Mike's Standard Rant #7: Designing for Gamism Ain't Easy

[Mike Holmes]

Response 2 in series.

Can I just say I LOVE this sort of thing. I don't know where I am in terms of gamist play style, but I think Diablo II is pretty gamist. Now, in that it's quite easy to fall back every time you take a little damage, or just sit still and let regen heal you to full. But you don't, because its not fun. Then again, that's a sub optimal choice to go in without full health when its easy to get...and sub optimal choices aren't really gamist IMO. So where does that leave most Diablo II play?

There are days I love it, and days I hate it. I play Angbad and other games like that. There are moments when I am thrilled to have gotten ahead, and others where I die and lose my progress. Sometimes I play it easy and take the slow and steady rout. I'm glad to advance, but it's dull.

The thing is that it can be challenging constantly with the right tactical inputs. Even that simple orc should have some challenge in dealing with it correctly.

And I got the impression the idea is that while the PC's are away, the challenges can re-inforce as well, perhaps changing challenge types, etc (within a reasonable degree though). But its possible this has no effect on what your getting at.

Right. You can always make retreat a failure condition (the Princess gets sacrificed). But that doesn't make the tactics any more interesting.

Mike

[Mike Holmes]

Response 3 in series.

BL>  I think that the key here is that Gamism is about Challenge, rather than about Victory.  Observe the Go player who plays with a handicap to equal the playing field.

Diablo II has the added effect of trying to maximize benefit over time played, which is more complicated and not (I think) directly relevant to TRPG Gamism.

I think that games like Diablo and Angbad have advantages in the amount of play you can get through. The strategy becomes tactical if you will. This is why these games still exist (especially Angbad with no graphics), IMO. It's their advantage over tabletop games. That and the objectivity of the experience.

Mike

[Mike Holmes]

Response 4 in series.

Things that usually make for good gamist design include:
-Multiple strategies/tactics to success
-Chance for the sudden turnaround/comeback
-Significant Karma based mechanics
-Reward, reward, reward

Good points, Chris. The comeback one is hard to implement well. Even harder to make jibe with a sim side are "balancing" mechanics that give advantages to the side that's losing or handicap the side that's ahead. These are awesome mechanics, but I don't know that I've seen any in a RPG before.

Instead in designs, you tend to get what I refer to as "the Steamroller Effect". One side gets up a head of steam, and then they're impossible to stop. Take the popular game Warlords for instance. Once you have more Castles than your opponent you're going to win. Success reinforces success making the early success critical to the game. TROS suffers from this, sorta. Actually Jake makes it work by saying, in effect, the first hit wins mostly. This makes combat both realistic and short. Note that getting that first hit is the trick, hence why it's a good game.

Mike

[Mike Holmes]

Response 5 in series.

Martin, good explanation of Strategy and Tactics, thanks. Some would say that Strategies are a set of tactics that you choose to employ to make an overall plan.

Most Real Time Strategy games - especially Blizzard's herd of the StarCraft and WarCraft lines - exploit this mercilessly. And, on the extreme opposite end of the scale, most of the 1-on-1 fighting games (Street Fighter, Soul Calibur, et cetera) use a version of this as well (High Attack beats Low Block and so on).

I was just talking to a board game designer not too long ago, and he was lamenting that this was so common. We played the boardgame "Age of Mythology"in which this is just rife. Myth kills mortals, Heros kill Myths, and mortals kill heros, genarally. Some people think of this as a good thing, and it does mean that there are balanced considerations. But often too balanced. When this is the case the best play is random because the only way to win is to outguess your opponent. And if you use any predictable pattern, you are beatable. So, if all it promotes is random play, then it's not really that useful in making for tactical analysis.

The counter of this argument is that people claim to be able to predict other's actions, which therefore makes it a skill again. And round and around it goes.

Anyhow, RPS is one of the classic games of Game Theory because it promotes just this discussion.

Mike

[Valamir]

Take the popular game Warlords for instance. Once you have more Castles than your opponent you're going to win.

Not to be too tangental, but if you're referring to the PC game Warlords (I through IV) the steam roller effect only really applies to the end game where most conquest games that don't have alternate victory conditions break down.

For most of the game the steam roller effect is mitigated by the up keep cost to garrison all of those castles (or cities in later versions).  There is a signficant period of time in the mid game where he with fewer castles has an easier time concentrating his forces and can use this advantage to achieve victory.  In fact, growing too rapidly typically results in a paper empire that is easily overwhelmed by a smaller empire with better infrastructure (Warlords IV is the weakest of the games by far and doesn't do this nearly as well as II or III).

