Wednesday, August 3, 2011

More about Transparency

What I Am Working On Right Now
I'm rationalizing the Powers armor list. It isn't that big but there will likely be some thought into how it is organized. I'll talk about that in a moment.

One note: I'd like to start cleaning up the web-site in conjunction with working on JAGS Revised Archetypes. One thing that has stopped me is lack of good web graphics collateral. I'm not sure where to find good, cheap web-style artists who can help me with (a) a revised logo and (b) some graphics of specific sizes and color content that will look good on the web. I realize I can do a lot of it myself (take a graphic, run some kind of "button-ize" filter, and then "save for web"). I'd rather have someone else do it for me.

Likewise, I'd like to update the logo but only have vague ideas about what to do with it. So that's holding me back.

You can see the original post here. Transparency is defined as sharing with the Players what the GM knows. Making all rolls "on the table" is an example of higher transparency. Keeping the effects of damage or how badly hurt an NPC is from the PCs (save for what the GM declares they can see) is an example of lower transparency.

We tend to favor higher transparency as a rule. Let's look at a specific example-case that makes it especially interesting.

Partial Coverage ADP
One rule that we're playing with right now is "partial coverage ablative damage points." In JAGS if a target has partial armor (like just a breast plate) it is assigned a 'coverage number'  (usually 3 or 4). If you hit by less than or equal to that amount you automatically go against the armor. If the roll to hit exceeds the coverage though, the attacker has a choice: apply the full amount you hit by to the armor and go against it or reduce the amount you hit by and hit around the armor, ignoring it.

Remeber: in JAGS, the more you make your to-hit roll by the more damage you tend to do. With penetrating weapons (like swords or bullets) this is even more dramatically the case.

Ablative Damage Points are a specific kind of defense: unlike "normal" Damage Points (DP) which everyone has, when you suffer ADP you don't ever have to make a wound effect roll (to see if you were Stunned, Dazed, knocked Unconscious, or worse). ADP, unlike armor, is reduced by every hit: if I started a Round with 20 ADP and got hit for 12 pts of damage, I finish the round with only 8 ADP remaining. If I then get hit for 10 damage, I'll lose all my remaining ADP and suffer 2 actual Damage Points worth of damage. ADP doesn't prevent "extremely high damage" from good hits with Penetration Damage attacks but it's a good way of having a character who can stand up to a lot of punishment even if they'll tend to drop more quickly once their ADP is gone (this is perfect, for example, for a lot of monsters or bosses).

When dealing with ADP with "partial coverage" it means that the ADP has a coverage number (let's say 4) and that if the attacker hits by 0-4, their blow is automatically applied to the ADP. If they hit by 5+, though, they can choose to hit around it.

The question is: what's the logic a Player (or the simulator) employs when deciding whether or not to hit around the ADP?

That depends on how much they know--far more than partial coverage Armor does. Here's why:

  1. ADP, unlike Armor doesn't prevent Penetration. When a PEN damage attack (like a sword) hits the target makes an Armor Save. If they succeed the damage is converted from dangerous PEN damage to far less dangerous Impact damage. Unless the attack totally eclipses the armor there is a pretty good chance of a save. With ADP,though, there isn't--so deciding whether to hit around it or not isn't such a no-brainer.
  2. If the attack will break through the ADP and do DP damage it may force a roll. Whether or not it forces a CON check to see if the target suffers debilitating wound effects will depend on how much damage you are doing, how tough the target is, how badly damaged they are, and whether or not the ADP will totally absorb the damage or not. This is a complex decision made harder if you don't know how much ADP the target has left.
  3. Good Hits with a PEN weapon tend to "double." If you hit by 4+ (A "good hit") with a PEN damage weapon your damage tends to be 2x the base damage of the weapon (a 9mm handgun has a base damage of 6--a "good shot with it" will tend to do about 12 points). If you "hit around" the ADP your attack will tend to hit by less than 4 (unless you hit very well, in which case you might otherwise see 3x or greater damage!) so it's a tough decision since the attack will likely Penetrate anyway.
So What Do You Know?
Usually in our games we don't tell the players exactly how many DP or how much ADP the target has (you can see it's badly damaged--but you may not get an exact count). On the other hamd, the simulator will 'know' so do we use that in the calculation?

