Forum changes: Editing of posts has been turned off until further notice.

Main Menu

New 3D Model

Started by Mike Holmes, August 09, 2004, 01:13:09 PM

Previous topic - Next topic

Mike Holmes

GNS has limitations, I think we'd all agree. One of the limitations is that it doesn't address the concept of player control, other than to say that without control you have simulationism. This leaves a bad taste in some people's mouths, because they see this as making sim a dumping ground. So, what if we put control into the model as it's own dimension? I think this would take care of a lot of concerns.

As a note, I'll bet this has already been proposed somewhere before (and I probably argued against it then).

Backing up to GDS, we have three goals of play. Roughly player challenge, theme, and internal causality respectively. Now, in GNS if you want theme, but the GM makes it, you have a form of Sim. How about a six mode model that takes into account player power as tangential to the desired product of play? Note when I say "low" player power, I could say High GM power, or something else - but any way that I say it, realize that it's not being judgemental about the modes in question. I think all six can be quite functional.

So, you'd have:
Challenge high player control
Challenge low player control
Theme high player control
Theme low player control
Internal Causality high player control
Internal Causality low player control

I really want to stay away from labeling these with currently used terms (to avoid exacerbating the GNS/GDS problem). So, for the moment, let's use GDS and H and L for power split.

What do these equate to in terms of previous analyses?

GH seems like classic Gamism at first, but GL could actually account for more Gamism than is realized. I mean, in D&D, for instance, the GM sets up the encounters so that they're on parity with the party. Does this really give the player the chance to test his abilities? In fact, it seems that the GM is enabled to make things "close" in most of these cases for drama's sake. This is an interesting fact. In "GH" a player would really have more ability to set up their own encounters, and "play the game," overall. Think more like a boardgame, where the GM is only a referee.

DH is narrativism, effectively. That is, the goal is theme, and the player is empowered to create it. DL was thrown into Sim, because the player didn't have the power to do one of the two other things. But here we see that the player can still want to have theme as the outcome, but may just want the GM to provide it. I have a strong urge to call this "Storytellerism," both for the game system, and for the nature of the product of play.

SH is what I've refered to as Open Sim (and also, probably, Virtualism), and what people are claiming is problematic with SL. By looking at it with this model, we can see where any problems between the two might occur. SL contains the mode that Participationism techniques are used to create, but also other forms of Illusionism (the latter intended to make SL look like SH, and have the advantages of each). The assumption, interestingly, is that in participationism and Illusionism, the GM's job is to keep things "Storylike" and to that extent is about creating theme.

See a pattern here? In the high versions, the expectation is that no real theme will get created, except for DH where the players are fully empowered to do it themselves. In the low power versions, the idea is to make it so that while the players do their thing as minor participants, the GM is simultaneously empowered to create theme overall (or, at the very least, Drama, if you go by the narrativism assumption that theme can only be created by the player).

This, by itself, creates an interesting model, I think. I think it does away with all of the problems created by player control being an issue with the other models (which I think are most), by addressing it directly. Note that, in terms of a spatial model, this takes the normal triangular GDS space, and makes a "prism" of it, with complete GM control at one of the flat ends, and complete player empowerment (with perhaps no GM at all) on the other end.

What's interesting is that I think it's accurate in some ways for play as it currently exists. In most of the texts there's something about how it's the GM's job to make things interesting. This is encouraging low player power, and the GM being in charge of making sure that theme is created. It would theoretically be possible, however, for the GM to be in charge of creating, say, challenge himself. That is, instead of creating challenge for the players, he'd create challenge for himself. In fact, when the design is effective PvGM, this is sorta the case. There's the question of whether or not the player is empowered still, making the possibilites multivariate. Very complicated. Fortunately, at this point, I think this is a small part of play.