[Mike Holmes]

Response 6 in series.

Something occurs to me about a design element no one else has mentioned, but which you bring up numerous times: the players never know what the GM has up his sleeves in terms of resources, so it makes for unsatisfying Gamist play -- the challenge isn't up front, or rather, the use of tactics are purely shots in the dark for an overall strategy.

You have that right. In Diablo, for instance, you can sorta "feel" what's coming up. Things get gradually harder, and you know about what you need to proceed to the next level (though sometimes it's a shocker). In most RPGs, you can get some of this feel in terms of "level" and such. But what you can't usually know is how many "encounters" are left. Again, the best you can do is to assume that the GM made it a tight squeeze, which means that a standard strategy then forms, which means no decisions, etc, etc.

If that sounds right about what you're saying to you, I present that in ORX...

Slightly better, but you still don't know about the strategic situation, overall. What's the next confrontation going to look like. It's not really a lack of tactical information - players obtain that quickly enough by back calculation in most games (Hmmm, a fifteen hit, but a fourteen didn't = AC 2!). Again, it's "encounters". Is there a goblin in the last room, or a giant?

Mike

[Mike Holmes]

Response 7 (6 was broken off by Ralph's reply).

First, Kirt, that's a good point. All games can benefit by considering any other games. Simple as that.

Not to be too tangental, but if you're referring to the PC game Warlords (I through IV) the steam roller effect only really applies to the end game where most conquest games that don't have alternate victory conditions break down.

I'm thinking in terms of a two player game (the AI can't beat a human player with the least bit of sense). The player who does better earlier will win almost all of the time. There's not enough other tactics to allow for comebacks when you get behind. Note that if you give the computer the advantage this way that it then can beat you, or at least extend play interminably.

For most of the game the steam roller effect is mitigated by the up keep cost to garrison all of those castles (or cities in later versions).  There is a signficant period of time in the mid game where he with fewer castles has an easier time concentrating his forces and can use this advantage to achieve victory.  In fact, growing too rapidly typically results in a paper empire that is easily overwhelmed by a smaller empire with better infrastructure (Warlords IV is the weakest of the games by far and doesn't do this nearly as well as II or III).

Note that you're saying that the person is paying for their expansion. In that case I agree. But if I take one castle with no losses, and you take two, who's ahead? My point is that, beginning or end, having a steamroller effect usually leads to single tactics being the best. Note how lacking Warlords is in this respect (I've not played 4, but from what you say it's actually worse)?

Mike

[Valamir]

Note that you're saying that the person is paying for their expansion. In that case I agree. But if I take one castle with no losses, and you take two, who's ahead? My point is that, beginning or end, having a steamroller effect usually leads to single tactics being the best. Note how lacking Warlords is in this respect (I've not played 4, but from what you say it's actually worse)?

That depends.  If we start with an identical stack of 8 and assume that you need at least a stack of 4 to be effective at attacking, and we assume that we each leave 2 units in captured cities for garrisons, then you have an army of 6 and I have an army of 4 left after our respective conquests.

If we are near each other, you are now in a stronger tactical position (even more decidedly stronger if I take a third city).  Your army of 6 (with its more potent combination of bonuses) is in an excellent position to take one of my undergarrisoned cities and completely reverse the situation.  Expanding too fast is a common newbie mistake in Warlords, and one I exploited many times in opponents who thought they were winning because they had alot more cities than me.

If we are not near each other its a little more in my favor because I have time for the added production to increase my army size faster than yours.  Still not a 100% certain advantage however, because the more spread out I am, the more difficult it is for me to defend what I have.  I will need stronger garrisons to make sure you don't slip into a lightly defended area and simply raze half my empire to the ground.  I will need more offensive stacks to be able to react to your attacks.  This effectively means that simply having alot more guys than you (which is what having more cities get me) doesn't mean I have more offensive power than you.  It is extremely easy in Warlords to get into a situation where I'm financially broke and have a huge army, but can't get the army where it needs to be fast enough to prevent you from ripping into me.  

The game Age of Wonders (heavily influenced by Warlords) is actually even better at this than Warlords.

Now after a certain critical mass...yes, an overwhelming advantage in cities will win through sheer attrition.  But this is not "as soon as one side has more castles".  There is a pretty sizeable and tactically enjoyable mid game where the winner is not determinable simply by counting castles.  Which is why the game is one of the most successful and popular franchises in game history.

In other words I'd hold it out as a particularly effective gamist design.

