[MUD-Dev] Cognitively Interesting Combat (was Better Combat)

[Paolo Piselli]

It seems to me that the discussion of "better combat" is diving
depth-first into tangents on chess, randomness and game mechanics
without taking a look back and asking: what is it that makes combat
"better"?  I don't think any one of "make it like chess", "make it
more random", or "make it more mechanically complex" are good
general-case solutions.

Lets try to restate the problem.  "How do we make combat better?"
depends on what "better" means.  I'm guessing that we could at least
agree on "better" meaning "more interesting", but this is still too
poorly defined.

Making the mechanics more complex makes combat more "interesting" to
the few explorer types who first do the analysis, however (as stated
before) if all those mechanics are reducible then optimal combat
"solutions" will be discovered and the majority of players will
still find combat just as uninteresting as before.

Adding more variance to combat outcomes may make it more interesting
in the same way that slot machines are "interesting", however
something tells me that taking players on a stochastic
roller-coaster would do more to conribute to user fatigue than to
make them more engaged in combat.

Making combat more like chess?  This suggestion has some merit, as
chess has the quality of interesting-ness, but chess itself is not
the solution.  As stated, something on the complexity level of chess
is too complex to be played for every combat running on a server,
also the search space is too large to be keeping around a giant
pre-computed game tree to point mob combat-states at.  The
discussion of chess so far has touched upon what I'm getting at, but
I want to lift it into the greater context as that thread seems to
be heading off into a discussion of branching factors and minimax
algorithms which, although interesting, will not answer a thing
about how to make combat more interesting for human minds.

When we ask "how do we make combat more interesting?", we have to
consider the context.  We are trying to make the *process* of
in-game combat more interesting to the mind of a *human* player.  To
me this cries "cognitive task analysis" - not "search trees" or
"game mechanics" (I guess those with hammers will always see nails).
In my estimation, Puzzle Pirates has you all beat for this one
reason: "combat" via those puzzles is far more cognitively
interesting than "debuff, stun, nuke, nuke, nuke" or whatever
variation thereof is the optimal combat procedure for class X in
game Y.  IMO the question we need to answer is:

How do we make combat more cognitively interesting?

To answer this question, we need a few things: First, we need some
initial constraints to bound our ideas.  Second, we need some way of
assessing how cognitively interesting a given combat system is to a
certain player.  Third, we need some way of evaluating wether or not
some change makes a system more cognitively interesting.  Finally,
we need to address how the cognitive capacity of a player changes
over time (i.e. player learning) and how this affects the interest
level of combat.

So here are some questions to lead off discussion on cognitive
constraints:

  - what kind of mastery of the keyboard and mouse should the player
  have?

  - how fast of a reaction time will the player be required to have?

  - how frequently will the player have to make decisions?

  - how much time will the player have to make strategic decisions?

  - how much information will the player have to remember during
  combat?

  - how many cognitive structures (production rules, decision-tree
  nodes, or whatever model you prefer) will the player need to
  maintain in order to be successful at combat?

...and a thought to lead off discussion on cognitive interest: - It
is my opinion that making combat interesting involves finding the
sweet-spot of cognitive-load on the player where she is neither
bored nor overtaxed. This means that the player has meaningful
decisions to make and actions to take, all in response to various
combat events over the course of the entire combat.

....and a thought to lead off discussion on changes that will
increase interest: - Wether its RPS fighting-stances or alpha-beta
decision-trees, I see these things simply as ways of increasing
cognitive-load, and evaluating their effectiveness at improving
combat amounts to evaluating how they impact the cognitive tasks
involved.  You could argue that in-game-context devices such as
fighting-stances are neccesary, but I'd point to Puzzle Pirates as
an example that out-of-context devices can be used to make combat
interesting.

....and a thought to lead off discussion on player learning: - well,
since IMO its all about cognitive-load, then the natural conclusion
is to increase cognitive-load over time.  However, our constraints
put a limit on this in order to keep the game accessible.  I'd
propose to allow players of all capacities to be successful at
combat, yet give benefits to those capable at performing at a higher
level (less downtime, more XP, whatever).

Anyhow, thats just my 2+ cents.  -Paolo

=====
Paolo Piselli
ppiselli at yahoo.com
www.piselli.com , www.bestcoastswing.com
_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev