Rethinking the core mechanic

[cyl]

Another thing I didn't mention, but was kinda obvious, with this roll/keep system is that people will hyper specialize instead of keeping some sort of versatility that they can have through skills and charms.


I mean when you have Charisma 2 / Manipulation 2 / Appearance 2 and 2 is the maximum number of dice you can keep after your roll... you don't even try to take some social skills / charms, it's going to be a waste of time and xp.

 

[wordman]

The notion at play in the system I've suggested is sort of that your own limitations (i.e. your attributes) limit how much benefit you can get from training. I've also considered some other variations on the same basic roll and keep idea.


One is that you roll Ability + Attribute + Specialty and keep Essence + Virtue, where the virtue is your rating in a virtue that you must explain how it applies to the test. (Channeling would double the Virtue.) This makes Virtues much more front and center, and avoids some of the issues mentioned above (everyone has similar Virtue ratings). Seems a little forced, though.


Another is the opposite of the main one, where you roll Attribute + Essence + Specialty, and keep Ability. The idea here is that you are limited by skill; your innate make up only gets you so far. This variation has the same issues mentioned above, in reverse.


I suppose you could do a sort of hybrid of roll Ability + Essence and keep Attribute + Specialty. This has a less clear "philosophy" behind it, but makes specialties a lot more important, and gets around some of the issues with mortals that people have been mentioning.


In any of these variations, including the original one, you could use Essence twice, both as dice and as part of the limit.


One last variation, I'm not sure has much of an effect, but it might. Roll as you would in canon, but only use Ability + Attribute dice of the result. This would probably only matter when a lot of gear was involved. It is a more subtle change, though.

 

[cyl]

One thing you could do is to roll the classic pool and keep a minimum of Essence if the attribute is lower than Essence.


Rolling Ess k Ess in the extreme cases (when attribute+ability < Essence), ensures a certain viability of some characters particularily inept in some fields.


The higher the essence, the more chances you have at succeeding, but you still need training to rise above your own inner potential... and that may be another problem.


Some players will be content with their 4k4 and will not try to increase their dicepool via attribute or ability training.


All in all, I do not think that the R/K system could be applied to exalts, unless you change the system of successes (numbers vs 7s)

 

[Gylthinel]

L5R rolls attribute + ability, and keeps (attribute). It balances out the value of abilities by giving various bonuses for high abilities. Rather interesting all in all, but would take a lot of homework to emulate in Exalted.

 

[cyl]

But as I noted previously, L5R (in all 3 editions) does not encourage versatility, on the contrary.


You focus on raising your primary attributes and skills, and too bad for everything else, this is why there is a group and there are many classes / family / school.


Exalts on the other hand can achieve great versatility (favored/caste/aspect skills or attributes), and they need very little time.


Even if you put in place Emphases, like in L5R, you would limit the progression of the characters too much so they can benefit from anything as exalts (like they do now).


Perhaps another solution would be to do the following:


- classic rolls


- cap the maximum successes by attribute+ability+specialty


- allow virtue /wp channel or excellencies to increase this cap


- raise the difficulties threshold: 1/2/4/6/8.


This way you can preserve the mechanics of the game (this is not core-shattering) and can add this "your own limitations limit how much benefit you can get from training" philosophy behind it.


So until you get a decent level of mastery, feats that are legendary are forever out of your reach.

 

[wordman]

Another variation just occurred to me, which sort of changes my thinking about the whole notion...


When you use a roll and keep system, how the "keep" part works allows you to tune your game to focus on behavior you want to see. That is, if, by doing X, a player can raise their keep limit, then the player is given a very strong incentive to do X. As has been mentioned, in the original system discussed, X is "raise Attributes". That doesn't seem like a great thing to emphasize. The also mentioned notion of X being "channel Virtues" is more like it.


With that in mind, what about this variation:


Players roll Attribute + Ability + Specialty, just as they do in canon. By default, they keep twice Essence dice; however, stunting changes such that, instead of awarding dice, every "level" of the stunt adds two (or maybe three) to the keep limit. Channeling virtues adds the virtue to the limit (and, possibly, no longer costs Willpower). I'd still keep the notion of one excellency per type of exalt, similar to what I listed before:

• Solars, Abyssals, Infernals: 2m buys 1 sux, adding no more than Ability sux.
  
• Lunars: pre-roll, 1m converts one die into a success. Limit of Essence motes spent like this for any given test. (Attribute based)
  
• Sidereals: 3m allows 9s to generate two successes instead of one on the roll.
  