Warlords IV is worse.  In addition to being 1) taking the basic game play of Warlords and coming up with a particularly joyless product and 2) coming out with a game that is noticeably inferior in play, options, and production value than its older competitor AoW II, the AI is particularly non aggressive.  A non aggressive AI is easy to beat with the steam roller effect.  Earlier Warlords versions had AI that was aggressive enough to always leave you wondering if you could really afford to strip your garrisons down for an additional attack stack.

[greyorm]

It's not really a lack of tactical information - players obtain that quickly enough by back calculation in most games (Hmmm, a fifteen hit, but a fourteen didn't = AC 2!). Again, it's "encounters". Is there a goblin in the last room, or a giant?

That's the beauty here -- it doesn't matter. If all the GM has left is a d10, or if all he has is a d4 and a d6, then players can plainly see that -- his resources are all up front, and that's all he has to go after the players with (I think ORX has changed a bit since you played it with me, where I don't think that rule was implemented yet).

Simply, you can see his pieces and where they lie. There is absolutely no difference between a d4 giant and a d4 goblin -- Color is completely seperated from the mechanics of a situation. Same dice, same rolls, no difference.

I may be misreading your criticism here, so help me out here if I'm on the right track or not -- using this information relies on some guess work by the players and some gambling, but so does chess ("Is he going to move his King or his Pawn?") and most other pure strategy games.

{EDIT} Adding a question for clarification: by your estimation, is this good Gamist design or failed Gamist design? And why? If the latter, what would make it a better design?

I ask all this in an effort to better understand the criteria and thus the rant.

[Callan S.]

Meaning that much of the challenge becomes a matter of bravado, or seeing how far you can push into the adventure before being repulsed by it.

Can I just say I LOVE this sort of thing. I don't know where I am in terms of gamist play style, but I think Diablo II is pretty gamist. Now, in that it's quite easy to fall back every time you take a little damage, or just sit still and let regen heal you to full. But you don't, because its not fun. Then again, that's a sub optimal choice to go in without full health when its easy to get...and sub optimal choices aren't really gamist IMO. So where does that leave most Diablo II play?

BL>  I think that the key here is that Gamism is about Challenge, rather than about Victory.  Observe the Go player who plays with a handicap to equal the playing field.

Diablo II has the added effect of trying to maximize benefit over time played, which is more complicated and not (I think) directly relevant to TRPG Gamism.

yrs--
--Ben

Good point. Can I ask, though, is it percieved that part of gamism can be  about choosing your goals (and the harder the goal, the harder the challenge. Thus by choosing the goal, you choose your challenge/handicap?). I tried to read a forge essay on gamism, but I just wont be able to absorb it till I've printed it, so excuse any questions here.

For example, could a player decide to himself that game time (not real time) is important to him, and thus change a lot about the games strategic and tactical levels. I mean, do all gamists have the same goals in play? Can one judge a system to not really be gamist, when it might be, if one is an advocate of certain particular goals?

Must print out those essays soon...

[Callan S.]

*snip*

Right. You can always make retreat a failure condition (the Princess gets sacrificed). But that doesn't make the tactics any more interesting.

Mike

Well, I was about to ask if there wasn't any trickle down of interest from the strategic to the tactical, when the princess can get the chop. But it isn't you point, really. It doesn't make the tactics any more complicated.

So lets look at one of the orignal posts points
The only tactic that makes sense in attrition games is to reduce the enemy's numbers of attacks first. To that extent, "good play" becomes making sure that PCs gang up on opponents as much as possible. After that you try to make sure that characters with more vulnerable and valuable resources are exposed to the least danger (fighters in front, wizards in back). Then you match up attacks with susceptibilities to attacks. This is so elementary that most master it after only a session or so. If they haven't, they're not "Gamists", and don't care.

Doesn't changing the environment do this though? If you just have three orcs, okay, the group rumbles one, then the next, then the next. Simple. But what if they are three orc snipers that you haven't spotted yet. Just being able to spot them will require menouvering, perhaps.

And if it is just a matter of running up to three orcs with swords...well, how does chess handle this. After all, in that game its just a matter of walking up and bang, you take the piece. Well chess basically handles it with 'covering fire'. Each piece is covered by another piece. In the orcs example, if only Jimmy the PC can resist fire, then he's the only one to fight the magically flaming orc.

Of course, once you've figured this out, the tactical part runs dry. How would chess answer that? Well, the opposing side would move in in anticipation. Eg, the flaming orc falling back to an ambush point.

