Initiative System

[giblin]

Hi all,

Any ideas on how to make this more streamlined and playable?  I always hated regular initiative systems where one person runs across the room and attacks while time freezes for everyone else until its their turn.  I have been attempting to make a time-phased or action ordering system but it seems from limited playtesting that it slows down the game too much.  Here's the guts-

A number, called Delay is calculated after every action and added to your Time which is when your next action goes.  Baically, 1 Time Unit =  1/2 a second.  The Delay is calculated as 13 + action cost - (dex+D6).  

Reasoning for the formula for calculating Delay(some assumptions);

-A combat round is 3 seconds or 6 time units hence the d6.
-Normally, 1 action per round so using the d6 determines when in the round the action goes.
-Movement is done on a incremental basis with the referee calling out 1, 2, 3...etc.  All combatants may move a number of squares for that phase(the number called out loud by the referee) based on their speed and is listed on the character sheet like-

P M
1 2
2 1
3 2
4 1
5 2
6 1

Where column P is the phase count and M is the number of movement points available for spending on movement for the phase.

To simplify for understanding, say you roll a d6 to determine when your non-movement action takes place(ie: next attack).  Then you get to spend points each phase on movement.  Thus you no longer have to "stand still" while someone crosses 60 feet of ground and attacks you or pulls the lever...ect.

If your un-engaged in melee then you attack when you move close enough to strike your target(ie: 2" for a spear)-the phase you move into range is when your attack happens.

It also enriches the vision of the scene..."Tor: I draw my sword and it will be out on phase 3.  Ork-charges Tor with ready spear and is 4 squares away = phase 1 ork moves 2 squares, Tor backpeddles 1 distance is 3 squares...phase 2 ork moves 1square, tor up against wall so distance is 2 so the ork attacks Tor before Tor can pull his blade out to help defend himself with it.  Or the "Do I come to the rescue in time?" question or do I pull the lever to let the gate drop down in time?  Or even, do I run out and slam the door before the dragon breathes on me?

One thing I thought of as a derivitive is to use "normalized numbers".  Normalizing to setting everything to decimal(in this system).  EX: it takes 6 aciton points to attack and I have a speed of 9 so my initiative is 6/9 or .67.   Urf the Ogre has 6 and his attack costs him 5 or 5/6 which is .83 therefore I attack before Urf.  Movement would be coordinated the same way.  I have a movement of 5 then each square I move costs me .20 initiative points...

Im writing this at 12:30 so I may have written this well.

What do you think or does it need clairification?

Thanks all,
Dave

[Mark D. Eddy]

Ok, I actually had a diatribe written up, and then I reread your first sentence. There are a few things that I would suggest to make this system more usable:

1) Enforce timing. Call out numbers in three second intervals,and if the player doesn't respond in their turn, keep going. This will help stop the Starfleet Battles problem of (less than) one-second impulses taking five minutes to resolve. You can make exceptions to this for self-defense, unless you really want to use Paranoia's "Dramatic Combat  System."

2) Precalaculate as much as possible. Before you drop into combat time, know all of those decimals. PC's in this sort of game probably will anyway - put it on the character sheet, if nothing else.

3) Only drop into impulse-driven combat time when it's absolutely necessary.

[Ron Edwards]

Hi Giblin,

Welcome to the Forge.

"Initiative" is one of those topics that gets ahead of itself in game design. Many RPGs have worked well without it, or at least without a special roll or other determinant.

You might be interested in this material, which I mentioned in my essay and then got developed in this thread:
The four steps of action

It's been discussed further elsewhere as well.

Best,
Ron

[Le Joueur]

Hey Dave,

Any ideas on how to make this more streamlined and playable?

Well, the short answer would be to get rid of the initiative system completely.  I put the long answer here.

I always hated regular initiative systems where one person runs across the room and attacks while time freezes for everyone else until its their turn.  I have been attempting to make a time-phased or action ordering system but it seems from limited playtesting that it slows down the game too much.

...calculated after every action and added...

...is done on a incremental basis

...calling out 1, 2, 3...

...points available for spending...

To simplify for understanding, say you roll a d6...

...you get to spend points each phase...

It also enriches the vision of the scene...

