Dice Pool Crunching

[snakefing]

I tend to agree with you that I think dice pools of around three represent a fairly good average starting point. Maybe two dice for the least skilled, or for skills that need lots of extra training.

1 Success = easy task, generally the kind of thing that would be fairly routine for a decently trained and skilled individual.

2 Successes = moderate task, the kind of thing that would challenge a reasonably trained individual.

3 Successes = hard task, something pretty tough even for well-trained individuals.

4 Successes = very hard task, something difficult even for talented experts.

As for modifiers, there are a bunch of cases to consider:

Opposed task: Here the acting character and opposing character may both have modifiers that cancel each other out. You might expect this to have modifiers in the range -2 to +2, which is quite a wide range. (For example, if the base target number was 5, then the -2 modifier makes it impossible, while the +2 makes it very easy.)

Straight up skills: Some skills won't have a modifier (the ones that might otherwise be based on metal stats). Some skills will. The overall modifier would normally range from +2 to +5, where it exists. Otherwise, it's just 0.

Combat: Combat is a slightly different case, because you need to both hit and do damage. The multiple die rolls leads to a more complex situation that needs to be explored in detail. But generally this is fairly similar to the opposed skill case, at least in concept.

[snakefing]

One thing to think about is using modifiers to add or remove dice from the pool instead of modifying the dice roll. This might simplify the matter of setting base difficulties.

I want to compare the dice rolls and stats needed to complete an attack in our system vs. in d20:

d20:
Attacker's BAB, determined by class and level, as modified by feats, strength, and weapon bonuses.
Attacker's damage dice, determined by weapon, as modified by feats, strength, and weapon bonuses.
Critical threat range and multiplier, determined by weapon, as modified by feats.
Defender's AC, determined by armor, as modified by dexterity , natural armor if any, and armor bonuses.
Defender's DR, if any.
Situational modifiers, such as spells, flanking, aid another, etc.
Die roll d20 to hit.
Die roll (damage dice) to damage.

Our system:
Mostly TBD
Attacker's COORD modifier (feats to modify?).
Attacker's dice pool size, determined by skills and possibly weapon type.
Attacker's STR modifier for damage.
Attacker's damage dice pool sice, determined by weapon type and probably modified for skills.
Defender's REACT modifier (feats to modify?).
Defender's TOUGH modifier for damage resistance.
Defender's Armor for damage resistance.
Dice Pool roll Nd6 to attack.
Dice Pool roll Nd6 for damage.

I'm sure I missed something in there. I just want to make sure we're not creating something meant to be simple that turns out to be complicated.

[beejazz]

Meh. DnD rolls attack vs. AC. Then rolls damage.

So do we, except damage based on # of successes.

No biggie, right?

Anyway, "when I can buy what" just bears a passing resemblance to levels, in that it can be used to categorize gross levels of power. There's a lot more subtlety and gradation here, not to mention a universal system for advancement (as opposed to feats AND skills AND class features all drawing from diferring pools, plus static attack and save progressions).

Again, tactics that focus on the melee/ranged divide and the ground and movement bonuses/penalties are good, IMHO. One opponent wants melee, one wants ranged, control of ideal ground is important.

As for skills, just straight up abilities. two to five (start two in each and three to distribute). All there needs to be.

Alot of this is that it'll be a tougher system to balance, not that it will be tough to play.

[snakefing]

We aren't using that much more in the way of dice rolls. But we do have some different ways to modify them. I'm just trying to keep track of it so it doesn't get out of control.

Mechanically it is probably sound to limit spending on any one item to some fraction of total experience. But I'm not sure that keeping track of how much total experience you have, and how much you've spent on each thing, is necessarily the best idea. I mean, I'd like to think that once I spend XP on increasing my attack dice, after that all I need to care about is what my current dice pool is, not how much it cost me to buy it.

For that reason, I'd like to try to keep those limits in the way of simple procedural things, like max buy at one time (say, only one improvement level at a time), prerequisites, etc.

[beejazz]

So one level two die requires four level one dice, I suppose?

[snakefing]

Not sure. It seems reasonable. At one point I floated the idea of package-type prerequisites. I didn't get much response. In this instance, it might work something like this:

Anyone can buy one die in their favorite weapon(s). That represents training or experience in a particular weapon.

The next step Advanced Melee Fighting. This requires at least one die bought in some weapon, plus some additional prereqs (undetermined at this time, effectively some other fighting feat) that represent branching out a little into general experience as a fighter.

