The Forge Archives

I've been pondering this game design problem for a while now, and thought I'd ask for some input from the Forge crowd.

I'm trying to devise a system such that it can be played with dice (or other randomizer), and also diceless, without requiring any change to mechanical entities, only the resolution systems that interact with them.  The dilemma I've reached is one of elegance of mechanics vs. complexity of play; while many modern diceless systems have something like a resource-allocation mechanic to produce variation of results, a diced system doesn't need such a contrivance.

http://www.indie-rpgs.com/viewtopic.php?t=6877

This thread is a discussion that jburneko and I (with some comments by others) had about making Sorcerer (a diced game) diceless through changing as few mechanical entities as possible.  The discussion sort of trickled off, but before it did there were a number of excellent ideas that, depending on how your system is made up, might at least give you some ideas.

I hope that's of some help.

There's an easy way to do this -- make the dice system a resource system as well.

Consider this quickie system:

Characters have a pool of points. You spend points to attempt a task -- as many as you want, but before you know for sure how tough the task is. The idea is to beat a certain difficulty. In the diced version of the system, spending a point allows you to roll a d6 -- add the results together and compare to the difficulty. In the diceless version, each point spent is worth a "roll" of 4 -- add them all together and see if you beat the difficulty.

So the diceless version of the system is like the diced version of the system, with slightly-above-average roll results.

Does that make sense?

I have been investigating exactly this sort of arrangement. I think the important thing to consider is that the diceless system must encompass the same range as the die roll. Of course, this is the real challenge. My solution has been to use a curved die roll, such as 1d6 curved to produce the following results:

1 = -2
2 = -1
3,4 = 0
5 = +1
6 = +2

or 2d6 curved as follows:

2 = -3
3,4 = -2
5,6 = -1
7 = 0
8,9 = +1
10,11 = +2
12 = +3

or 2d10 to produce results +5 to -5, etc.

Then, for the diceless component, the player gets a pool of points with a fixed starting value such as 2, 3, 5, or 10. The player may then assign modifiers within the range of the dice. Of course, the player must keep the pool built up. If it reaches 0, the player is out of points and must take a penalty. Alternately, the player can be allowed to deficit spend for one result, and that deficit becomes a penalty until the pool is returned to 0 or higher. It's a managed resource scheme, but I like it because it gives the player a high degree of control.

I do make a certain number of assumptions with this method. Attributes are assumed to run from 0 to 10 or along a similar range of small numbers. Also, the die roll generates positive and negative numbers so the player doesn't need to jump through hoops in order to refresh the pool. It refreshes as play progresses. Results are compared to a target number.

Edit: Typos

Quote from: xiombarg...
So the diceless version of the system is like the diced version of the system, with slightly-above-average roll results.

Does that make sense?

While ingenious, I think this overlooks one potential advantage of diceless systems, which is that game mechanics are merely the starting point for resolving an encounter. They merely who has an innate advantage based on ability and skill, but the situation and how the characters approach it actualy determines how the encounter plays out.

For example in a sword fight, the character's relative swordsmanship abilities might dictate that one was a better swordsman, but that does not mean that character must win. It merely establishes that this character has an advantage, but how the contest is narrated will determine the outcome, which is not deterministic.

Merely replacing randomisers with a resource allocation system still leaves a game in which game mechanics controll outcomes, rather than game narrative.

Having said all that, there's nothing inherently wrong with employing resource allocation systems in RPGs, and as I said this is an ingenious way to replace dice.


Simon Hibbs

Well I had an interesting thought last night which is not exactly what is described by hopefully might permit a new angle.  Speed speed speed is a major criteria for me and so I was wondering how to eiminate multiple die rolls in a single exchange.  The reason I want multiple die rolls is so I can have multiple inputs impinging on the resolution, and the reason I don't is that they add handling time.  So after a bit of tinkering I thought, why not make a pack of cards each of which has a matrix of results that the system my generate.  So a given card for Vampire, for example, would have outputs for die pools of each size, a humanity roll etc; one card is one batch of results with all special cases taken into account (e.g. exploding dice).  If multiple rolls are required, a single card flip gives you discrete outputs for each roll to be made simultaneously.