• Terrestrials: 1m adds two dice to the pool, max of Ability + Specialty
  
• Alchemicals: 2m raises the cap by two, only 2m per test. (Attribute based)
  
• Spirits, Ghosts, Mortals: 1m adds one die, up to Ability extra dice.

What you have then is a system that provides serious incentives to stunt, which I like quite a bit.

 

[cyl]

Hmmm I have several concerns with this proposition:


1- what if Ess x 2 > dicepool of the character ? do they roll & keep Ess x 2 as a minimum ?


2- what about low essence users / mortals with high attributes (Ess x 2 < dicepool of the character) ?


3- your original concern was that the legendary feats were too common in exalted, and now you are talking about limitating the dice kept, but with a system of stunt that give +0k2 per level of stunt, and +0k(virtue) without the need for a wp channel...


So let's take an example:


mod: big joe, 5+5+3 (ess 3) so big joe has a dp of 13k6. He stunts at 1, his pool becomes 13k8, and channes virtue (4), 13k12.


canon: 13 + 1 + 4 : dicepool of 18.


Is 13k12 better than a dp of 18 when your goal is to make legendary feats more legendary and less common ?


Honestly... not much IMHO.

 

[Gylthinel]

But as I noted previously, L5R (in all 3 editions) does not encourage versatility, on the contrary.
You focus on raising your primary attributes and skills, and too bad for everything else, this is why there is a group and there are many classes / family / school.

Well, I haven't played the game, but the rules don't seem to agree with this statement. Your school rank is limited by your Insight, which goes up faster when you're versatile, and slower when your focused. I.e. you can have a master swordsman w/ a high attribute+ability, and a low insight/school rank and, thus, his school rank would be less than somebody that paints with a broader brush.

 

[cyl]

Insight is based on Rings mainly (1 Ring = 2 paired attributes), and then skills.


The formula is Sum of rings x 10 + skills (so really skills are minor here).


You can't be too focused in the long run, but at chargen, it is more likely that you are going to raise 1 attribute to 4 and leave the other to 2, so that you can increase your Ring at a cheaper cost (raising 1 attribute from 2 to 3 and from 3 to 4) than if you had chosen to leave both attributes (therefore the ring) at 3.


Plus there is the question of the viability of the character... in L5R not everyone has to fight for their lives, in Exalted, everyone must be able to survive a fight.

 

[Gylthinel]

Well, this isn't really a thread about L5R, I'll just say that I haven't played, but my GM reports on the system to me every other day and states rather frequently how spread-out character's insights go up quite a bit faster than dice hording munchkins. I find the mechanic to be a bit forced, but.... what do I know?


As this thread is about rethinking the core mechanics, I'd like to jump onto that topic again. I think that the roll/keep idea could be fun and functional, but ultimately I don't really see the point in it, other than simply crunching old numbers in new ways. My interest is in streamlining the game, making it faster and easier to play without losing the depth in character creation/advancement. I have this nebulous concept that doing something like the ORE (One-Roll Engine) system would function to streamline the game and make it flow faster.


In my experience, I've found the game to stall rather frequently once the players start rolling with several different modifiers. Specifically, I call for a roll, they respond with a stunt, I give them some kind of stunt reward, they select their charms and apply their stunt and roll it. Now, this seems pretty simple, but I find this much back/forth to become tedious during events that involve a ton of dice rolling (i.e. combat). Having this much interaction on each roll starts to soak up oodles of time, and thus I find combats to drag out into painful crawls, rather than clipping along at a brisk and fun pace (like they do in, say, Silhouette).


Of course, a lot of this issue comes from the fact that we game on MSN messenger (we cannot use mics for various reasons). And a lot of the issue is that my players aren't fast with the dice or the keyboard even on a good day. As these are shortcommings I cannot overcome, I find the only option to be somehow revamping the system. Which I feel it needs regardless of the hurdles I personally experience.


The question is: what do you think? Do you think the system is too clunky? Is there a way to get a faster system without simply clipping out the fun little details?

 

[wordman]

mod: big joe, 5+5+3 (ess 3) so big joe has a dp of 13k6. He stunts at 1, his pool becomes 13k8, and channes virtue (4), 13k12.
canon: 13 + 1 + 4 : dicepool of 18.


Is 13k12 better than a dp of 18 when your goal is to make legendary feats more legendary and less common ?


Honestly... not much IMHO.

In this particular case, it matters a little. Slightly more when you add 10 excellency dice.