Of course, this might be considered the strategic level. But when you don't have sword blow by sword blow combat like in TROS and instead that levels fairly abstracted, then at the first point where you get to make choices is the tactical level. Even if its in a risk game. It might feel as if it should be called a strategic level game. But just because it describes you as using army against army, your just using object A against object B, which is at the same level of TROS sword/object A vs TROS object B. For the characters Vs the orcs, they might consider it tactical. But for the player, its tactical (eg, he doesn't have a say on how each sword blow happens, so the next layer up where he does get a choice, is his tactics: Choosing who to use the sword on).

Finally, an example. A TROS equivalent of the 'run up to the orcs, use the same old plan' would be bad guys who always use the same dice spread and no special moves or anything. That would pretty much kill tactics in TROS as well.

Man, I go on and on sometimes... :(

[Drifter Bob]

I think that, for me, the most interesting idea is that of moving strategy out of just the realm of combat and adventuring and into other realms like Politics, merchantilism, and most importantly "everything all at once".

Mike

You tapped on something I have been thinking for a long time, and something I touched in when i was talking about the idea of using realism as a tool, or as a pallate of options.  RPG's started out with this head start in combat realism, which they got from War Games.  This has been developed into in many cases the most challenging parts of many RPGs, the crunchy part where people can really feel challenged.  

Rather than ditch that, it seems to me like you can expand from that into many other areas of reality.  And it doesn't mean you have to be a 'slave to realism', you can still analyze what really goes on in the world (or has gone on in history) and bring it into the RPG at whatever level of abstraction you want.  For example, I love the way Dying Earth RPG, which is trying to emulate a literary genre and not reality, takes the dynamic of (their unique take on) combat and brings it into the realm of verbal persuasion between people.  The idea of the way different personality archetypes can trump each other (forthright vs glib or obfuscatory vs. lawyerly, etc.) is actually a fascinating insight into the human condition, while simulatneously being a very entertaining game dynamic borrowed from reality.  And with their single die system, it's hardly bogged down in detail.

I think all kinds of aspects of the human condition, especially the more interesting survival oriented aspects which RPG's often deal with, could be explored with more nuance which can be informed by reality, to build the dynamics which are more challenging.  Like when a "thief" (to use an old D&D archetype) is climbing a wall, or picking a lock, or looking for a trap, there could be many different kinds of approaches with different trade offs.  One way to climb the wall might be faster, another safer.  One type of trap detection or disarming technique might be suitible for one type of trap, but not as effective against another.  Knowlege of whoever set the trap could inform a tactical decision at this level.

I think this is at least one road for a possible future of RPG games, a way they could actually become more "grown up" they can actually inform us about life while we borrow from it to make game play more rich and involving...


JR

[Harlequin]

Two quick thoughts here.

Bob, yes.  There've been several discussions of this sort of thing kicked around here.  Several threads on Social Combat in the TROS forums, for example - I think this is one of the most promising examples, as well, which is why it keeps coming up.  Interesting to note, though, that nobody has yet managed to crack some of the key differences enough to present details, unless it's happened since I last looked for it; detailed tactical resolution remains the domain of combat for most games because although we focus on similarities (innuendo as attack, yadda yadda), the differences are subtler and often derail the design (what do you do to "parry" innuendo?) or simply pile up in a big Sim heap and defeat the purpose.

Not to say it's not doable, but that actually getting fun tactical Gamism out of anything but combat doubles Mike's original point: Gamist systems are harder to make than we think.  Noncombat gamist ones doubly so.

In terms of the Diablo analogy: This is indeed a relevant note, because that sort of gamism is indeed wildly popular.  Sure, you could play methodically, and I know some who do (frequently including me), but the fun level actually increases as you buy into the design choice that nonmethodical play, though sub-optimal, is actually more fun.  I think there are two elements overlapping, there; Gamists love a challenge, and by increasing the challenge due to a strictly OOC decision, you're increasing that challenge without having to sacrifice your sense of triumph.  [As a counterexample, agreeing to have your PC downgraded by five levels would also increase challenge, but would be anathema.]  Also, pure challenge is not the only aesthetic of the play, and the (part sim-char, part pure esthetics of elegance etc) "don't go back to town just yet" mode of play satisfies both considerably better than "kill one monster, rest, heal, return."