So applying this to above, for this is still a randomiser, I wonder if there is a difference between heavy karma and most diceless play.  I don't think that in itself is radical (Conspiracy-X eliminates die rolls through karma), but tied to the card device above you could publicly reveal a set of cards which will, in sequence, be employed as a randomiser.  Play then revolves around who gets to utilise which space on the card containing a number when.  So, off the top of my head, you could do things like say each block on a card, as representative of a particular type of roll, can only be used once in a given exchange in a conflict.  Initiative would be important because you'd be able to lay claim to the best set of available results; losing initiative would be bad becuase you would have to deal with the leftovers.  On the other hand, a set of numbers might come up that make an apparently irrelevant ability useful in some manner.

So in any way, I guess the general principle is that a random system should be seamlessly convertible to a non-random system if the random element becomes Karma-ised.  You could in principle do something very similar by generating a page of numbers and employing other devices to determine who gets which when.  Everything will still be about decision and planning.

Quote from: Shreyas SampatI've been pondering this game design problem for a while now, and thought I'd ask for some input from the Forge crowd.

I'm trying to devise a system such that it can be played with dice (or other randomizer), and also diceless, without requiring any change to mechanical entities, only the resolution systems that interact with them.

I don't know! I can see how you could add dice as an extension of a diceless resolution system though.

You start with a simple resource allocation system that lets you assign points to different activities on a one point = one success basis.

eg. "I'll assign 6 points to Attack, and 1 point to Defence. Let's see if your defence can beat my 6 successes attack."

In the dice version you replace points with dice. Say d10s where 1 through to 5 generates that many successes, and 6 through 10 counts for zero (ala Universalis) means that each dice rolled is averaging one success, just like the point system but with a random element.

eg. "I'll assign 6 dice to Attack, and 1 point to Defence. I roll a 1, 1, 4, 6, 7, and 10 with my Attack dice. Let's see if your defence can beat my 6 successes attack."

So the core is a resource allocation system, the resources are straight successes, or dice that yield successes.

If I am following correctly, then we are talking about a system which can alternativelyly be used either with or without a randomizer, right?

If so, then I side with simon a bit in that a diceless system is more than just you don't roll the dice. I mean d20 has this with the take 10 or take 20 rules. There is nuance to diceless design that is missed when approching it as a dice design without the dice (both in design and in attempted play).

I also don't think the idea of the systems being compatable will really work. Some examples that come to mind are T&T solo vs group play, where solo characters, or the rewards of solo play are often disallowed. And Warhammer battle to WHFRP conversions (or any rules-to-rules conversions) which is always nice to have but IME rarely used, if ever.

Thanks for the thoughts, folks.  I'll be posting some stuff in Indie Game Design once I have a firmer grasp of the design I'm working towards.

Quote from: Chris, a.k.a. xiombarg, was on the right track when heThere's an easy way to do this -- make the dice system a resource system as well.

Let's imagine that you're doing base score plus resource against unknown score plus resource.

In the diceless version, the unknown would have a value ascribed to it that was secret; that is, overcoming this obstacle would involve beating a base of 4 plus a resource expenditure of 3, but that would be fixed in the scenario. The player has a base of 7, so the expenditure of 1 would be sufficient to win, but he doesn't know that. He has to decide how to ration points from a pool which must be replenished by some means provided in play.

In the diced version, there are no points, and the targets are now not fixed. The obstacle has a base of 4 plus a resource of +d6; the player has a base of 7 plus a resource of +d6. Each side rolls and adds its results to the base, and the numbers compared.

The difference between them is that the diceless version requires strategic consideration of how to ration points, and the diced version provides an essentially unlimited number of points but limits use to the number rolled.

--M. J. Young

Back when I had a website, I had rules for Diced Nobilis (I like the diceless version myself, they were largely written to show that such conversions were possible) that essentially worked the way most people are suggesting -- they converted the diceless attribute totals into a roll and add pool of dice.

But I was thinking -- this is essentially adding a minute failiure chance into a dice-based resource system -- it isn't that interesting.  And then I had an idea for a mixed dice / diceless mechanic which interested me.