With that many steps, I can see why this slows things down.  Calculate and add, go by increments, call them out, spend points (or not), all after rolling a die?  It certainly parses out the sequence of combat quite clearly.  The problem is, the more you 'chop things into little managable pieces,' the more there is to deal with and the longer it's going to take.  That's unavoidable.

As far as the "time freezes" effect, that is usually the result of two things; the detail of application of the resolution system and the simple act of taking turns.  There's no simple answer to how much detail (or how to handle that) you should put in a action-resolution system; everybody has a different opinion.  One thing never discussed is the act of taking turns.

Superficially, there doesn't seem to be any other way to handle things other than taking turns.  Until you get to Wall Street or real combat.  I remember a cute little Parker Bros. game called Pit; it was an fast, extremely abastract, card-based, commodities trading game...with a bell.  Everyone just started talking at once trying to trade the cards they didn't need for the commodity they chose; when you corner the market on anything, you ring the bell and win.

I was just thinking about the 'live-action' component to table-top gaming (unlike 'net games, you've got real human beings right there).  It is possible to simply have everyone handle their own combat.  Provided that the game mechanics could be operated by anyone (without the need for a referee or gamemaster) individuals could self-select whom to battle and engage that person in conversation/melee.  You might have some kind of time keeper to call out the rounds on a clock-based schedule (if you haven't resolved it by 'next call' it fails).  You'd need representatives for each combatant player or non-player, but otherwise the whole thing would resolve rather quickly.

That's the only thing I can think of to eliminate 'taking turns' and the inherent "time freeze" it includes.

One thing I thought of as a derivitive is to use "normalized numbers".

More calculations?  I thought you wanted "streamlined and playable" combat.

I agree with Ron.  Could you express why you believe you need all the technical detail?  (I usually assume it's for a rigid 'fairness' aspect, but you might think otherwise.)  Your stated conflict is between mechanical detail and "streamlined and playable" play; can you say which is more important and why?  What about the reasons for not 'giving up' on the other side (if you choose "mechanical detail" as more important, what is the reason you don't jettison "streamlined and playable")?

Only when you clarify your priorities can anyone comment on how you might acheive them.  I look forward to this discussion.  Good luck!

Fang Langford

[Matt Gwinn]

Back a few years ago I was working on a game called The Keys of Solomon which had a similar system for combat.  In KOS each round consisted of 60 seconds which was then broken down into action phases.  Each Phase filled a varaiable time frame based on the character's initiative roll.  For example, one character may get 5 action phases per round in which case each phase would be 12 seconds.  Another, faster, character might get 20 Action Phases of 3 seconds each.  All the action would take place at the same time, jut at different speeds.  Actually working through a combat round was very difficult to explain in writing which eventually led to the system dieing, but your idea kind of reminded me of it.

I'm not sure if that answers you question, but maybe it will give you something to think about.

you can find the rules here
http://www.angelfire.com/games3/errantknight/keys.html

,Matt G.

[Walt Freitag]

Hi Dave, and welcome.

Your mechanism reminds me of the way most computer simulations work: advance time in small increments, and for each component of the system (in this case, each character), determine what happens during each increment. Of course, computers are really good at updating multiple calcuations quicky. As for people, that isn't usually their strong point.

If you've decided that you want combat resolution to work this way, then your problem isn't so much game design as it is efficient execution of the system in play. You have to think like an industrial time-and-motion efficiency expert.

To start with, settle on one single time unit and stick with it. If intervals between actions are specified in action points, and characters have different numbers of action points per round based on their speed, and a round is so many time ticks, convert that all to ticks (rounded to integers) beforehand. During the combat, the players must have in hand how long each action takes in ticks. If there's a die roll involved, then the base number of ticks without the die roll should be known beforehand, and the die roll should add ticks to the time, otherwise new calculations would have to be done every action.

Next, make all ticks equal. Since you know how long each action takes in ticks, there's no need to count rounds, and there's no need to count which tick you're at within a round. Get rid of that move 2 on phase 1, move 1 on phase 2, etc. thing for movement. Instead, determine in advance a movement per tick for each character, and allow that to be a fraction. You're using miniatures (or else this would all be quite unmanageable), so just do the movement according to the scale distance per tick. Physical templates, like a popsicle stick marked or cut to the length of a character's scale movement per tick, can help. (Avoid using rulers; they're too long and too hard to read quickly.)