Another advantage of this sort of play is that it gets rid of the perceptual problem that I think exists about what the GM does in play vs what the players do. That is, I think that a GM and players can employ narrativism, but nobody would claim that the results look the same. This model allows us to consider the difference, again. Thus Illusionism as a technique is seen as the GM's side (player accepted for coherent forms) of the equation. It's what he does to achieve the SH/SL feel for the players.


Member of Indie Netgaming
-Get your indie game fix online.

Hunter Logan

I think your shorthand should be CTI and HL where Challenge maps approximately to G; Theme to D/N depending on HL; and Internal Causality to S. Initial impression: I like it.

Andrew Morris

Okay, this makes perfect sense to me, whereas I've been struggling with GNS ever since I came to the Forge. I don't have anything to add, just wanted to say it sounds good from where I'm sitting.
Download: Unistat



I see similarities in this to GENder.  The C, T, and I are top tier concerns, while H and L are bottom tier.


Interesting. I was going here myself in a sense. I've always felt that there's an invisible "slidebar" on some sort of control axis that gets slid one way or the other without being acknowledged.

I think someone suggested that a term for this be fidelity. See, a sim game (I'm assuming my most recently posted game in Actual Play would be judged to be essentially Sim) may not have a lot of "control" in the player's hands--it's not "here's a map, there you go." There are forces at work that they'll highly probably want to deal with (but are not forced to deal with).

Neither is a scenario that looks anything like the set up for Lord of the Rings (the dark army is going to march and win--but you have Sauron's achilies heel in your pocket).

In this case the GM's commitment is to fidelity (what-if) but the level of player control over the world is limited (Frodo can chuck the ring in a lake and go and open a tobacco shop and the GM and players can go through with that for a while--until the orcs come--but it'd be hard to explain how that happens in an actual, functional game).

So I might either create another axis of player input or just rename player control with "group commitment to fidelity."

You still get the same groupings but depending on the role of the GM the open-sim section may be a little less open ... and the high-fidelity Gamism section may turn out to be horribly unbalanced (the players go somewhere they're really not ready for and the GM lets the slaughter go down).

It's unclear what low-fidelity means with respect to theme (which is not a fidelity issue) but you could replace theme with Premise and see if that works any better (I'm not sure it does).

But, yeah, I think that axis exists and doesn't get sufficient mention most of the time.

I like it.
JAGS (Just Another Gaming System)
a free, high-quality, universal system at:
Just Released: JAGS Wonderland

Christoph Boeckle

I'm only just starting to get into GNS, but your new dimension makes great sense to me.

It could also be used to describe linear vs "open" scenarios.

For example, I ran a game where the characters are patients in some weird asylum. All I told them in the first session was that they woke up as complete amnesiacs. From there on, they explored the asylum, getting to know who the personnel was, what schedules had to be respected (and what happened when they didn't), etc.
They had high Internal Causality control, if I understand your idea correctly.

I believe it's a quite important part of the Creative Agenda (correct me if I'm getting this wrong, I still feel awkward with some terms) to make it clear how big player control is going to be. Lots of frustration can arise from there (it's also an impression I get when reading other forums).

I also came up with another thought:
The models are going to evolve quite surely, with different axes being added as new aspects of game are being identified.
And I'm quite sure as well that some matches will produce inherently flawed game types, just as some elements aren't stable on Mendeleiev's periodic chart.
I was wondering if you would find an example for all 6 game types you just defined* (because what is a model worth if it doesn't descrinbe something real?), but finally came to think that either it would be a new game type to try out (often models predict things that have not yet been discovered), or as I said in the last paragraph, be a flawed game type (and there's nothing wrong with that, imho).

*and for that matter: all 21 game types someone else might come up with

Hope this can be of interest to someone ;)


Hi Mike -

This looks basically sound to me. You make a minor error in discussing D&D when you write "I mean, in D&D, for instance, the GM sets up the encounters so that they're on parity with the party." This is a big bone of contention among some members of the O/AD&D community, actually. I come from a school that says you challenge the players, not the characters - meaning you might give them a totally unbalanced encounter and expect them to either figure out to run away or come up with some brilliant way of beating it - making sure that you gave them the tools to do one or the other up front, at least in a subtle way.

