When I was stuck in Korea last summer, I had come up with an idea for a game about dreams and psychological healing that I finally got around to committing to words. I admittedly was majorly inspired by Inception -- the idea of operating as agents in someone's dream was intriguing. However, the main catch of the game is that the player characters are dream-constructs that do not really exist outside the dreams.

The main conflict of the game would be in how they deal with the tiny universe within someone's head that they are stuck in. The way I'm seeing it now, there would be a higher level conflict in resolving the dreamer's mental/emotional trouble that gave rise to this phenomenon in the first place, with their repressed urges, unconscious agendas, emotional backlashes and whatnot. Because of the ephemeral, non-concrete nature of the game's setting, I would like to make this game as rules-light as possible -- with immediate conflicts being resolved via the players' creativity guided by a handful of stats. If the player characters are able to warp the (semi)reality around them, it would seem silly for their actions to depend on simple probability.

However, it is true that dreams are often random and chaotic and the system would be better for reflecting this. Of course, I'm at a loss -- trying to find a randomizer that is not as concrete as numerical dice yet provides enough wiggle room to simulate the unpredictability of dreams is harder than I wanted it to be. :(

If I may ask, what would be an ideal solution for this? I'm looking for a randomizer (I think there was a Greek-sounding word for this...) that can reflect the capricious nature of dream reality. One idea I had was Rorschach tests -- the player whose actions are being tested will attempt to guess what the Dreamer (with GM acting in character) sees from the blotches. It entertained me at first for a while since it seemingly captured the idea of psychological randomness I was hoping for... But the fact is that there's still too much ambiguity. If the GM really wanted to punt the player in one way or another, they could simply decide that the player's guess was or wasn't close enough for the PC's action to pass the testing.

There was also that it would be really messy to do in actual tabletop play. :[

Blimey. It doesn't seem that I'm able to edit my previous post...

As the rules thread asked, here's a link to the page for Waking to Dreams -- as of 04/18/11, this contains only a bare fragment of what I've toyed with in the last 9 months. Being as crunch-weak as I am, the actual rules part will more than likely change quite violently. :|

There is a game system that came out a while back known as MURPG.  It's a superhero system, originally, that used what was an "Action Resolution" system.   Where the character had a pool of energy and could put as much energy into a single task as their skill in it.    it removes the randomness from the situation and bases it solely on ability and willingness to complete the task.  

the system died a long time ago and after talking to the company they don't mind me making it more generic and "rereleasing" it as a free product so I've been working on that, rewriting and using it in many different genres,  So far i've done Fantasy and Anime with great success.  here is the basics.

http://www.facebook.com/home.php?sk=group_122324284490415&view=doc&id=161300523926124

Now that you've pointed me back to entirely diceless, I've got an idea I'm still kicking around...

I don't know how effective this will be, but could the source of randomness and unexpected results come from the player's lack of knowledge about their characters' exact capacities?

With how I see the game right now, the stats of the characters will be something that changes every now and then, midst the scene and action. And while the players can definitely lead the character's immediate development to a direction or another, how the GM interprets the character is played will play a role in how the stats -- and what they can safely allow -- will change as well.

I'm thinking that this could work because low stats don't stop anyone from doing anything -- it just means that doing so will be riskier. If the players had direct access to the character sheets in a system as simple as I see it (basically 7 stats total with no resource system), the risk might not mean much. Though if the players did not know how much their characters' sheets changed since the last time they saw it, they might be motivated to take some risk (since it's the only real way for the characters to grow, narrative-wise and stat-wise)...

...but, of course, this might not make any sense when I'm not in my mind. Help? D:

Hi, sz!  Welcome to the forums.  :)

szp wrote:
However, it is true that dreams are often random and chaotic and the system would be better for reflecting this. Of course, I'm at a loss -- trying to find a randomizer that is not as concrete as numerical dice yet provides enough wiggle room to simulate the unpredictability of dreams is harder than I wanted it to be. :(

If I may ask, what would be an ideal solution for this? I'm looking for a randomizer (I think there was a Greek-sounding word for this...) that can reflect the capricious nature of dream reality. One idea I had was Rorschach tests -- the player whose actions are being tested will attempt to guess what the Dreamer (with GM acting in character) sees from the blotches. It entertained me at first for a while since it seemingly captured the idea of psychological randomness I was hoping for... But the fact is that there's still too much ambiguity. If the GM really wanted to punt the player in one way or another, they could simply decide that the player's guess was or wasn't close enough for the PC's action to pass the testing.

Sorry about the confusion! I'm still naive to the terminology regarding game design... The fact that everyone has their own definition, as was stated in the rules thread, doesn't help either, I guess.

What you have is about right -- with the critical point being that Rorschach blotches are too open to interpretation. Originally, I was hoping for something like Jenga in Dread, though with something to do with how mutable dreams are like...