Secondly, as some hits will probably hit around the ADP the target may have taken both DP and ADP damage while still having ADP left (if the ADP is standard it'll all be gone before the target takes 1pt of DP damage). In this case the questions will be things like "if I hit around, what are my chances of forcing a CON roll?" and "What are my chances of moving the target to a higher/worse condition category?" These are things the simulator can calculate exactly, a player might know to a degree, and an experienced player would probably have a better grasp of than an inexperienced player (based on what the target looks like, what the GM is like, etc.)

Do we make the simulator all-knowing? Canny? Random? What represents the true-value of partial coverage ADP?

Finally, there is a consideration that is even more complex: if you are fighting on a team your decision about what and how to attack can (and, in our experience, often is) based on what you know about your team-mates and the opposition--especially where they go in Initiative Order (it is more advantageous to attack enemies earlier in the order since if you can disable them, they can't attack and may give your peers a chance to attack rather than waste action points defending).

Whether to hit around ADP or not would count double here since ADP damage alone won't take anyone out--but if you hit around it and deal damage to DP you have at least a small chance of taking out any semi-reasonable target in one shot.

This, of course, gets even more complex when you add armor with partial coverage--but we're presently not trying to do that (the game will allow it--but I wouldn't go trying to write simulator logic for it).

It is Likely
It is likely that the 'correct' algorithm is something like "if the average hit around would force a CON roll but attacking the ADP would, on average, not, then hit around" but, as I've said, I don't know how likely the average player is to have the information to figure it (not to mention the ability: I don't want the cost to be based on your mathematical acumen).

The Question of Transparency
This brings us to the fundamental question about transparency: how would we recommend handling targets taking damage from a theoretical standpoint?

There is one example I'd like to look at since it's interesting: what would I do in a situation where a PC is encountering something unusual in the world and does damage to it. Imagine a mundane PC police officer who enters a building and comes face to face with a Chi fighter. He opens fire and hits twice--the Chi Fighter is either immune to bullets or otherwise highly resistant--but the character only sees a tattooed hugely built bare-chested man and the Player may have read the Chi rules but is unaware of the character's stats in the game.

What would we want the player/character to know?

Firstly I'd want to minimize the gap between Player and Character knowledge as much as possible. That's a personal preference but outside of serious "take me all the way out of the game" meta-gaming I don't have a particular soft-spot for watching people work on "not playing the knowledge." If the guy is literally immune to bullets I would tend towards saying that the bullets "seem to go right through him. He doesn't even blink."

If the guy is just hugely resistant to mundane damage (let's say Chi Fighters in this world have 100's of DP) then then I would want to describe the effects as minimal (blood like a pin-prick as the bullets seem to indent the skin but then fall away. He blinks--but that's it). Again: the idea is to impart information to the player and the PC that is in keeping with the game mechanics.

What if the situation is specific: the guy is immune to metal but not flesh or the night-stick? What if the immunity requires a "block" action to "absorb" the bullet's energy?" In these cases I see things as though it were in a movie: the director is spending money for the special effect so you'd want to show him sort of "catch the bullet" or maybe have the "metal pass through him as though he were a ghost to it."

But There's One Situation ...
However, the above being said, there's a situation where I would want to keep some information from the players: if the encounter was being played (by my side as the GM) as the initiation of a tension building scene where I wanted to imply there were bullet immune things but not spell out what they were or whether that was truly the case then I might employ less transparency.

I believe that would would be my right as the GM--up to and including: making rolls out of sight or forcing Perception checks to glean any information about the bullet impacts. I might, for example: not tell the Players what their negatives to hit were ("you see a silhouette on the fire-escape: either you take a super-quick shot or it's gone!"). I might even tell them, after the roll, that there was a negative that I'd applied due to the quickness of the shot.

This Is Outside The Rules
In JAGS there are no rules (currently) for mysterious negatives to hit. A Player will usually know if they hit or not--however, there is also no rule that says the GM must divulge all the modifiers to the Player. The book (so far as I know--and I wrote it but I don't remember every line) is silent on that. I think that's as it should be: ideally there is some good guidance and discussion on those points ... but failing that, I believe leaving it in the hands of the participants is the right way to go.