By the way, make sure all the miniatures are on firm bases that stand easily. Having to delicately re-balance a miniature on its teeny tiny molded-in pewter base every time someone touches it will slow things down intolerably. Also, make sure everyone, especially you, can easily reach all parts of the field. You probably can't do that sitting behind a screen. Stand up (or even use a separate miniatures table of standing height) if necessary. Oops, can't see your notes or the tables on the screen while you're standing so you can reach the miniatures? That's a problem. Solve it.

Now, to speed up the movement handling, give each character a destination marker. (Color-coding to the miniatures' bases can be helpful). These are placed on the miniatures playing field. Characters move toward their marked destinations unless blocked. The destination mark cannot be changed during a tick. It can be changed freely between ticks. This cuts down on indecision delays during a tick, by making the movement during the tick truly simultaneous. (A character's destination mark can be placed on another character's base, indicating that the character intends to close in on or pursue that character.)

Poker chips. The object is to cut down on the need to store and recall information, especially numbers, in short-term memory during play. When a character is required to spend some number of ticks in recovery, during which movement is disallowed or reduced, plunk down a pile of that many chips next to the miniature (or, if that space is too crowded, in front of the player). That character's player will remove one chip per tick until the recovery is done. Same procedure if you're counting up prep time, such as when a character has to stand in place for some number of ticks until a spell or fired missile goes off.

The goal here is that on every tick except those during which an action such as an attack is being resolved, each player should be able to resolve their movement or waiting or preparation actions in three or four seconds of real time, all at once. The GM, when controlling many characters, will take longer, but if he uses destination markers or otherwise clearly states what the NPCs are doing ("they all charge directly toward you") then the players can help with this task.

Resolution of actions involving success rolls, effects rolls, determination of the recovery time (and placing the chips), and so forth will take longer, of course. But these are the procedures you'd be doing one at a time in any case, and for this style of play they're the interesting part, so that shouldn't be a problem.

Finally, practice practice practice.

This isn't my favorite way to play, but if you're going to do it, concentrate on making the execution of the system a model of efficiency that even the computer would envy.

- Walt

[Mike Holmes]

First, see TROS.

The other solution to Fang's dilemma of taking turns is to have everyone go at the same time. This is a much better way of simulating what happens in combat than any turn based resolution. One really positive effect is that it allows the idea of skill to be put into initiative, which is by far the most important thing in RL initiative.

How do you do this? Simply make all combat exchanges a contested roll between the combatants in question. Winner gets his result. This goes for movement as well. Want to beat someone to the gate? Roll their running vs yours. This is cool because it introduces the RL element of randomness to movement. And nobody "whiffs", they just aren't always the one to hit. Something is always happening.

The only concern is the rate of progress of combats that are outside of each other (I try to finish Bob quick so that I can also join the race to the gate). But there are subtle ways around that as well. Make the player rushing to finish take some number of penalties to do so based on the running player's successes modified reasonable for the distances in question. This is a great way to make combats dramatic.

It's all in the currency. See Zenobia for a free example of this solution in action.

Mike

[Le Joueur]

The other solution to Fang's dilemma of taking turns is to have everyone go at the same time. This is a much better way of simulating what happens in combat than any turn-based resolution.

Aaahh..."Fang's Dilemma."  It kinda has a 'ring' to it; don't it?

Anyway.

The 'everyone handles their own rules' example was meant to be exactly an 'everyone goes simultaneously' practice.  Simply having everyone go at the same time becomes cacophonous in anything but the smallest groups, so you'll have to deal with that.

As far as "Fang's Dilemma," I think it should be for things like, "It can't be realistic because in real [combat] everyone doesn't take turns."  (Change the bracketed items to any game mechanics.)

Ha!

Fang Langford

[Mike Holmes]

The 'everyone handles their own rules' example was meant to be exactly an 'everyone goes simultaneously' practice.  Simply having everyone go at the same time becomes cacophonous in anything but the smallest groups, so you'll have to deal with that.

