[MUD-Dev] Game Economies

[Caliban Tiresias Darklock]

On 04:01 PM 6/9/99 -0500, I personally witnessed Koster, Raph jumping up to
say:
>
>We did something like that on UO for a while--the total quantity of gold in
>the system was tied to the total size of the playerbase, and goods/gold on
>monsters (or issued by shops) was tied to how much remained in the "bank"
>(eg, not in play, either on monsters or in players' hands).
>
>The system broke because players hoarded *everything*. Soon no monsters were
>worth killing because they had no loot whatsoever, and every shop was out of
>stock and out of cash.

I started thinking about this sort of thing for UU to make up for the
open-ended playerbase, but I saw some problems when I really examined the
concept:

  Allow every player a cash value of 20m where m is any random constant.
Start at 0m.
  Player A logs on and plays. Pot increases to 20m. 
  Player A makes 10m very rapidly. Pot is now 10m.
  Player B logs on. Pot increases to 30m. 
  Player A continues playing and makes another 10m. Pot now 20m.
  Player B plays and makes 5m. Current pot now 15m.
  Player A continues playing and makes another 10m. Pot now 5m.
  Player B continues playing and makes another 5m. Pot empty.
  Player B plays and gets nothing. Pot is empty.
  Player C logs on. Pot increases to 20m.
  Players A and B immediately grab 15m of it.
  Player C plays and makes 5m.
  All players play and make nothing. 
  Player D logs on. Pot rises to 20m.
  Players A through C make 10m, 5m, and 5m in play. POT EMPTY.
  Player D makes squat. Pot is empty.

No matter how much you increase m, there is some number D at which the
ability of existing players to acquire wealth has increased beyond the
capacity of the game to attract new players. At this point, players begin
leaving the game, which in turn DECREASES available wealth. Let's assume
that D now decides never to play again. When D is eventually idlenuked,
players A through C now have 20m in money that does not exist. Assuming
their ability to earn remains constant, all players signing on will now
encounter the same problem D encountered: their money, brought in for their
use by the game, is snapped up by the existing players before they get a
chance at it. Players A through C continue to become richer and richer.
Players D through whatever remain paupers and probably think this game
really bites. 

In fact, once you hit D, *everyone* probably starts thinking this game is
really pretty pointless -- since they've whacked the ceiling, it's barely
worth playing anymore. If anyone ever figures out that number of players
increase available wealth, you can probably expect a flurry of
"up-the-ante" players to be created purely to raise the pot for existing
players. If earning ability increases with experience, you hit your ceiling
and suck out the additions to the pot even faster.

>From a purely mathematical viewpoint, this approach will *never* work. Even
if the player is forced to put money back into the system, he does so
primarily to increase and extend his earning potential -- the warrior
spends the 50m on the powerful sword that allows him to kill three monsters
guarding 20m each, and now you have a 10m deficit. 

(I just saw JC's post stating that it creates a "negative feedback through
hoarding and value consolidation", which is more or less what I just said,
but I'm going to post this anyway because if I saw that statement when I
first joined this list I'd wonder why and how. Maybe this isn't as elegant
as JC's statement, but it might make more sense to someone newly examining
these ideas.)

Basically, I've found that a lot of apparently good ideas fall apart when
you start examining them with variables plugged into the various places. I
was planning a "production" concept with UU, which I mentioned here, but
earlier today I pegged a similar problem in it. Basically, the thought was
that you have a limited number of any given device, as follows:

A device exists in quantity Q. It is considered scarce when there are S of
them, and abundant when there are A of them. On any given day, a number of
devices P is produced for sale. All numbers are unsigned, so there is no
negative value problem to worry about.

If Q >= A, the device is abundant, and none are produced -- supply exceeds
demand. 

If Q <= S, 2 * P are produced -- demand exceeds supply.

If Q > S, production is more complex: P is scaled by where we are in the
range of scarcity to abundance. This is done with the formula of ( ( A - Q
) / ( A - S ) ) * 2 to obtain a double precision floating-point number
ranging from 2 to 0 depending on existing supply, which P is multiplied by.
The fractional portion is discarded. (Note that since we have established
previously that S < Q and Q < A we know that S < A and there can never be a
divide by zero.)

Whenever the number of a given device changes, the price is recalculated.
The recalculation is a bit more complex, since we can't allow it to drop
past 1/2 the existing price. 

So we take the existing price, and calculate ( ( ( A - Q ) / ( A - S ) ) *
1.5 ) + 0.5 for the scaling factor. 

Selling price is the opposite scaling factor: ( ( A - S ) / ( A - Q ) ) in
order to pay nothing when abundant (won't buy) and full normal price when
scarce. Median is half normal price. 

Now we hit the problem. The player goes and finds an expensive item that's
abundant. He buys several, making them very scarce. Then he sells them at
the scarce price, making them abundant again. Then he buys again at the
abundant price. Looking at the math above, you can figure out reasonably
quickly that every time he buys the items, he buys at half normal cost C,
and every time he sells them it's at full normal cost. 

So he makes ( C / 2 ) * Q in cash with each buy and sell cycle. Low-supply
items are particularly vulnerable to this, since if 5 of a device is
abundant and 1 is scarce we're only talking four purchases here. Such
devices also tend to be expensive, making it a lot easier for the rich to
get richer. 

The obvious solution I'm considering now is to set the price once, after
production, and then not change it over the course of the gameplay until
the next maintenance cycle. However, this still allows players to do the
above once a day: buy an object into scarcity, and then sell it the next
day. Not as bad as the initial aspect, but I'm still not wholly comfortable
with it. 

-----
| Caliban Tiresias Darklock            caliban at darklock.com 
| Darklock Communications          http://www.darklock.com/ 
| U L T I M A T E   U N I V E R S E   I S   N O T   D E A D 
| 774577496C6C6E457645727355626D4974H       -=CABAL::3146=- 


_______________________________________________
MUD-Dev maillist  -  MUD-Dev at kanga.nu
http://www.kanga.nu/lists/listinfo/mud-dev