Advanced Melee Fighting is a prerequisite for the second die in a weapon. This represents advanced training. It would also be a prereq for some of the more advanced fighting feats/skills.

Repeat this with Expert Melee Fighting, for access to the highest level of abilities, and possibly Epic Melee Fighting for access to near-miraculous levels.

If that didn't make sense, imagine it this way:

At the top of the page, we list all the basic fighting stuff: proficiencies, adding one die to the dice pool, all (or most) of the initial feats in the feat trees. These are things that anyone can take, any or all of them.

Below this, draw a line representing Advanced Melee Fighting. You have to get this before you can get any of the advanced abilities. The requirement for Advanced Melee Fighting is that you have to have, say, three of the abilities from the basic list.

Now we list all the advanced techniques: 2d6 added to dice pool for a weapon, feats from higher on the feat chains, etc. Anyone with Advanced Melee Fighting can take these, mix or match.

Draw another line for Expert Melee Fighting, then list all the expert techniques, etc.

I'm not sure this works well, I haven't worked out all the details. But the basic idea is that you can't just keep buying dice in a single weapon - to get to the Expert levels you have to be at least a little committed to combat.

Then repeat the process for Advanced Social Skills, Advanced Ranged Attack, Advanced Firearms, Advanced Spell Casting, etc. In some ways, it is kind of like a class system, since it encourages characters to choose one or two areas to advance to Expert levels in. But at least all the basic skills are open to anyone.

[beejazz]

As for the extra dice tech tree, it could work.
Rather than specific prerequisites, though, let's just divide skills into types.

Die two of a weapon proficiency requires a certain number of combat skills with at least one die.

Rather than having one die feint, one die trip, or what have you.

Specific abilities, on the other hand, might have specific skill prereqs. "Dirty fighting" for example, may require "one die feint."

Numbers of dice are the basic stock and trade of the system, and shouldn't be too complicated or specific.

[beejazz]

Also, I'm worried about ability scores and defense values...

If it's even possible for defense values to be low enough that automatic success happens against them, then some characters will get critted for 100% damage 100% of the time.

What ability score generation system and static modifier (like the +10 in DnD) could make this work? Without the automatic success worries.

Or would changing critical success to "four sixes" be a better idea?

[snakefing]

I'm thinking it is time to put it together a bit, to see how things play out in various scenarios.

Proposal:
[table=Dice Pools]
[tr][td]Non-proficient[/td][td]2d6[/td][/tr]
[tr][td]Normal[/td][td]3d6[/td][/tr]
[tr][td]Advanced[/td][td]4d6[/td][/tr]
[tr][td]Specialist[/td][td]5d6[/td][/tr]
[tr][td]Legend[/td][td]6d6[/td][/tr]
[/table]

Characteristics range from 2 (fair) up to 7 (legendary).
Attack by rolling DicePool + COORD against 3 + Armor DV + TOUGH
Normally, 2 successes required. Possibly some defensive feats might increase this, but for now I'm not considering this.

So with evenly matched attributes, the target would be 3 + defensive value of armor. This would have to be limited to 1, 2, or 3. (Armor would have a higher impact on the damage side.)

Anyway, there's something concrete to look at. How does it play out if the attacker or defender has superior attributes?

[snakefing]

[blockquote=CYMRO]
The more I look at this, the less I like dice pools.  Not when there is a simple opposed roll system in place with a simple and convenient 5% increment built in.
[/blockquote]

As I've said before, I'm not the world's biggest dice pool fan. But they do appeal to some people. Right now I'm interested in seeing what is the best we can do with them. My biggest beef with them is that if you get to where people are rolling 6,7,8 dice or more, that's really no easier than doing d20 + 17, compare with 31.

From the simplicity standpoint, you can simplify a lot by not modifying the dice rolls. Use modifiers to add or remove dice or change the successes required.

After some discreet questioning around several different kinds of boards, reduction of complications in rules is the biggest objective for those looking beyond d20.
This does not do it.

I'm always a little cautious about taking people's word for it on stuff like that. What they usually mean is, "Get rid of all the stuff I don't want, and keep the stuff I do." Which works fine for them, but no one really agrees on what is necessary and what isn't.

When I worked on Ysgarth, the main author was just dead set against the tactical display, miniatures, or anything of the kind. Too wargame-y for him, I guess. On the other hand, I like it. Not necessarily the strict grid and movement rules of D&D 3.5, but at least some visual display makes it more interesting for me. I don't need miniatures - just Monopoly pieces works great. But something.

Simplification is good - until it removes the stuff you care about.