And I think that Gamist games, to return us to Mike's topic, could benefit from this example - some of them, anyway.  Because you don't always have to design to resist the "twinks" who will methodicalize their way through the module without regard for other aesthetic choices.  Those who are really getting their Gamist kicks will subscribe to voluntary challenges (how long can you stay on Level Fifteen before using Town Portal?), and those whose enjoyment is based on a less purely Gamist agenda will have other aesthetics to satisfy simultaneously (my character wouldn't back off yet!, or the simple elegance/story aesthetic that says that one epic journey is superior to fifteen short recon trips).  Really, many Gamist designs build too much for the "Pseudo-Gamist" type... those who are ego-stroking, not playing against the challenge for the enjoyment of it.  

For example, not all weapons need to be statistically balanced, Sim considerations quite aside; having one (often the rapier, for example, or the whip) with lesser stats and superior "cool-factor" is a Gamist choice, not a Sim one.  Sure, you could burn through that dungeon with a spiked chain.  But can you do it with nothing but a set of brass knuckles?

And as always, System Does Matter.  Even with regard to aesthetics - maybe even especially then.  If you want to replicate the Creative Agenda of good Diablo play, and encourage players to do better in X time than others might, then build toward that.  One thought, just as a toss-out from a fictitious example game, that could accomplish that:

The spell "Goodnight," available to all classes, brings you to the party's HeartCamp, a pocket realm where all your damage can be healed and your spell slots regenerated.  However, the spell is named "Goodnight" for a reason.  When you use it, the game session is over, for you, today.  You won't return to the ArenaWorld until the next game session.

Sure, "most efficient" play would have you return frequently to the HeartCamp, rather than risking your life once your resources drop.  But by using issues of pride and bringing them right to the front, that's definitely sub-optimal Gamist play.  Optimal play is, indeed, "can you hold out (a) 'til last, and (b) 'til the game is called because the GM/players are tired, instead of due to the last player casting Goodnight?"  Which is much closer to the indisputably enjoyable, indisputably Gamist agenda shown in Diablo and its cousins.  Design for effect.

- Eric

[Mike Holmes]

Hmm. Somehow I misses several replies.

That's the beauty here -- it doesn't matter. If all the GM has left is a d10, or if all he has is a d4 and a d6, then players can plainly see that -- his resources are all up front, and that's all he has to go after the players with.

Yes, that makes for a good strategic level of planning. Still doesn't do anything tactically.

using this information relies on some guess work by the players and some gambling, but so does chess ("Is he going to move his King or his Pawn?") and most other pure strategy games.

That's true, but is basic to most games with more than one opponent. By hidden information, usually I refer to something like not knowing how many dice the GM has. But to an extent. If the player has no idea at all, then there's no way he can plan. But if there's some way he can estimate, then that becomes the skill of interest in determining the best strategy with regards to the hidden information. So, basically, your game goes to the other extreme in making all information known. It means that the player has something to work with, but the "calculations" are less difficult than if he had to do them estimating how many dice were left.

How you'd accomplish getting the pool to be partially understood, I don't know.

Adding a question for clarification: by your estimation, is this good Gamist design or failed Gamist design? And why? If the latter, what would make it a better design?

Well, to be truthful, I think that there are too few variables for the game to become very deep in a Gamist sense. OTOH, I think it has a very good "Beer & Pretzels" level of Gamism. See the other thread for clarification of what I mean. In any case, it gets it's Gamism pretty elegantly, which is a plus.

Mike

[Mike Holmes]

Can I ask, though, is it percieved that part of gamism can be  about choosing your goals (and the harder the goal, the harder the challenge. Thus by choosing the goal, you choose your challenge/handicap?). I tried to read a forge essay on gamism, but I just wont be able to absorb it till I've printed it, so excuse any questions here.

For example, could a player decide to himself that game time (not real time) is important to him, and thus change a lot about the games strategic and tactical levels. I mean, do all gamists have the same goals in play? Can one judge a system to not really be gamist, when it might be, if one is an advocate of certain particular goals?

Somehow people are getting the opinion that "good Gamism" must mean one and only one thing. That's patently not true. There are probably infinite ways to achieve it.

That said, almost all Gamist RPGs allow the players and GM to tailor the level and types of challenges to some extent. This is, in fact, the one (and maybe only) really big advantage that PnP Gamist RPGs have over other forms of games. To the extent that this is enhanced, sure it's probably a good thing.

But there's a danger as well, which is that in leaving things open you run the risk of the game becoming untennable as a challenge in some way. Hard to come up with an example, but lets say that a player makes up a character who is just too crippled to take on anything that the game has designed to throw at the player. Well, then you have a problem.

As long as that's kept in mind, however, I think that allowing for variance in the challenges presented is a valid goal.

Mike