Here is a Monte Carlo simulation comparing your scenario to canon, using 1 million rolls. In this simulation, the standard pool (13 dice) was rolled, and the same pool evaluated under both systems (with the stunt and virtue adding five more dice to standard system, and setting the cap on the roll and keep):

(The dark boxes are canon, the label is messed up.)


This is the same simulation run again, but this time with the standard pool assuming 1 additional dice from the 1st Excellency. That is, 23 dice evaluated under both systems, with the stunt and virtue adding five more dice to standard system, and setting the cap on the roll and keep:

Without virtue channeling, stunting or excellencies, the same case is a bit more interesting:

Note how "legendary" five successes is in any of the cases above.


It's a bit closer to earth for the case of Essence 3, Ability=3, Attribute=3, no specializations, no virtue channeling and a two die stunt?

 

[wordman]

I have this nebulous concept that doing something like the ORE (One-Roll Engine) system would function to streamline the game and make it flow faster.

You'd probably like ORExalted.

 

[cyl]

My conclusion: your proposition of Roll / Keep is not a significant improvement of the chances to succeed on a roll with the legendary difficulty.


Even with only 13k6 the chances are still pretty good, and we're only at essence 3.


IMHO, if you want to change this "players reach legendary and + too often" fact you got two options:


- increase the difficulty


- change the whole system (which is your choice)


I am sure that if you adopt the L5R system you'll start to close in on your objectives.


The basics are the following:


- Roll / Keep (attribute + ability K attribute) TN, and raises (gambles, increasing the TN by +5 / raise).


- maximum 10 die roll. 2 cases if attribute + ability > 10,


1) add +k1 per die above 10


2) if attribute + ability > 20, every die above 10 is a free raise.


Raises are an interesting part of this system because it helps making rolls (simple, contested, social or combat) more interesting.


You have the difficulty, and TN, and you can play on both side to win (or loose) the upper hand.


Just a few example of how raises ® work in L5R3e combat:


Disarm: 3R (+15) + contested roll


Extra Attack: 4R (+20)


Feint: +1R (+5) gain a free raise on your next attack


Guard: +xR (+5/ TN granted) protect someone from someone else


Increase Damage: +1R for +1k0 +2R for +1k1 on the damage roll etc etc


Knockdown: +3R


Both systems are very similar, and a conversion from Exalted to L5R system would be really easy.

 

[cyl]

Forgot to mention the TNs with L5R ranging from 5 = very easy to 50= impossible. (and then you have +5 TN /Raise)


That's very high considering the roll / keep system.

 

[Gylthinel]

I have this nebulous concept that doing something like the ORE (One-Roll Engine) system would function to streamline the game and make it flow faster.

You'd probably like ORExalted.

Thanks for the link. I'll have to take a look, but I'm not going to buy Reign just to try this out. I've read over Wild Talents and wasn't that impressed by the ORE system, but I like the IDEA of the ORE system. The idea of a singular roll being used for multiple things is captured in the rules for the upcomming Bount Head Bebop "Inverted D20" system as well.

 

[cyl]

Forgot also to mention that the big problem with L5R was that counting the results of actions slows down the game.


But a mix could be done beteween roll / keep and successes with but a few steps.


1- lower the TN to 6 for determining successes (10 still counts as double)


2- conserve this roll/keep thing.


3- set that if Essence > attribute, then an essence user keeps Ess dice instead0


4- use the difficulty ranging from 0 to 10 (DV is unchanged).


5- implement this Raise thing.


6- modify the excellencies: 1st excellency = +1k0 2nd excellency = +0k1, 3rd excellency = unchanged.


7- wp channeling add +1 suxx


8- virtue channeling adds + 0k(virtue)


9- stunts add +1k1 / stunt.


This way, mortals can kiss butt kicking exalts goodbye, essence user have the upper hand thanks to their essence rating, and above all the impossible stays difficult to reach and can truly become possible only if you spend some motes.


Note: you'd have to adapt the stats of the weapons also to this roll/keep thing... but that's no big deal.


Thoughts ?

 

[wordman]

I'm not going to buy Reign just to try this out.

Don't.


Instead, buy Reign because it is a really neat game. While its setting doesn't do much for me, nearly everything else about it does.


(Oh, also, if you want to tinker with the ORE, I analyzed its probabilities a while ago. You might also look at how A Dirty World uses the ORE. I've been thinking about how that might work for Exalted, but haven't gotten very far.)

 

[Gylthinel]

(Oh, also, if you want to tinker with the ORE, I analyzed its probabilities a while ago.