But the minor error gives way to a much more important truth: that GL is in fact much more common among roleplayers and much more commonly facilitated by gamist designs. You only have to sit down to a game of Great Ork Gods, which is solidly GH, to appreciate this.

I wasn't able to make peace with what some other people on this board say about Narrativism until I started to regard DL as Narrativist play. Many earlier discussions on this board do suggest a confusion between these two things though. 'Narrativism' sometimes means DH and sometimes means DH and DL in past discussions.

So for the past couple of months at least (since those 'social mode' threads) I've come to see GNS along the lines of your CTI and treat player control as a totally separate axis. What you control is not what you're getting a charge out of and many people can get charges out of things they have little or no control over. There's this minor art form called 'literature' that speaks pretty highly to that possibility, I think.

Jack Aidley

I like it. I do echo the Hunter Logan's request to use CTI instead of GDS though.
- Jack Aidley, Great Ork Gods, Iron Game Chef (Fantasy): Chanter

Emily Care

Quote from: Jack AidleyI like it. I do echo the Hunter Logan's request to use CTI instead of GDS though.

Hear, hear.  This is very useful, Mike (and getting away from the gns etc letters would be helpful).  I'd tend to think of the high-low player control split as centralized vs. decentralized. But that's me.

Koti ei ole koti ilman saunaa.

Black & Green Games


I agree that using the term GDS is probably a prelude to massive confusion, but I think we need to answer another question before switching to CTI: namely, is this a more refined analysis of GNS, in which case it might make the most sense to keep the old letters while remembering these insights, or is this a new if related theory, in which case a switch to all-new letters might be a propos? I was pushing for the former with my post, but maybe that's not right.

Emily Care

The advantage to using gds is that people (generall) already know what it means. However, is that really what's implied by the theory? Is step on up part of the G axis here, or solely challenge?  Likewise is there more to D that is not implied by theme? IC and sim?  

In fact, it seems that this model takes certain parts of gds and creates a structure out of them.  It may be more accurate to use cti.
Koti ei ole koti ilman saunaa.

Black & Green Games

Mike Holmes

Emily, Centralized/'Decentralized is, indeed a more correct generalization. And CTI is better. So to recontitute the chart (with examples for Christof that are likely to be disagreed upon as are all such model assignments):

    [*]CD - Player Challenge goal with Decentralized control (Great Ork Gods - thanks for the example, Sean)
    [*]CC - Player Challenge goal with Centralized control (D&D)
    [*]TD - Thematic goal with Decentralized control (Sorcerer)
    [*]TC - Thematic goal with Centralized control (Pendragon)
    [*]ID - Immersion goal with Decentralized control (No game that I can think of specifically has this as part of the design - at best, GURPS and such might be played this way without problem)
    [*]IC - Immersion goal with Centralized control (CoC)[/list:u]

    Christophe, yes, "linear" adventures are a great example of design requiring a relatively high level of GM control. And, yes, I agree about the player control thing being big. GNS makes a big deal of this, but by making it part of one axis, it makes it harder to get a grasp on. Maybe.

    Sometimes it makes sense for a model to evolve, and other times it makes sense to merely replace the old model. So I can't say that more axes will make sense here. But who knows. In any case, this is a new enough model that the new terms are merited, I think.

    Quote from: SeanSo for the past couple of months at least (since those 'social mode' threads) I've come to see GNS along the lines of your CTI and treat player control as a totally separate axis. What you control is not what you're getting a charge out of and many people can get charges out of things they have little or no control over. There's this minor art form called 'literature' that speaks pretty highly to that possibility, I think.
    Hmmm. I think this is sorta true, which is problematic again in terms of crosses. But I think some are incoherent. Basically there are three goals and four control possibilities. For at least 12 types. Hmmm. Examples:

    Immersion Goal High Player Control of Immersion
    Immersion Goal High Player Control of Challenge
    Immersion Goal High Player Control of Theme
    Immersion Goal Low Player Control

    Note that given that the Goal is what the players are seeking, it's my contention that the first and last are fine but the middle to are probably incoherent. Meaning that they're likely to produce play that doesn't match the goal. In fact, this is an easy way to look at incoherence in design, is the design giving the power to control some element that's not a goal.

