Re: How to hide this formula?

From: ulim@coli.uni-sb.de (Ulrich Mayring)

Date: Sat, 26 Jun 1993 23:14:43 +0000

js138@cus.cam.ac.uk (John Sloan) writes:

>|> This example taken from an actual game gives you 16 pieces of data for one
>|> unit. If you give it all away, you simply list them in a table. Do this for
>|> all 152 types of troops and that still leaves you with an impressive
>|> knowledge base that takes very, very long to master, even for veteran
>|> players. And that although the data is all there.

>This strikes me as excessively complex, giving a challenge to even the best of
>optimizers!  Not my thang.

Well, I haven't even begun to describe the complexities of economics in
this game :-) Even someone who can run a mental minimax over the troops
table might lose badly, because of his lack of strategic ability and
creative economics he brought fewer troops than anyone else to the
battlefield. The one way to prevent minimaxing is to make the game
excessively complex, but (not like ASL) still playable very easily. This
means you have to add lots of variety in few concepts, rather than lots of
concepts with few variety (-> ASL)

>No, you would have 152 entries in a table form, which might look like:

>Roman Legion Infantry
>Cost : High
>Cost in weapons/armour : High
>Armour Class : Good
>Morale : Exceptional
>Ranged combat : None
>Experience : Good.

This leaves no room for hard-hitting economics, because if you want to
recruit, you'll always have to have some more money than you actually
needed, if you knew the exact price.
Also, I prefer to handle this low-end variation with random numbers and
giving away the range for them.

>You still have your table, but you define ranges of values, rather than exact
>values.  Things like the exact cost would approximate to a value, but would tend
>to vary according to where you recruit, the time of year, your popularity in
>the region (say) and the price of tea in china (a random figure), so you couldn't
>simple average it over time to find it out.

Yes, I find it very desirable to have this kind of variation. IFF you
disclose the formulas. Constraints like time of the year, popularity,
location of recruiting etc. are all known, so I can figure it out if I
want. All this excepting the one random number, whose range should (and in
the case of this game) is known.

>And would the game be much worse if it was invoked like that?

Actually, it would be much better. Sadly it is not being developed
anymore, because the designer feel that it is the best historical
simulation PBM on the market. And whaddaya know, they may be right. But
not for long anymore :-)

Ulrich

References

How to hide this formula?
ulim@coli.uni-sb.de (Ulrich Mayring)
Thu, 24 Jun 1993 09:29:04 +0000
Re: How to hide this formula?
js138@cus.cam.ac.uk (John Sloan)
Thu, 24 Jun 1993 10:28:55 +0000