Essentially, you have your base scores and your resource pools ala Nobilis and such.  When resolving an action, you add your spent resource points to your base score and the total gives you your effectiveness. OR, you can roll a number of dice equal to your point expenditure, take the highest, and add your base score (ala Sillohouette.)

So, essentially, you can get things cheaper if you risk failure.  However, past a certain point, luck will no longer avail you, and you must blow your resources.

Is this interesting as a resolution mechanic?

yrs--
--Ben

When I was working on my Matrix RPG, I had a dichotomous system that avoided the congruity problems by making them mutually esclusive. Essentially, anyone still hooked to the matrix rolled for success. (xd10 where x is the number of relevant stats, highest result over difficulty is success).

However, if a character is freed from the matrix, their sense of self-confidence is such that they automatically succeed actions whose difficulty is below their freedom rating (which can get fairly high) + the ratings of any other relevant traits. An actions with a higher difficulty, like defying gravity or fighting an agent, start out with the same number plus an expenditure of points from their freedom pool.

Shreyas started the thread, and he's satisfied with the outcome. For that reason, I considered it ended; but now I see it's not ended. I seem to have gone over the top here. Maybe someone will find the response useful.

IMO, people who are considering designs that combine dice and dicelessness, or even people who are contemplating one option over the other, need to look at the topic more formally and consider the underlying reasons for each decision.

Dice serve a function. They provide both an element of risk and a sense of realism. The element of risk comes with the chance that the dice can bless or screw the player on any given die roll. The sense of realism comes with the fact that we as people really don't know exactly what will happen when we do something. We may have a high degree of certainty about it, but there is always a chance events will turn out better or worse than expected.

Resource-based dicelessness also serves a function. It provides an element of control and a sense of self-determination. The element of control comes from knowing that the player can allocate the points to affect the outcome. The sense of self-determination comes with knowing the ramifications of every decision. If the player allocates the points now, they won't be available again until later.

Now, recognizing that these two options serve different functions, it seems to me that a designer must make some self-evident and  different assumptions about using each one. When the designer chooses dice, the designer is choosing an element of risk, an element of surprise, a desire to gamble, and a desire to accept the unknown as part of the equation. When a designer chooses dicelessness, the designer is choosing player control and a right to self-determination.

With this in mind, I assert that mixing dice and dicelessness (die roll + resource in the same mechanic) is a poor design decision because it muddies the water. If you roll dice, you agree to take a risk. At that point, spending resources just dulls the sense of risk. If you spend resources, you are demanding self-determination. At that point, rolling dice merely diminishes the player's degree of control.

If strategy is important to the designer, it will play a role in the decision-making process whether the game's mechanics use dice or not. In any game, players can look at the situation, the character's ability, and the risk in deciding on a course of action. Strategy is not the crucial issue. The crucial issue is desire to gamble vs. value of character ability vs. desire for control.

MJ made an interesting point when he said that the "diced version (of a game) provides an essentially unlimited number of points but limits use to the number rolled." As a broad statement, it's wrong; but it's true within the constraints of a game session. The statement is wrong because, when dice are rolled enough times to accurately generate a statistical distribution, the average of all the results always migrates to the middle of the range. The point pool is effectively limited. This is the whole basis for spending resources: Though the numbers always average, the player can limit the occurence of really unpleasant results and choose the moments when his character will look really cool.

The statement is true because the number of times a player will roll is usually much too small to guarantee that the die rolls themselves will (as a group) conform to expected statistical distribution. In this case, the player really can't control what comes up when, or even necessarily predict the likelihood of certain results within the game session.  So, we're left with a high degree of risk and the possibility that the numbers a player actually rolls would be impossible to achieve using the rules for a point-based resource. In other words, Roll a d6 1000 times, and the average result will be 3.5; but let a player roll that d6-cube 8 times in a session, and he could conceivably roll eight "1's" or eight "6's" in a row.

