This is a follow up to the original thread regarding my game in progress: Silent Sound.  The original thread can be found here:

http://www.indie-rpgs.com/forum/index.php?topic=27077.0

Last night we played the first session of Silent Sound.  It went well in terms of the fiction and in terms of productive play testing.  It didn’t go so well in terms of the mechanics doing their intended job.  Such is the perils of trying to move a game from Alpha through Play-testing.

First off I made a GM blunder in terms of starting the game off.  Part of the intention of the design is to get the characters to interact with reflections not only of their own crimes but of the other character’s crimes as all.  The idea is to get a cross pollination of behavior across the characters.  I think my situation setup is fine in creating an interlocking web of characters that have overlap between elements of the character’s crimes.

The problem is, that in order to foster a creepier or more subtle tone I started off two of the characters encountering material relevant to their Memory rather than their Lures.  It turns out that Memory and Lure do NOT carry equal resonating weight with the players.  It was suggested that I reverse my thinking.  My attempt at “subtle” instead turned into the players feeling a little lost and disoriented.  I should hit the players immediately with stuff relevant to their Lure and their own crimes and then introduce elements form their Memory to draw them into the cross pollination.  Fortunately, I think situation I created is fine for that I just have to use it differently.  As I put it last night, “The instrument isn’t broken; I just have to play different notes.”

The mechanics on the other hand are not working as intended at all.  They aren’t driving the fiction forward in any way.  If anything they’re grinding it to a halt.  Colin felt the mechanics actively support turtle-ing.  I’m not sure I agree fully with that but what I did observe was that the players were very eager to engage certain aspects of the mechanics much earlier than were “necessary.”  Engaging with those mechanics carry a risk and thus the players felt they were being punished for engaging in something they thought was cool.

In particular I’m referring to the Flashback mechanic.  In the game you can frame Flashback scenes to situations involving your crime.  Mechanically this allows you to increase your base stats.  What you’re risking is an increase to your Shadow score which is bad for you come endgame.  The intended dynamic was that the GM would hammer the players with opposition which decreases their base stats thus *forcing* them into Flashbacks to recover.

Here’s the thing, doing that would take probably six or seven scenes and the players, fictionally, wanted to introduce flashbacks much, much sooner.  Hence they seemed like a punishment because now they were risking Shadow for no reason other than introducing cool content.  I have to think about this because I like the idea of a reflective “refresh” but if that’s the case then stats need to go “down” much, much faster or I have to decouple Flashbacks from “healing.”

Possible Solution: Instead of being fixed numbers, players spend their base stats like a resource.

With regards to the GM providing opposition the system has a serious problem here.  There are two scores called Town Influence and Guilt.  These two scores specify the base number of dice the GM rolls in a conflict.  These numbers go up and down based on conflict outcomes.  From day one of my design I saw the problem of the game becoming a death spiral for the GM so I added this mechanic called the Shadow Pool.  The Shadow Pool is basically a resource the GM spends to add dice to his pool.

I’ve NEVER liked the Shadow Pool.  First of all, it’s always been a tacked on hack to prevent the death spiral.  Second of all, I’ve never really known how large to initially set it or a good way to refresh it.  If I could, I’d get rid of it and last night it became apparent that I HAVE to get rid of it.

Here’s the thing, it is providing one thing that isn’t just about the death spiral hack.  I like to be able to lend “emotional weight” to my die rolls.  The limited resource allows me to demonstrate how much something is “worth” to me mechanically.  Even if I fix the death spiral problem I’m not really sure I’d be satisfied with a static number that I had no way to influence.

Possible Solution: The GM can just add dice to his pool, but has to give the opposing player Judgment on a 1-to-1 basis.  Judgment is currently a resource players can spend to influence the outcome of other player’s characters.

Another observation is that the endgame is WAY too compelling.  The game has epilogue constraints based on the outcome of the final conflict die roll.  I don’t mind glancing at the state of the mechanics to decide when to shift tone or direction or how to play your next scene to setup something else.  But the players were looking at the system in a way that looked like eight or nine moves ahead like a chess match and that’s getting too “gamey” for my tastes.  I want mechanical choices to be about the state of the fiction and what the player wants to express in the short term.

This is largely because the system as currently formulated is a giant machine which is clearly too complex.  In fact, I’m pretty sure it computes logarithms as a side effect of play.  I need to reformulate the machine so that (a) the long term effects are still present and visible but are a back burner concern and (b) the immediate short term decisions are more interesting.

Part of the problem is that there are a lot of levers that I deliberately put in the player’s hands because as GM I wanted to focus on playing the components of the town and not have to worry about certain pacing concerns.  The problem is that these levers when pulled set off a series of gears that have layered consequences.  It was my intention that the levers were to be pulled based on the immediate emotional desires of the players and that the machine would happily do its job unregarded.  Turns out this is a little bit like me placing the lever of a guillotine into someone’s hand and saying, “Pull this when you fall in love.”  Their rather natural response is, “But it will cut my head off.”  And me saying, “Never mind that, just pull it when you fall in love” probably isn’t going to cut it.

There are NUMEROUS examples of this through out the system and is probably the biggest design flaw.  The largest is the way Shadow works.  You don’t want Shadow because it reflects on your character badly in the endgame.  There are two ways to gain Shadow and both are entirely voluntary.  The first is the Flashback mechanic I described earlier.  The second way is by taking on Shadow to earn bonus dice on conflicts while in the Shadow world.

This is where Colin’s “turtle-ing” comment is probably most appropriate.  I assumed that the conflicts at hand would eventually take on such importance to the players that doing these voluntary things would become worth it.  Turns out they don’t.  Or at least don’t fast enough (again see my problems with pacing and opposition above).

Possible Solution: I have been considering swapping the labels of the GM stat Guilt and the player stat Shadow for some time because that more accurately reflects their function.  I mentioned this and it was suggested that rather than having Shadow thrust upon the player through opposition (which was the intent) make Guilt something the player starts with and has to proactively “burn off.”  I like this idea.

It goes along with that discussion in the first thread about whether or not fictionally it’s important for the characters to feel guilty about their own crimes.  I realize that I setup the whole Shadow mechanic with the assumption that the character’s had largely forgotten about their crimes and now Silent Sound has shown up to “remind” them by thrusting it in their face.

Oddly, I think this comes from me misinterpreting my source material, the video game Silent Hill 2.  In the game James’s crime is a reveal that happens about 3/4th of the way through the story.  For some reason I had gotten it into my head that James is in denial about his crime.  That effectively he blocked out what he did and the ordeal of Silent Hill draws it out of him.

But that’s not right.  I’m mistaking the audience experience of not being fully in the know regarding James’s motives with James himself not being in the know about his motives.  I’m further mistaking his eventual admission with remembrance.  On reflection it’s clear that James’s guilt is driving him the whole time.  I think that confusion comes from the weird identity confusion that can happen in third person video games.  If I don’t know something James must not know something because I am James, right?

In any event, I include that as a note to other designers basing games on the thematics of established source material.  Make sure you really understand the thematics of your source material.

The group is taking a two week hiatus to give me time to think all of this over and retool the system and then we’re going to take another stab at it.