    I don't even want to get into hybrids here, yet. :-)

    Marco, heh, fidelity was my term for the "sim" axis in the Beeg Horseshoe theory. It seems like you're using it for something else here, however. Fidelity in terms of Genre Expectations and drama. But, yeah, in general that seems to be what it's about. That's usually the goal. At this point I'm not sure that there aren't other reasons to give the GM more power - I think there probably are. So I'll leave it at power for now, and we can speculate about what it's used for.

    In any case, I don't think that there is a Low Fidelity as an opposite (high player power isn't low fidelity). That is, it's merely a question of who's expected to require it. The question of how much is, again, I think extraneous, and will vary from group to group. Even a GM with power to do so will only stress it so much. Complete control by the GM doesn't neccessarily mean reading LOTR, but telling any story he likes.

    Or do you see this as a third axis? That I could see. But, again, not as a primary mode thing. That is, I think fidelity is very fluid, and not likely to cause automatic conflicts of interest.

    It is somewhat like GENder in some ways, but with a cross instead of tiers.

    As for Theme, would you suggest Drama, instead? Or something else? Hmm. I'm starting to wonder if I shouldn't use more specific terms for these so we don't fall into the trap of "that's not what's meant by theme when real people use it." Any technical suggestions?

    Member of Indie Netgaming
    -Get your indie game fix online.

    Emily Care

    Great, Mike.  But, are you sure you want to go with immersion rather than internal causality?  Sounds problematic. I think you were spot on with IC.

    Marco's fidelity sounds like my take on verisimilitude--fidelity to a specific referent (a genre, text, etc).  This is similar to baseline--as in "realism" as an underlying assumption (which is basically just a form of this "fidelity" with the real world as the referent).

    I'm not having any brainstorms about theme. Drama seems equally loaded. Same for narrative. Meaning would be too much, but that's often how I think of it. Diegesis would be accurate but way too didactic.

    QuoteMeaning that they're likely to produce play that doesn't match the goal. In fact, this is an easy way to look at incoherence in design, is the design giving the power to control some element that's not a goal.
    Think about that folks. Wow.
    Koti ei ole koti ilman saunaa.

    Black & Green Games

    Hunter Logan

    IMO, new terminology is long overdue. The overlap between GDS and GNS caused significant confusion and difficulty. GENder, for all the good things that Scarlet Jester brought to the table, somewhat increases the confusion.

    Theme adequately describes an actual purpose of play while neatly avoiding the D/N trap.

    I suggest Immersion is best considered as a technique, not a goal for play. Immersion can be used to some extent in any other mode. Internal Causality works as a goal for play. It indicates a desire to make in-game actions and outcomes conform to the notion of "what should happen" given the known facts about the characters, their situation, and the game world rather than "what I want to happen" or "what makes the best story."  This is probably the most difficult and troublesome term to define and refine, because people have different notions about what constitutes internal causality and different thoughts about "what should happen."

    Adding more axes? No thanks. The original set (cti-hl) offers a very elegant way of looking at the situation. Resist the urge to overdevelop it.

    Ben Lehman

    So, in the first major challenge to the model -- what about genre pastiche?

    What I'm looking at here is something like Toon, Marvel Superheroes (chart edition), or vanilla Teenagers from Outer Space -- the basic point is not immersion, but creating something that is "just like" a particular genre, with the rules providing a strong emulative framework.  In the context of GNS, is this Sim.  In the context of GDS, I'm pretty sure it's D.  In the context of CTI/DC is this T or I?  It doesn't really seem like either, to me.