Now, about those diceless results... I'm no expert of d20, but I've heard about the "take 10" rule. If I have it right, the player starts with a 10 point base, which represents the average die roll of a d20. Then, he gets a 10 point pool to manipulate the results. On each roll, he can add or subtract points from the pool to determine his "die roll". Since this is a resource-based diceless means of resolution, I maintain that this changes the player's expectations. The use of a diceless resource is not about gambling or risk; it's about control and self-determination. The value of the "take 10" rule is that the player, if he plays carefully, can live blissfully free of the dreaded "I rolled a 2" fumble, and that's just fine.  

Now, when discussing this sort of mechanic, xiombarg suggests allocating the points before knowing how tough the task is. MJ in his example suggests keeping the obstacle (target number) in the diceless game secret. This works fine when a character is locked in conflict with another character (or important npc). The player still has his wits about him. He knows what constitutes a good result and what constitutes a bad result. In the diceless scenario, he controls his results and that control strongly affects the outcome (whether the character wins or not). For other situations, this is still an acceptable approach, but I think it could be abused as a ploy to make the player waste points. I think the player should have at least some idea about how tough the task is so that the player can make an intelligent decision about how much effort to put into accomplishing the thing. The GM doesn't necessarily have to reveal the actual target number, but I think the GM should give the player a hint.

OTOH, if the GM does simply give the player the TN, this may not hurt anything. Simon was talking about narration of outcomes. He asserted that "Merely replacing randomisers with a resource allocation system still leaves a game in which game mechanics controll outcomes, rather than game narrative." Well, yes and no. "Yes" because game mechanics do still determine the outcome; but "no" because in the diceless version, the player is determining the outcome. The player is making an active decision about the outcome of every action, and that's a lot different than just rolling dice. If a designer really wants narrative to determine the outcome, then dice and numbers may both become irrelevant. That designer is possibly better off looking at a game like John Tynes' Puppetland or something and finding a way to make a game that is based primarily on player assertion, not numerical mechanics.

From memory the Eden Studios unisystem was designed to be diced or dicless by just using averages. So the damage for a weapon in their book would be listed as 1d6[4], for example, meaning to roll 1d6 or just use the rounded-up average of 4, etc.

Brian.

Just a note: Don't consider that I've closed the thread, since it's clearly a subject of ongoing and interesting discussion.  In fact, some responses to Hunter,

Quote from: whoIMO, people who are considering designs that combine dice and dicelessness ... need to look at the topic more formally and consider the underlying reasons for each decision.

Dice serve a function. ...
Resource-based dicelessness also serves a function. ...

Now, recognizing that these two options serve different functions, it seems to me that a designer must make some self-evident and different assumptions about using each one. When the designer chooses dice, the designer is choosing an element of risk, an element of surprise, a desire to gamble, and a desire to accept the unknown as part of the equation. When a designer chooses dicelessness, the designer is choosing player control and a right to self-determination.

With this in mind, I assert that mixing dice and dicelessness (die roll + resource in the same mechanic) is a poor design decision because it muddies the water. If you roll dice, you agree to take a risk. At that point, spending resources just dulls the sense of risk. If you spend resources, you are demanding self-determination. At that point, rolling dice merely diminishes the player's degree of control.

...

Now, about those diceless results... I'm no expert of d20, but I've heard about the "take 10" rule. If I have it right, the player starts with a 10 point base, which represents the average die roll of a d20. Then, he gets a 10 point pool to manipulate the results. On each roll, he can add or subtract points from the pool to determine his "die roll". Since this is a resource-based diceless means of resolution, I maintain that this changes the player's expectations. The use of a diceless resource is not about gambling or risk; it's about control and self-determination. The value of the "take 10" rule is that the player, if he plays carefully, can live blissfully free of the dreaded "I rolled a 2" fumble, and that's just fine.  

Quote heavily snipped

In response to this all:
So clearly, if you're going to choose a mechanism that allows you to choose dice or diceless as the situation demands it, you are compelled to make it clear what this means.  Are you going to give your action to the winds of chance, or are you going to take responsibility and control over your action?  This could be an interesting thematic issue...

As for taking 10... in DnD3.0 you don't get to vary your result, you get to take 10.  Maybe there's some variant where this resource pool happens, but baseline didn't have such a thing.  Perhaps you're thinnking of some other branch that does.