If the GM wants there to be a lack of clarity around whether or not a shot hit I think that all things being equal the GM should deserve (and therefore have earned) the Player's trust. If the relationship is dysfunctional (the GM is using the rule to punish a player) then I don't have a lot of respect for that and I wouldn't expect the Players to either.

What About Changing the Dice Rolls?
So what about, in conditions of low transparency, changing the dice rolls? Would I "cheat" and change a roll? The answer is almost universally no. It isn't universally 'no' because I've done it. The only situation I can remember was ignoring some rolls at the tail end of a long session where the PCs were battling radioactive zombies: they PCs had clearly won and I was interested in terminating the battle. Instead of just cutting to "you mop them up" I instead had the last two rolls "take out" the remaining zombies without checking or applying the rules.

I don't consider this a best practice but I did do it. I also don't consider it harmful: I'm pretty sure no one at the game would've complained if they'd known. It's even possible someone got some mild satisfaction out of it. I'm not sure.

One thing I wouldn't do is change the die rolls--even hidden ones--to meet my feelings about what a better story would be or to save or kill a PC. While I'm not actually absolutist about that--I simply am not enough of a theory zealot to hold an absolute opinion--I consider doing that to be a fairly weak approach to solving whatever problem I'm trying to solve and believe there are better tools ... and that having and allowing the negative outcome is preferable in the long term for everyone involved.

So while I might change the level of transparency I would not use a low level of transparency as a 'shield' behind which to manipulate the game rules at a meta-level.



  1. I'm not going to comment on the second half of your post because I'm close enough to in total agreement about,well, everything to be no nevermind. About the question of first half, though...

    It really does turn on how the game is going to assume ablative DP is going to be expressed, in-game. If its not something that its going to assume is really visible, then I'd either go with the simulator either being blind to it, or (if practical) slowly pay attention to it over iterations. I think that'll get closer to the reality of how the decision making will be made in the actual game, at least on the part of PCs.

  2. This is a very good point: exactly what /is/ ADP. The answer is that we have several different definitions of it and we're not going to specifically say which it has to be. There will be some discussion about this in the back of the book.

    The question is "how meta is your game going to be?" At some levels of meta, ADP can be mobility (in those cases you should lose all your ADP temporarily if you are immobilized or grappled). In other cases it'll actually be physical mass--just of a "different sort" than DP (i.e. for monsters).

    The same applies to negative Damage Modifiers (it could be magical resistance or hyper-defensive movement ... or luck ... etc.). We don't want to be fully prescriptive because /any/ of those answers could be legitimate for some game and for some players and it's worth the page of discussion for us to leave it up in the air.

    As for what we're doing with partial coverage ADP: we really kind of do know what it's worth from other experiments. At coverage 5+ it's almost worth the full amount. At Coverage 3 it's a little under half. As that applies to armor it should also apply to ADP (albeit to a slightly greater degree since even a very low-damage hit will degrade ADP).

    So while it's hard to be exact or, indeed, to know how players will react to it, I think our mental-math approach will really work pretty well.

    The post was generated by discussion about what the logic would be and me going "well ... that depends on what we expect the simulator to know ... you know, that's a really good question!"


  3. I can see the desire here, but its one of those things you need to make some kind of default assumption for if the cost is going to be right. An invisible defense, a gradually revealed defense, and a visible defense are just not going to be of the same average value; even if you assume the game permits some meta-knowledge, all that does is change why the assumption you decide on exists, not the fact you need to make some assumption.

    Its like you referenced in your most recent entry; unless you're in a situation like a lot of superhero games where everyone knows you're Superman and can bounce bullets, there's going to be a big advantage at first in being the guy who just has bulletproof skin and the guy who who grows a steel carapace. Until and if word gets around, a lot of opponents are going to waste time finding out a tactic is really bad on the former who'd take one look at the latter and let the guy with the rocket launcher deal with you.

    So you either need to deal with that issue somehow mechanically, or assume some level of default in evaluating it.