Wow, my brian imploded about 1/4 of the way into this post. I do enjoy dice probabilities, but I'm afraid I'm not that smart.

You might also look at how A Dirty World uses the ORE. I've been thinking about how that might work for Exalted, but haven't gotten very far.)

Arg, more money!


Ultimately, the issue that I have w/ ORE is it doesn't include what I want it to include into the single dice roll. I'm not interested in a single die roll that handles speed, damage, hit location and hit probability all in a single roll (since speed is well represented by weapon speed IMO, damage is easily handled even w/out a roll in the core system, and hit location is irrelevant in most RPGs). What I want from my single roll is for it to encompass both Attack and Defense. I imagine ORE might be able to do that, but I'm not entirely sure how. Perhaps through the assiduous application of ORE's multiple action rules.

 

[cyl]

Meh... realized my own stoopidity 3 days late.


On the last post I wrote I stated that with the mod I proposed the Difficulty should range from 0 to 10, but that would actually be quite silly since you can't roll more than 0k5 without magic/stunt.


I'm wondering if DV would actually have to be changed or not... but this roll / keep thing might make the players a little less paranoid and less inclined to have a DV of 10 at chargen, helping characters to be a bit more balanced.

 

[wordman]

I'm finally getting around to posting some things I figured out a while ago about this roll and keep system.


The problem with my most recent scheme for roll and keep is, as cyl mentions, the results are pretty close to canon most of the time. This is because the "keep" number is almost always pretty close to (or greater than) the pool size. It's about twice as big as it needs to be. When proposing it, I was a little wary of doubling of various things in the calculation, but did it because otherwise mortals get hosed.


And then I realized, I don't really give a fuck about mortals in Exalted.


So, I played around with my simulator a bit more, and changed and tweaked the rules that drive it. Unlike my initial proposal, I now understand what happens to the numbers. I kind of like the behavior in the end result, but more because how it differentiates the different types of exalted than due to the "difficulty scale".


The "difficulty scale" issue is a bit better under this system, but is still "stretched" compared to the notion of "five successes == legendary". That's probably OK.


Current system is this:

• Roll Attribute + Ability + Specialty dice.
  
• Keep Essence + 1 of die.
  
• Stunts add dice to the pool.
  
• Virtue channelling adds to the number of dice you keep.
  
• Virtue channelling does not cost Willpower
  
• Solars, Abyssals and Infernals have only one excellency: Add 1 sux/2m, max Ability successes.
  
• Lunars have only one excellency: Pre-roll, convert up to Attribute die to successes, 1m/die.
  
• Sidereals have only one excellency: Pay 3m to add one to the number of dice they keep, and count double successes on 8 and 9 as well as 10. (This sounds insanely powerful, but turns out not to be.)
  
• Terrestrials have only one excellency: Pay 1m to add 2 die to the pool, max Ability + Specialty dice.
  
• Alchemicals have only one excellency: Pay 1m to raise the number of dice they keep by two. This can be done only once per test.
  
• Spirits, Fey, Ghosts, etc have only one excellency: Pay 1m to add 1 die to the pool, max Ability.

What took so long to post this is that I decided to turn it into a blog post. You can see more detailed analysis and pretty graphs here.

 

[cyl]

No 10 dice rolled limit ?

 

[wordman]

No 10 dice rolled limit ?

No. That would actually defeat the properties of the math.

 

[cyl]

So...


- Essence becomes the most important trait of the game


- Lunars are really screwed unless they can convert dice up to Attribute, regardless of the Essence+1 limit, that was not clear.


I would change the spirit exc to this: 2m / adds +0k1, max virtue / essence for universal charms.

 

[cyl]

Hmm you are still considering this for 1e right ?


I have a sudden doubt about the 2 dice rolled by mortals...

 

[wordman]

- Essence becomes the most important trait of the game

It already is. Well... better Essence than Dex, at any rate.

- Lunars are really screwed unless they can convert dice up to Attribute, regardless of the Essence+1 limit

The conversion is not affected by the limit. Once converted, those dice are not rolled, so step outside of the realm of "roll and keep". You can see from the graphs in the post that lunars are far from "screwed". Generally they fare only slightly worse than solars (while spending only half the motes, at least in the examples in the graphs).

I would change the spirit exc to this: 2m / adds +0k1, max virtue / essence for universal charms.

Why?

you are still considering this for 1e right ?

Yes, though only because I haven't really thought about how it interacts with DV yet.