Re: PBM design and formulae - a fresh start

From: saschreu@cs.vu.nl (Schreuder SA)

Date: Fri, 02 Jul 1993 14:04:09 +0000

gym@dcs.ed.ac.uk (Graham Matthews) writes:
>In an effort to take some heat out of this debate I am going to make
>a fresh start on this thread. I am going to spell everything out in
>detail and use some examples to get some specifics.

Fine! This discussion became too much: "You don't listen", "You're crazy",
and so on.

>Assumptions :-
>
>a) we are talking about simulation games.
>b) the type of game could be anything (eg: boardgame, PBM, ...)
>
>My position: games, being simulations of something, should be as
>realistic as possible up to the level of the simulation. Players
>should be told the general rules of play, the scenario etc, but beyond
>that they should only be allowed to learn things about the game and
>how to play etc through the actions of the entity they represent in
>the game. As a corrolary therefore I do not think exact formulas
>should be published. The reasons I take this position are twofold :-
>
>a) it adds to the realism of the game. In the real world people
>don't know exact equations for things and yet seem to do ok.
>A simulation should therefore capture this imprecision - if it does
>not, then it is not a good simulation.(even the formulas themselves
>are only approximations to the reality!).

I agree partially with this: Sometimes I play games not for reality,
but to show other people I'm good in it. I'll say it again: I'll
play some games to show I'm good in analysing turnsheets! That's part
of the fun for me: Being better than other players in using all things
that are possible (Excluding exploring obvious bugs).

>b) it makes the "playing field" as level as possible. People who
>are not math or CS majors are not disadvantaged too badly by not
>understanding the formulas. Moreover everyone has the same basic
>learning mechanism, namely the actions of the entity in the game
>that they represent.

I agree with the first part: To make the playing field as level as
possible. But I don't think it can be done by hiding the formulas.

So we have two things:
a) reality
b) playability

Somewhere there has to be a good mixing of those two. But we disagree
where it is. No problem: I will play other games than you do, and
we are both satisfied.

>Now there are caveats in this position, namely the caveats of :-
>
>a) experience. The more players play the more experienced they get
>and hence the better they get. This is inevitable - publishing
>the formulas would not solve this problem because the non math
>people would not be helped by knowing the formulas for example.
>The only way to counter this is to let the players treat experience
>equitably by helping newbies or by playing with people of the same
>experience.

Publishing rules won't give experience! I knew the formulas of Galaxy,
I've analized them, but still I have made many mistakes. But because
I knew the rules, I knew what went wrong, and now I won't do it again.

There is nothing against experience. I guess even David has made some
mistakes in Galaxy when he started. Now he has (much) experience,
and he won't make them again.

>b) statistical analysis. Again there does not seem to be all that
>much that can be done about this. Sure lots of statistical analysis
>will help you gain approximations to the formulas that are being used.
>But sufficiently complex formulas, random walk, and many aspects of
>the game that don't depend on numbers (eg. alliance making), together
>with sensible relationships between things (eg. food growth depending
>on season and not on the colour of the king's hair) will tend to
>reduce the advantages of the statistical analysis.

Here I have to disagree: I don't know how good you are in statistical
analysis (I guess not that good), but you don't seem to realize what
is possible with the right tools. Random walks can be excluded by averaging,
alliance making is player-interaction, and that can never be predicted
(except by assuming they will play as well as possible too), and
sensible relationships make analyzing only simplier: it gives clues
which parameters have (almost) no influence.

>Now to an example. I will stick to one example for the moment.
>We are to simulate a game of tennis. Now one immediately notes
>three things :-
>
>a) the ball obeys some equations, parameterized by court speed,
>air pressure, spin, racquet speed etc.
>b) practically no players know the equations and yet still manage
>to play somehow.
They do know, in the form of a neural network (brains). Maybe they
don't know the mathematical form, but they have analyzed years
of playing, and now have experience.

>c) knowing the equations would not help a player play better because
>by the time he/she had solved the equations the ball would have
>gone past them.

I have to object against this: A monkey will hit the ball in a random
direction. From Physics one can conclude that the ball will travel
in an almost straight line, and curve downwards. Now supose somebody
with this knowledge plays against a monkey. The monkey will loose,
because halve of the time, it won't hit the ball in the direction
of the net!

Now our physist (P) will play against an average tennis player (T).
P will have a better chance against T than the monkey has, because
P knows how to hit the ball. So knowing the rules gives some compensation
for the lack of experience.

And of course, no one is able to calculate it on paper, but everyone
can predict where the ball will land. And that is the same as calculating
it.

>So if we are to write a good simulation of tennis we should not
>make it equation based, or rather we might have the ball obey
>equations, but the interface with the human would not be through
>equations. We would not ask tell the player - oh the ball is coming
>at you with the following parameters ... what parameters would you
>like to reply with. What we would do instead is say have a small
>picture of the racquet showing its tilt etc (no figures just the
>picture). Through practice the players would get to know how to
>position the picture to hit certain shots.
>
>Now you will note that this is imprecise. The players don't know
>exaclty the formulas for the ball motion, and even if they did it
>would not help them as they don't interact with the game through
>figures. But its also realistic!

And this reality can never be obtained. We are talking about text
and computers, which are non-continious by nature. So there are a
finite amount of distinct situations (maybe very large), and
a formula will exist.

Besides, to have an exact answer, one has to include relativity,
the exact mass of each sand particle, etc. But a close aproximation
can be obtained by ignoring this. To prevent such things from happening
in games, many parameters should be important, which makes intuition
impossible.

To summarize: analysing the formulas can compensate lack of experience
to some extend. And that is _my_ reason to publish them. After all,
I want to have somewhat equal equiped players in the game.

>I will await some comment before moving onto another example.
>
>graham
>
>PS: when you agree with something please say so as it makes the
>ground a little firmer for both of us.

Sjoerd Schreuder
saschreu@cs.vu.nl

PS: Yes, I play Galaxy, but I'm also roleplaying in RL.

References

Re: PBM design and formulae - a summary
gym@dcs.ed.ac.uk (Graham Matthews)
Thu, 01 Jul 1993 11:40:27 +0000
PBM design and formulae - a fresh start
gym@dcs.ed.ac.uk (Graham Matthews)
Fri, 02 Jul 1993 12:01:01 +0000

Referenced By

Re: PBM design and formulae - a summary
gym@dcs.ed.ac.uk (Graham Matthews)
Thu, 01 Jul 1993 11:40:27 +0000
PBM design and formulae - a fresh start
gym@dcs.ed.ac.uk (Graham Matthews)
Fri, 02 Jul 1993 12:01:01 +0000