The difference is that I am not advocating getting rid of turns. You still do turns, and each character only gets one declared intent in each turn. You just roll everything simultaneously, and apply the results. It's just as ordered as normal, I go - you go play, you just avoid the problematic cases of running by people and trading blows and whatnot.

That trading blows thing is a biggie. Using simultaneous resolution, you must always remain tactically sound. Trading turns leads to things like the Rolemaster first round berzerk strategy. Essentially, if I know I am going first, I'll ignore defense and put everything into attack. Given that my attack is now relatively certain to take yours away from you (stun), I'll win most every time.

Randomizing the initiative to avoid this just begs the question. As soon as you play simultaneous (or use something like TROS) you find that you have to constantly re-evaluate your strategy as things change. Which is fun for the whole family.

Mike

[Le Joueur]

The difference is that I am not advocating getting rid of turns. You still do turns, and each character only gets one declared intent in each turn. You just roll everything simultaneously, and apply the results.

Ya threw me there Mike.  When you said "have everyone go at the same time," I took it literally.  And just for a bit more clarification, I was tossing turns but keeping rounds.  I don't know what would happen if both went.

I wonder what would happen if you had in-game time set to simultaneous (and out-of-game time in turns) and used something like Godlike's heightxwidth dice mechanic.  Sure both go at the same time, but only one usually hits first.

Well, anyway, I'm just talking out of my hat.  I'm a big believer in quick, turn-based combat with a really simple die roll mechanism over two or three layers of application.  (One layer is 'tactical soundness' as resource management; another is whose actions leave them with an advantageous position; a third could be flurry-of-actions as resource management.  Otherwise it's essentially 'beat me on 2d10,' reg'lar Scattershot.  Y'know, I never looked at it like that before...thanks for the stimulus.)

Fang Langford

[giblin]

First, see TROS.

The other solution to Fang's dilemma of taking turns is to have everyone go at the same time. This is a much better way of simulating what happens in combat than any turn based resolution. One really positive effect is that it allows the idea of skill to be put into initiative, which is by far the most important thing in RL initiative.

How do you do this? Simply make all combat exchanges a contested roll between the combatants in question. Winner gets his result. This goes for movement as well. Want to beat someone to the gate? Roll their running vs yours. This is cool because it introduces the RL element of randomness to movement. And nobody "whiffs", they just aren't always the one to hit. Something is always happening.

Thanks Mike I took your advice and checked out TROS quickstart Realt-Time Combat article but I don't think they explained movement.  Is it everyone moves(or the roll-off idea, which I like btw) and they all combats take place?

How would you, in TROS, a dragon breathing fire and you want to see if you make it thru the door and close it before he nails you?  Would it be like the roll-off example with breath fire skill vs running I take it?  Same would go for any missle fire right?  This works great when there's only 1 on each side but when you need to know how far or where every combatant is at when something happens this system isn't a solution...unless there's some way to normalize the initiative(skill) rolls in timing contests??

Dave

[giblin]

2) Precalaculate as much as possible. Before you drop into combat time, know all of those decimals. PC's in this sort of game probably will anyway - put it on the character sheet, if nothing else.

Yeah, I was thinking something like this for movement after someone posted getting rid of how many AP's you have to spend per impulse to decimal format.  IE:  a Movement of 9 in 6 impules = 1.5 where the decimal is rolled each move to see if you make the additional square of movemtent(extra-rolling though) or you keep trak of impules on paper and do the old D&D thing IE: 3 per 2(like the old fighter attacks).  In other words on even numbered rounds you get 2 move points.

The other thing you'd do when moving is divide your SPEED number by the cost to enter a new square.  IE:  1.5 is my speed and it takes 4 to enter the terrain in the sqaure I want to move to so thats 3/8 chance I make it.

I think this could then apply to timing of everything then where each "phase" you roll to see if you've completed your task based on the cost.  IE: if attacking cost 6(assume 1 attack per 3 second round with setup time for the attack) and I have 9 for attacking speed(divide by 6 phases) = 1.5.  Then 1.5/6 or 25% chance per phase of getting to attack.  This works out to the expected value of 4 phases for every attack but allows for realistic delays(he's not open, he's not open..Nows my chance!) and for realistic bonus attacks for going faster than normal-I catch his gaurd down, I strike, he parry's but is off balance and I capitalize on his stance/arms being out of position and get to make a follow-up attack the very next round(I succeeded in rolling initiative 2 "phases" in a row).

