| abstract
| - This page is about the mechanics of how a Treasure Trail's rewards are determined, and calculations from this to determine drop chances of each reward. Most clues operate on a series of drop tables with associated chances to access. They start with an initial table which contains all the rewards common to all levels of trails - sweets, meerkats, god pages, etc - and a chance to enter the clue's rare table, containing all of the clue's specific rewards (some of which may be on subtables within the rare table). The vocabulary used here can be confusing, so here are a number of definitions of terms used on this page:
* DroptableA list of things that can be dropped and their associated chances. Can contain another droptable.
* (Droptable) slotOne of the things on the droptable. This may be an item or another droptable.
* Reward spotA place on the clue reward interface which can contain an item. A clue generates at least 2 of these per reward. Each tier of clue has a minimum number of reward spots it will create, and then it will add 0-2 extra spots randomly.If the same item is generated for multiple spots, they will combine into one stack (commonly occurs with runes), which may make a clue appear to generate less than it should.
* 1/xThe value given in the column is the x value in the 1/x. So a value of 15 in the column means the chance is 1/15, or 1-in-15.
* x SFThe value in the column has been rounded to x significant figures.
* RollGenerate a random number in order to choose a droptable slot, or other number needing to be generated randomly. Each number is equally likely to occur, giving rolls a uniform distribution.
* Any text with a dotted underline has hover text with further explanation. In order to calculate the per-clue chances, one needs to use a binomial distribution, taking the number of trials n to be the average number of reward spots of that clue tier, and p to be the per-spot chance of getting the items, then calculating the probability of receiving greater than zero of the item. In statistics notation: Luckily, most spreadsheet programs can do this natively. One can verify these values using a Wolfram Alpha query, such as this for barrows dye from a hard clue; in Microsoft Excel or Google Sheets, this is performed using =1-BINOMDIST(0,5,3/25600,true) (see documentation: Excel, Sheets).
- This page is about the mechanics of how a Treasure Trail's rewards are determined, and calculations from this to determine drop chances of each reward. Most clues operate on a series of drop tables with associated chances to access. They start with an initial table which contains all the rewards common to all levels of trails - sweets, meerkats, god pages, etc - and a chance to enter the clue's rare table, containing all of the clue's specific rewards (some of which may be on subtables within the rare table). The vocabulary used here can be confusing, so here are a number of definitions of terms used on this page:
* DroptableA list of things that can be dropped and their associated chances. Can contain another droptable.
* (Droptable) slotOne of the things on the droptable. This may be an item or another droptable.
* Reward spotA place on the clue reward interface which can contain an item. A clue generates at least 2 of these per reward. Each tier of clue has a minimum number of reward spots it will create, and then it will add 0-2 extra spots randomly.If the same item is generated for multiple spots, they will combine into one stack (commonly occurs with runes), which may make a clue appear to generate less than it should.
* 1/xThe value given in the column is the x value in the 1/x. So a value of 15 in the column means the chance is 1/15, or 1-in-15.
* x SFThe value in the column has been rounded to x significant figures.
* RollGenerate a random number in order to choose a droptable slot, or other number needing to be generated randomly. Each number is equally likely to occur, giving rolls a uniform distribution.
* Any text with a dotted underline has hover text with further explanation. In order to calculate the per-clue chances, one needs to use a binomial distribution, taking the number of trials n to be the average number of reward spots of that clue tier, and p to be the per-spot chance of getting the items, then calculating the probability of receiving greater than zero of the item. In statistics notation: Most spreadsheet programs can do this natively. One can verify these values using a Wolfram Alpha query, such as this for barrows dye from a hard clue; in Microsoft Excel or Google Sheets, this is performed using =1-BINOMDIST(0,5,3/25600,true) (see documentation: Excel, Sheets).
|
![[RDF Data]](/fct/images/sw-rdf-blue.png)