[beejazz]

I'm thinking it is time to put it together a bit, to see how things play out in various scenarios.

Proposal:
[table=Dice Pools]
[tr][td]Non-proficient[/td][td]2d6[/td][/tr]
[tr][td]Normal[/td][td]3d6[/td][/tr]
[tr][td]Advanced[/td][td]4d6[/td][/tr]
[tr][td]Specialist[/td][td]5d6[/td][/tr]
[tr][td]Legend[/td][td]6d6[/td][/tr]
[/table]

Characteristics range from 2 (fair) up to 7 (legendary).
Attack by rolling DicePool + COORD against 3 + Armor DV + TOUGH
Normally, 2 successes required. Possibly some defensive feats might increase this, but for now I'm not considering this.

So with evenly matched attributes, the target would be 3 + defensive value of armor. This would have to be limited to 1, 2, or 3. (Armor would have a higher impact on the damage side.)

Anyway, there's something concrete to look at. How does it play out if the attacker or defender has superior attributes?

Okay... at first I was going for the two-to-five range for a cleaner look... but three to seven may work (we have to work out the consequences)...

For base dice, I've always been partial to three. With a range of possible target numbers. The lowest requiring the highest ability score to roll a two. The highest requiring the lowest ability to roll a six. In defense values, I would go for something similar. I like your idea to have both a base modifier and armor values. And since these are static values, they could easily be written on to the character sheet.

As for analysing the superior values... I'm a little short on sleep at the moment.

[beejazz]

As for "six, seven, eight dice"... I would almost be willing to propose a cap somewhere around six total.

[beejazz]

And the "vs. tough" roll determines damage by number of success.

It's the "vs. ref" roll that would determine hit/miss.

Otherwise, why have ref at all?

[beejazz]

EUREKA!

Abilities range from three to six.
All abilities start at three and then there are three additional points to distribute.

The base defense modifier is three.

Armor values range from one to three, with corresponding required successes. Maybe a max reflex for each (5 for armor value 1, 4 for armor value 2, 3 for armor value 3).

Difficulties for skills still range from seven to nine. Still between one and three required successes. Gives us nine difficulties to work with.

And critical successes on three natural sixes.

Dice still priced on the 10(x^2) formula. First die has no requirements. Second die requires three to five skills of the same type (combat, movement, technology, etc.) with at least one die. Third die is the same, but requires two dice in three to five skills of the same type. Default dice, as always, is three.

Weapons: Each weapon group is a skill. Nonproficiency starts at three dice, like any other skill. Exotic weapon group allows players to apply their extra dice to exotic weapons of the appropriate group. A similar feature may allow players to apply their weapon proficiencies while fighting in a mech... or for gun turrets or what have you.

Advantage/Disadvantage: Tactics rely on two things really: movement and movement. Action points are the most easily modified numbers for circumstance. Players might spend five to act out of turn, trip an opponent to prevent him from attacking or moving in the next turn, etc. Depending on a character's build, melee or ranged might be more ideal... hence the importance of engaging in or avoiding melee.

Alternate hit points idea: Rather than have "number of successes" to hit at all, have three teirs of hit points. This would work kind of like vitality and wound points, but there would be superficial, moderate, and vital damage. Number of successes (in the attack roll, not the damage roll) determnes which teir is hit, and if a teir is filled, damage spills over. This could be kept track of whith checked boxes on the character sheet. Depending on which teir is damaged, characters may get action point penalties. -2 for superficial, -4 for moderate and -6 for vital seems like a good baseline. Just a thought.

[snakefing]

Armor values range from one to three, with corresponding required successes. Maybe a max reflex for each (5 for armor value 1, 4 for armor value 2, 3 for armor value 3).

This means that if your REACT score is high, increasing armor won't decrease your chances of getting hit. The increase of armor defense is cancelled by decrease of effective REACT. Of course, it would still decrease the chances of being damaged. That could work, but it would be something to watch in terms of balance.

Alternate hit points idea: Rather than have "number of successes" to hit at all, have three teirs of hit points. This would work kind of like vitality and wound points, but there would be superficial, moderate, and vital damage. Number of successes (in the attack roll, not the damage roll) determnes which teir is hit, and if a teir is filled, damage spills over. This could be kept track of whith checked boxes on the character sheet. Depending on which teir is damaged, characters may get action point penalties. -2 for superficial, -4 for moderate and -6 for vital seems like a good baseline. Just a thought.

Hmmm. Not sure I like the additional book-keeping.