Re: PBM design and formulae - a fresh start

From: markiel@callisto.pas.rochester.edu (Andrew Markiel)

Date: Tue, 06 Jul 1993 09:07:25 +0000

In article <218qqs$4pk@taloa.unice.fr> gilette@obsidian.unice.fr (Gilette Serge) writes:
>In article <gym.741614461@dcs.ed.ac.uk>, gym@dcs.ed.ac.uk (Graham Matthews) writes:
>
>A wargame as 2 type of units : infantry and cavalry, which can be either
>regular or elite, and we want to statistically know what are the formulas.
>we'll have to experience in the game at least 1000 x 4 x 4 combats
>to have an idea of what the formulas are, which is completly impossible
>in a game, as it's in real-life.
>
Well, the number isn't that big. First of all there's only 10 distinct
combats (4x4 double counts 6 of them). And, only 10 of each would give
me a good estimate of the probablilities; 20 or 30 would nail it down
rather closely. So that's on the order of 100; not excessive.

>So i doubt it's possible to made such a study during a game, as it's in
>real-life (as circumstances aren't ever the same twice), and hence i'm
	    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>against showing the formulas, (for you're not able to find them by
>experience). That might be a bit unclear :)
>

There is an excellent point here. In real life, circumstances are
never the same, because there are always certain small random
effects that change the result. In a computer moderated game, however,
the circumstances *are* the same: every time you put the same numbers
into the formula, you get the same result out. Thus, by observing
long enough a player can *exactly* determine the formula. This may
be difficult if the formula is complicated, but in theory can be
done perfectly.

Now, you could insert a random element into your computer formula,
to simulate such random effects. But then knowing the formula does
not allow you to calculate the exact number, so then the exact
knowledge argument doesn't seem to imply that the formulas should
not be revealed.

You (or others) may be implying that it is a bad idea to use explicit
formulas in a game: and this is a reasonable comment. However, in
line with the original question posed, *if* you are going to use
deterministic formulas in your game, then I think that it is
generally superior to reveal them, rather than keep them secret.

>
>Serge Gilette  | University of Nice |There's no difference between theory
>------------------------------------|and practice in theory, but often a
>phone: 92.94.26.41                  |great deal of difference between
>email: gilette@obsidian.unice.fr    |theory and practice in practice!

References

PBM design and formulae - a fresh start
gym@dcs.ed.ac.uk (Graham Matthews)
Fri, 02 Jul 1993 12:01:01 +0000
Re: PBM design and formulae - a fresh start
gilette@obsidian.unice.fr (Gilette Serge)
Mon, 05 Jul 1993 09:04:28 +0000
PBM design and formulae - a summary
rob@ccc.govt.nz (Rob McNeur)
Mon, 28 Jun 1993 09:53:02 +0000

Referenced By

PBM design and formulae - a fresh start
gym@dcs.ed.ac.uk (Graham Matthews)
Fri, 02 Jul 1993 12:01:01 +0000
Re: PBM design and formulae - a fresh start
gilette@obsidian.unice.fr (Gilette Serge)
Mon, 05 Jul 1993 09:04:28 +0000
PBM design and formulae - a summary
rob@ccc.govt.nz (Rob McNeur)
Mon, 28 Jun 1993 09:53:02 +0000