Vincent was talking about Otherkind-ish mechanics in this thread. In that game the values of the dice were assigned at the instance of play rather than being predetermined by the system.

(Uuurk! Can anyone tell me how to do hyperlinks with BBCode?)

That just struck me as being a really interesting idea. What if, instead of having the meaning of the dice be determined by the system you let the meaning of the dice be decided by the group?

I see to possible ways to handle this:

1) The GM decides what elements are important to the scene, and lets the players decide how many dice they want to devote to each.

2) The players decide what elements are important to the scene *through the act* of devoting dice to them.

Vincent's game was an example of the first way: he made a list of important issues (PCs getting damaged, henchmen getting damaged, wall falling, etc.) and had the players roll dice for them.

Way 2 would have the players themselves constructing the list... it would be a Donjon-ish currency mechanic. Each die would indicate one "fact" that the player feels strongly about. A player could devote more than one die to a single fact (indicating that the fact is more important to him than others) to get a better chance of success, but doing so would reduce the total number of facts he could create. Bidding dice on a fact causes that fact to exist... you aren't rolling for directoral power. You're rolling to see whether or not the fact goes the way you want it to. In a way it's like Shadows with multiple simultaneous rolls per scene. Of course, the GM can create facts at any time.

Example:

GM: You now realize what the ominous silence portended - you're surrounded by a mass of rotting zombies.

Player: Right. I notice that some of the big trees around here are old and dead. I want to see if I can fell one on top of the zombies... that should take care of them! I'll devote 3 dice to that.

GM: You'll also need to roll to see if you can get the job done before the zombies get to you.

Player: Well, I have some dice left, I'll devote 2 dice to that.

So the player rolls the dice... if he wins with the 3 dice, he drops the tree on the zombies. If he fails, the tree goes someplace else. If the wins with the 2 dice the tree falls before the zombies are anywhere near him. If he fails at least some of the zombies get to him before the tree goes over. (So, the tree goes over, but more rolls might be needed to keep from getting hurt, or to fight zombies while chopping down a tree, etc.)

Paganini wrote: (Uuurk! Can anyone tell me how to do hyperlinks with BBCode?)

This could be an excellent mechanic, but there's a lot more to do with it before it's useful.

Some places it needs work are (in no particular order)

1. Skills. How would characters be different from one another under a system like this? Ideas include limiting dice expenditure to areas of skill, numerical values of dice permitted in skill areas, giving bonus dice in areas of skill, determining all useful skills ahead of time and adding them up to determine the total of dice available.

2. Whiff. You'll need a fairly forgiving curve or a system that isn't based on the success/failure model to handle all these rolls made with just a few dice.

3. Determination of variables. What are the things that can change? Target difficulty numbers, size of dice, number of dice, addition or highest single die, and other mechanical fiddly bits would go here.

Keep after it.

Jack:

Thanks!!

Ring:

1. - The way I was envisioning it was that the number of dice you have to play with is determined by your character's effectiveness. Remember Mammon and Animus? Something like that would work great with this mechanic. It would hardly even be a stretch. Instead of using your scores as target numbers, use your scores as the number of dice you assign to the different "facts." Draining a Source would give you extra dice for a single conflict. Etc.

2. - Yep. However, I was thinking, the more dice you have, the more likely you are to succeed at *something,* depending on how you bid them. That is, if you have lots of dice, you can assign few dice to many facts (high chance of failure for each one, but the overall chance that *some* of them will succeed is high) or many dice to a few facts (fewer facts, but high chance of succeeding at them). So, as long as players have a fair number of dice to roll, it's no problem.

3. - So many ways to set these up. This is the fun part of design, IMO. :) Going with what I wrote above (number of dice determined by character effectiveness) I would say that the object is to roll well on a single die. If you're rolling more than one die, you get to pick which one you like best. The higher the roll, the better the result. Not sure about difficulty. I almost think that no difficulty is needed in a system like this. I'm pretty sure Vincent doesn't have it in Otherkind, nor Scott in Draconic. Instead, the value of the roll would indicate the quality of the result. Using d6s:

1 - Complete Failure
2 - Failure
3 - Marginal Failure
4 - Marginal Success
5 - Success
6 - Complete Success

It occurs to me that you might need a selection of (or at least one) fixed meta-game rolls. What I mean is, we still need a Narration roll to decide who gets to construct the narrative based on the results above.

How do you think a system like this would work for conflicts between characters controlled by players?

--Emily Care

Emily Care wrote: How do you think a system like this would work for conflicts between characters controlled by players?

Ring Kichard wrote: 1. Skills. How would characters be different from one another under a system like this? Ideas include limiting dice expenditure to areas of skill, numerical values of dice permitted in skill areas, giving bonus dice in areas of skill, determining all useful skills ahead of time and adding them up to determine the total of dice available.