameworld fact (as opposed to GM ad-lib)
  • my fellow players playing characters that make sense within the setting
  • my fellow players playing their characters consistently (as opposed to occasionally dropping some trait when it's inconvenient, thus reminding me "this is my buddy here")

The first point gets me most of the way to "really there"; the second gets me the rest of the way; and after that it's largely just a long-ass list of "dont's" in order to allow me to remain "really there" (provided Point #2 persists).
here's my blog, discussing Delve, my game in development

Paul T

Aside from the great suggestions above, I would recommend you check out any "simulation"-oriented systems written by Greg Porter (of BTRC). CORPS and EABA are two good examples. It's amazing how many "inputs" he can cram into a very very quick simple mechanic. You might find some inspiration if you don't find a better approach through this conversation here.

David Berg

Paul,
Thanks.  I checked out the free "lite" version of EABA and it does seem very efficient in a few ways:
  • any time a specific number (2 points of damage taken) is correlated with a specific other number (-1 to AGL), that relationship is already written on the character sheet
  • damage is tracked in boxes on the character sheet, with a slash meaning "non-lethal" and an X meaning "lethal"
  • when one input can produce 2 outputs, that is done, rather than using 2 inputs (i.e. one damaging attack produces two rolls for the two damage types)
Not having played it, though, nothing else caught my notice...  I'm sure this system would be faster than mine, but it also produces less specific outcomes (i.e., no mention of hit location), and CORPS seems to do likewise.  Can you point out anything to me that I'm missing?
here's my blog, discussing Delve, my game in development