Re: PBM design: give away the formulas? (*FORMULAE !)set seen

From: 11jkessler@gallua.gallaudet.edu (Jeffrey Kessler)

Date: Thu, 24 Jun 1993 02:10:25 +0000

In article <20a8j2INNj8t@bredbeddle.cs.purdue.edu>, sjc@cs.purdue.edu (Steve Chapin) writes:
> }} In article <20a6s1INN7vl@uwm.edu> bruno@cerberus.csd.uwm.edu (Bruno Wolff III) writes:
> }}
> }} The fun in the game should come from discovering reasonable things
> }} about the world, building up your characters/empire, making friends
> }} and enemies, helping your friends and defeating your enemies.
> }} Knowing the rules does nothing to take away from this.
> }}
> }} Knowing all of the rules does not mean knowing everything about a single
> }} instance of a game. There can be much to discover within a known framework
> }} of rules.
>
> The question wasn't "knowing the rules," it was "knowing the formulae."
>
> For example, if the following appears in the startup packet ("the
> rules"):
>
>     Combat is a dangerous and deadly activity, best undertaken by
>     those who are practiced in the arts of battle, and who are well
>     equipped with high-quality weapons.  While one can eventually
>     subdue an oppenent bare-handed, an expert swordsman can hold his
>     own against a mob of untrained peasants.
>
> Do you really need to see:
>
>     damage = c1 + (c2 * weapon_value) + (c3 * skill_value) +
> 	     (c4 * weapon_value * skill_value)
>
> Properly explained rules tell you what you want to know:  that skilled
> people with weapons kick the living daylights out of the average mug.
> Players know that as they buy better weapons and train their troops,
> they will be better fighters.

The problem with the textual discription is how skilled are the respective
forces, or simply "How skilled is skilled?"  Formulae allow the player to
decide at what point should the training stop and the butt kicking begin
much clearer than a simple text discription.  In multi-player, high
interaction the formulae really won't tell you what the other players are
doing and guesses or intelligence may have to supliment the formula.  But
this way the player doesn't need to waste his effort over-training.  The
formula are a way of asking the character to rate his skills versus
possible opponets without actually risking his life.

> A well-designed game will follow two of Lindahl's Laws: the principle
> of least surprise, and equal investment yields equal return.  The PoLS
> means that more expensive weapons should be more deadly.  The EIYER
> means that if I invest the same amount of time and money on a mob of
> rabble or a few highly-skilled, well-equipped swordmasters, they
> should be equally deadly.
>
> In short, a well-designed game removes the whole incentive to optimize
> (and the Super Improved Oly Combat Rules did just this).  Instead, you
> choose a style of play, and run with it.  Do you want to lead legions
> of peasants into battle?  Enjoy!  Do you want to craft the Elite
> Guard, a nimble and deadly merceneary company?  Knock yourself out!

I have a trouble with the concept of eaqually deadly in this example.  When
different strategic paths are followed than the should both have chance to
win.  But this chance doesn't have to be equal.  Only compartive depending
on random factors and other player actions and the skill of the player.

> Of course, the peasant mob can carry more than the Elite Guard, but
> the Elite Guard will be able to enter smaller areas, or move more
> quickly (it's a lot cheaper to mount 10 than to mount 1000).  Peasants
> are cheap to replace; the Elite Guard are expensive.
>
> I claim that informative textual descriptions and well-designed
> mechanisms obviate the need for publishing formulae.  It's easier to
> just make up a formula and publish it, than to work and get it right.
> Any formula that *can* be optimized violates EIYER (which can also
> paraphrased as: there shall not be One True Way to play the game).

Because in the *yet* to be published perfect game there should be more than
one formula.  Therefore the goal would be to optimize as many as possible
to win.  Or there should be a variety of optimalization points.  Want to
kill peasant mobs, optimize here, want to kill wizards, optimize here.
This also the player to tailor his/her strategy to his/her opponents.  Thus
winning relies on figuring out what one's opponent is doing and counter it.

> If you're going to publish optimizable formulas, why even bother to
> provide the players with options?  Have One Weapon, One Armor, and One
> True Way of Doing Things.  That's the one that everyone will adopt,
> anyway.  As an example balanced choices, the weapon and armor ratings
> from Oly are carefully balanced in terms of weight, protection or
> deadliness, and cost.  There is no single "best" weapon or armor for
> all three parameters (to paraphrase a friend's law of CS: light,
> cheap, effective: choose two).
>
> Present the players with a consistent, fair underlying mechanism, and
> let them choose a style of play.  I'm sure this is what Scott meant by
> a sense of wonder.  Don't force them into a predetermined style of
> play.
>

Clearly we both agree that games with one way to win are boring.  Luckily I
have never played a game that was this way.  Mostly I look to see if there
is only one optimal strategy and if so I pass.  I got started with
PrimeTime Simulation's World Conquest.  Very fun and high interaction with
many strategy's to win.

Now that I play diplomacy through the judges I only have time for Monster
Island. (I'll find those demigods if its the last thing I do.)

jeff kessler
Gallaudet University
11jkessler@gallua.gallaudet.edu

References

Re: PBM design: give away the formulas? (*FORMULAE !)
srt@aero.org (Scott 'Dr. Pain' Turner)
Wed, 23 Jun 1993 17:09:10 +0000
Re: PBM design: give away the formulas? (*FORMULAE !)
bruno@cerberus.csd.uwm.edu (Bruno Wolff III)
Wed, 23 Jun 1993 18:19:45 +0000
Re: PBM design: give away the formulas? (*FORMULAE !)
sjc@cs.purdue.edu (Steve Chapin)
Wed, 23 Jun 1993 18:49:06 +0000

Referenced By

Re: PBM design: give away the formulas? (*FORMULAE !)
srt@aero.org (Scott 'Dr. Pain' Turner)
Wed, 23 Jun 1993 17:09:10 +0000
Re: PBM design: give away the formulas? (*FORMULAE !)
bruno@cerberus.csd.uwm.edu (Bruno Wolff III)
Wed, 23 Jun 1993 18:19:45 +0000
Re: PBM design: give away the formulas? (*FORMULAE !)
sjc@cs.purdue.edu (Steve Chapin)
Wed, 23 Jun 1993 18:49:06 +0000