About the mechanics: It's something I'm still not too clear about! :D What I have in mind right now is that Ego is expressed by the nature of a character's being -- Marty Dreamcakes can punch the moon out of the night sky because he is that strong. With Drive, something similar could be done, of course. He could bat the sun out of the midday sky with a baseball bat, but this time because he can swing that well. With how I see it, there would be a lot of overlap -- but that's somewhat intentional. The stats should fordef be fine-tuned, but I'm wanting to give the player's creativity the upper hand in what their character can do, as opposed to the statistic scope and range of their abilities. (if that makes any sense... I myself haven't slept for longer than an hour at once since the week began, thanks to school. <_<)

Hmm. Another way to put it would be that Ego is self-concept and Drive is power-concept -- literally. Actions driven by Ego is powered by the character's confidence in their nature, while those by Drive is powered by the character's confidence in their abilities. Counteracting Ego would require somehow proving they are not what they think they are, which would usually be more difficult than proving that they can't do what they think they think they can.

...I'm not sure if either of these explanations make any sense. But this so far is how I understand the difference myself.

Re ego vs. drive: I personally still think the distinction is hair-splitting, but you've explained your position better and I can see where you're coming from.  Thanks!  I think you'll be well served to draw a clear mechanical distinction between them, if you keep them apart.

Re the resolution mechanic I hinted at: have you considered Tarot cards?  Each draw gives you not only a number, but also imagery/symbolism/interpretation, and a way to swing the results dramatically.  As a simple starter suggestion, maybe you have to narrate in a theme from the card to alter the base success/failure the numbers determine?  (If you draw a 10 of Swords, that 10 is numerically going to kick ass, but clearly the card itself implies a huge sacrifice or reversal ...) 

I also think a card-based resolution mechanic might be best for a system like this. If you are not afraid of putting in a some effort to the actual system of randomization, I wouldn't recommend using the Tarot Cards. Instead, you could come up with a deck of cards yourself. Despite the fact that Tarot Cards have an inherent aura of coolness, they are actually somewhat disappointing (in my opinion) when compared to the mystical image that is attached to them. They is also a little difficult to use as an effective randomizer since there are only 22 Major Arcana cards (the cards most people think about when they talk about Tarots) and the symbolism behind some of the cards is quite unintuitive or otherwise difficult to implement into a game like this. If you create the cards yourself, you could actually make the system fit your needs instead of forcing the system fit the Tarot Cards. Here's an example:

The "suit" of the card could determine an abstract theme of the card, for example dividing the cards to seasons (Spring, Summer, Autumn, Winter) or moods of the dream (Fear, Anger, Desire, Sorrow) or Elements (Earth, Wind, Fire, Water) or whatever.

The "face value" of the card could determine a more concrete theme that is divided into several sub-themes according to the suit of the card, for example:
1 = Time (Past, Present, Future, Eternity)
2 = Person (Adversary, Lover, Friend, Master)
3 = Element (Tool, Obstacle, Aid, Distraction)
4 = Fate (Doom, Victory, Loss, Death)
5 = Location (Home, Wilderness, Paradise, Hell)
etc. etc.

These themes might be too concrete, so you could use more abstract themes similar to Tarot Cards (Cards like Tower, Flower, Sea or Diamond could mean a lot of things, but contain some outlines to what the theme of the Dream is)

The cards could also have a number value similar to normal playing cards that could be used as a number randomizer whenever something like that is needed.
The system could work like this: whenever you start a scene, the GM draws a certain number of cards that determine the elements of the dream and the players or the GM determine what those cards actually mean. And whenever a more simple randomizer is needed the number values of the cards are used (whenever you would normally roll a die, you draw a card).

For example, the characters follow someone into a new segment of the Dream and the GM draws two cards and combines the themes to determine the nature of the Dream. The GM draws Death and Lover and determines that the characters appear on a bridge. As they continue their chase, the characters notice a woman standing on the side of the bridge, trying to gain the courage to jump to her death. One of the characters somehow recognizes the woman as a long lost lover (despite having no recollection of loving her before this encounter) and the characters have to decide what they want to do. Leaving the lover behind could have mechanical consequences (a loss of Ego for example), but letting the pursued NPC escape could have consequences to the overall story arc.

Of course, creating a whole set of cards would obviously require a lot of work (compared to using a Tarot deck or a normal deck of cards or just using dice). If using custom cards seems too troublesome, you can always use dice and tables just as well. (Cards are just cooler because they are physical things the players and the GM can fiddle with)

If you want to use stats like you described, one way of dividing them could be as follows:

Ego / Drive
Control / Vision

The stats on the top row represent things relating to the character. Stats on the bottom row represent the character's ability to influence the surrounding environment. Stats on the left column represent capabilities that are "acceptable" to the dream and are less effective. Stats on the right column represent capabilities that bend the Dream in ways that could cause a backlash from the dream (similar to what happens in Inception whenever the dream detects foreign elements within itself).