I know the obvious objection to this type of phasing system is people will be put off by more rolling. "I have to roll every round to see if I act and then I have to roll to see if I succeeded!, geeze!"

But on the positive side, you take out record-keeping and it's faster.  Also, you could look at the additional rolling as being an enhancement to the game.  How often do players not pay attention and are talking about something until it's their turn to go, roll the dice:

DM:  Bob your go.
BOB: {Rolls}I missed.  So ya, I told my girlfiend...

In other words this system is really a feature that keeps players focused on the game since they don't know when they will go they will have to pay constant attention to what's going on as they may have to change their tactics-"Oh crap, looks like I'll be in range of the Ogre's club next turn if he makes his initiative roll so do I keep trying to reload my bow/crossbow or drop it and pick up shield and sword.

I think a certain beat and excitement would develop.  I think proper explanation and example would be needed to show people who complain of ROLL playing vs ROLE playing this is good, not bad.

What do you think or am I off track?  I guess like any theory, the prove is in the playtest.

Gib

[Mark D. Eddy]

I'm not sure if this is a good or bad thing in your mind, but it's beginning  to sound like you're building a wargame, not a role-playing game. If so, there are several wargames out there that you may want to look at and see how they handle the questions you're asking. If not, you may want to step back and see if you really want/need to have this much detail in your game.

Does anyone else remember a game called "Phoenix Command," or something like that? IIRC, it was a role-playing game that went to an incredible level of realistic probability, including chances for death by hydrostatic shock for various weapons at various ranges in various hit locations.

[Christoffer Lernö]

Mark, I have a friend who loves Phoenix Command. He and his groups are playing all sorts of different campaigns with that combat system.
Personally I feel having a root canal would be more fun, but people are different. Anyway, I feel it's a nice call to mention that game.

Dave, are you familiar with Phoenix Command? If not it might be worth looking into. It's out of print but I'm sure someone here could lay out the basic rules for you.

(Incidentally my friend swears it doesn't need to be very slow. Of course they are used to quick scirmishes in Rolemaster which takes *groan* around 2 hour on average - I know from experience)

[Lance D. Allen]

How would you, in TROS, a dragon breathing fire and you want to see if you make it thru the door and close it before he nails you? Would it be like the roll-off example with breath fire skill vs running I take it? Same would go for any missle fire right? This works great when there's only 1 on each side but when you need to know how far or where every combatant is at when something happens this system isn't a solution...unless there's some way to normalize the initiative(skill) rolls in timing contests??

I think you may have gotten the wrong idea of TRoS, Dave. Initiative is only rolled in the case of two or more opponents squaring off with intent to injure or maim each other. The initiative roll has nothing to do with skills or any trait on the character sheet at all, and when you roll, you don't even bother with looking at the number. All that the die is for is to announce intent simultaneously.. attack or defense. When we duel online, we handle initiative by simply announce "red" or "white".

In the case you described, you would never even drop initiative. You would figure out the distance, compare movement rates, determine the distance the dragon needs to be from you to hit you with flame, and then just go. I'll give an example..

GM: The dragon is 300 yards away from you, and it's flame is jetting out about 10 feet. It's coming in fast (sprinting).

Player: Crap. I'm gonna run for that door. If I can get it closed before the dragon reaches me, I'll have a little bit of safety. How far am I from the door?

GM: Mm.. You're about 100 yards from the door. ::grins:: Time for a touchdown dash, my friend.

Player: Funny. Okay, my movement is 7, and I'll be sprinting, so it comes to x4, for 28. That'll take me 4 rounds to get there.

GM: The dragon (does some calculations, movement 22, x4 for sprinting for an 88) will reach where you are standing now in less than 4 rounds, but will be about half-way(48 yds.) to the door by the time you get there. You're safe for a little bit.

Player: whew.. So, how sturdy does this building look again?

That's a matter of movement, and there is nothing involving initiative at all. If the dragon were notably closer... The man would be a crispy critter, no bones about it. Unless the man turned to face the beast, initiative would never be rolled, and he would quite simply have to take damage from the thing while he continued to try to run away.