Re: GURPS Dungeon Fantasy 8: Treasure Tables
It's a finite number of treasures with conceivably infinite variation in a few of the types. You just have to do what particle physicists do and renormalize. In this case, counting each gem as "1" instead of as "an infinite spectrum of gem values" works just peachy. Events where you get very high values are highly suppressed anyway, just like the obscure physical phenomena that physicists tend to write off; if you only have a 1/11019960576 chance of getting 17 times the value, you can pretty much call that "0" in safety.
|
Re: GURPS Dungeon Fantasy 8: Treasure Tables
Quote:
It is assumed that the GM, confronted with a long string of "bigger" results, will exercise judgment and either examine the dice for tampering and say "screw this, it's worth a lot, let's move on." |
Re: GURPS Dungeon Fantasy 8: Treasure Tables
Quote:
Sorry, I had to say it ;) |
Re: GURPS Dungeon Fantasy 8: Treasure Tables
Quote:
|
Re: GURPS Dungeon Fantasy 8: Treasure Tables
Quote:
"Hey PCs! Yoo hoo! See this treasure? Isn't it shiny? I double dog dare you to get a reasonable deal on it. Anyone capable of paying even 1% of it's value could squash you like a bug and take it. After you've sold it, the buyer won't want it known they've bought it (for their own security), so there will almost certainly be doubt about the truth of the sale. Certain other folks who might want it will check if you have it 'just to be sure.' Some of them will check in unpleasant ways. Go on. Find a buyer. I'll be waiting behind my screen, smirking evilly." |
Re: GURPS Dungeon Fantasy 8: Treasure Tables
Quote:
|
Re: GURPS Dungeon Fantasy 8: Treasure Tables
OK, I was really just showing that a simple mathematical argument, without sufficently well defined criteria can be used to prove whatever you really want.
Media statistics are my favourite abuses! |
Re: GURPS Dungeon Fantasy 8: Treasure Tables
Quote:
There is no mystery here, even though it's easy to get things wrong. With a d2 instead of a d6 in "my" game the values increase too fast, so the mean value goes to infinity. (Then you get the "St. Petersburg paradox".) |
Re: GURPS Dungeon Fantasy 8: Treasure Tables
The minimum possible value of a log-normal distribution is 1/infinity higher than zero.
The maximum value is infinity. And yet, it's quite possible to get the mode, mean, and median of this distribution, and those values are finite. In a statistics program called Minitab, I just generated a normal distribution centered around 4 with a sigma of 1. Taking 10^[That number] gives a log normal distribution with an expected median of 10,000. In this case, the minimum was listed as "0," which is really "so small that the computer rounds it to zero," while the maximum is over 210 million. Hrm. I suppose this means that a truly infinite population for this disty would have a median (10,000) but no average, since in theory you can generate an infinite upper bound, but a finite lower bound. Hrm. |
Re: GURPS Dungeon Fantasy 8: Treasure Tables
Quote:
This can work just fine in any game, especially DF. Just decide what the best merchant has in his wallet, and it doesn't matter if the PCs find a $16,000 gemstone. Well, not unless one of them decided to buy Jeweler-25 so he can break it down into smaller ones without ruining anything. :) |
| All times are GMT -6. The time now is 04:42 PM. |
Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2024, vBulletin Solutions, Inc.