If you are open to other ideas, another way of thinking about stats could be to use something similar to the Ars Magica magic rules and have the characters possess skills like Create, Destroy, Change and Understand.

Overall, I like your concept and I hope this rambling gives you some ideas.

Have you ever looked into.  Universealice??? I think with some slight modifications.  Such as giving players a personal character who influences the world instead of a player being a storyboarder.
Buying potential based on ego/drive or control/vision. 
Main GM can have greater veto power based on dreamscape. 
I think it would give you more of the chaos factor your looking for. 

Dreams and their manipulation have always seemed like a fun idea, but I've never put my mind to something beyond that.  some creative re-working I think you could make it work quite easily.

Dang. I was gone from life for 5 days and that gave me a time to think things over.

On Ego & Drive -- Yeah, I did come to see that the distinction as is is just too vague. :s Still, I wanted there to be three concepts that provides foundation for a character. How I currently see it is that Ego covers self-statements where an adjective is the operative word and Drive covers those where a verb is the operative word. This is still a really vague distinction, but I think this reduces wiggle rooms enough -- the idea is that Ego is a more passive thing while Drive is active. Possessing Ego gives a psyche a "body"; Drive gives it energy to act; and Vision fuels it with willpower and want. That sort of thing.

Of course, I'll probably redefine this over and over again as I continue to add and refine stuff. Still, the criticisms given here helped me rethink this. :d

On randomizer -- You know, the cards idea really intrigues me. One idea I got right now is having a suit each for major primary emotions -- probably Fear, Sadness, Joy and Anger. I could tie this in with a concept I've refined over the downtime -- each point in a Confidence trait corresponds to a statement about the character, which amount to depict their self-image. Each of these Quality statements may be tagged with one of the four emotions. When the character enters into a conflict, a card from suit of the base emotion of the statement that dominates the triggering action is drawn. The "value" would be divided into a "mood map" combo, like "high energy-negative feeling" or "low energy-positive feeling". I guess that would mean a total of 16 cards... It could very well be replicated with 4 d4s, too.

I'd still have to think more about this, though. What I'm thinking is that whether Energy or Feeling resolves the conflict varies one case from another, though the other qualifies how that result was achieved (or perhaps whether the defending Quality becomes fragile -- a mechanic I'm still tinkering with.) Or, perhaps, it could be used to break a tie. In any case, I think this could provide a definite conflict resolution while also providing a narrative element... But there's probably something I haven't though about yet.

On resource-based system -- I had considered it, but I thought that it would require a system that isn't as fluid and mutable as the one I'm imagining at the moment. I expect stats to change every other action, depending on the actions of the characters. So, in a way, they already are more like temporary points than permanent stats in other games, I guess... Still, thanks for the reminder -- I'll revisit this when I change my mind about randomness.

Also: Thanks for all the comments! This is really helpful, especially since I'm not much used to designing games and such. :d

Going back to the Rorschach tests a bit, I'm wondering if they would still be viable as an influencing factor in given situations. Not having them guess what the dreamer is seeing, but having them act in character to say what they would see in the ink blot. This could then influence the nature of the action, or the power behind it. For example if one's character is angry, perhaps they are more likely to see something violent in the test, and even when trying to perform an unrelated action, that anger and perhaps violence might creep into it?

I'm pretty sure I'm not being as articulate as I want to be, so I'll leave it at that. Something really captures my imagination about what I've described, but it may be that it's entirely not what you're looking for.

The original idea with the Rorschach blotch was that the PCs being able to grasp what the Dreamer's mind thinks like allows them to have more influence on the dreams. I dropped this line of thought, though, because I felt that this was too vague to be a randomizer/conflict resolution mechanic...

After consideration, I actually decided to forgo random conflict resolution and use a point-based system similar to Nobilis -- ultimately, I did not like the idea of characters failing to do something, seeing that they are essentially reality warpers.

The current idea is this -- Whenever the characters take a special action (called Shaping), they must spend a certain amount of Clarity, which also happens to be their HP. Doing so absolutely creates the effect that the player/character desires, though this is not to say that they cannot be reverted or challenged indirectly. It is easier to manipulate elements that have not been introduced to the scene yet (it just happens to be that everyone forgot to notice the sword my character is holding) instead of changing something that already exists. The scope of the action characters may take is limited only by the resources the character is willing to devote to it. Effects that allows for more creativity (thus allowing the characters to bypass common sense) take more points of Clarity. The final cost is adjusted by traits the character possesses.

In combat or conflict situations, the end result of two diametrically opposed actions would depend on who was willing to spend more resources. While this could make combat simplistic, the scope of creativity the game promotes should help players to come up with actions that defeat the aggressive action without directly challenging it. Remember how we used to pretend-fight in playgrounds?

I've added on a more detailed description of this mechanic at the end of the wiki article (earlier portions need revision, though I'm waiting to do this until I'm certain I'm done), but they might need further clarification. What do you people think?