There is a web interface for Troll. This is the easiest way to use Troll, and is sufficient for most users.
You can roll dice and calculate probabilities. There are some limitations on probability calculations to avoid stressing the server, so calculations that use a lot of time. This will affect some (but not all) calculations involving many dice or rerolls. If you get into this limitation, you can run Troll on your own computer (by downloading and compiling the sources below), but that will be with a more primitive user interface.
I have combined my research into domain-specific languages and my interest in board games and role-playing games to design and implement a language for specifying dice-roll mechanisms.
The first version of the language Roll was released in early 2002 and a paper about the design and implementation was presented at the conference Practical Aspects of Declarative Languages in New Orleans, January 2003. Roll was used to calculate probabilities for several alternative dice-roll mechanisms during the development of the new version of the RPG World of Darkness system from White Wolf.Based on experiences with Roll, I in 2006 redesigned the language to make it easier to use and added a few new features. The result is Troll, which is the current version, alebit extended with many new features since its release.
In Troll, you can write a definition for a dice-roll method
like sum largest 3 4d6 (corresponding to a common method
for rolling attributes in Dungeons &
Dragons) and either get randomly generated results based on this
definition or get a probability distribution:
Value % = % >=
3 : 0.0771604938272 100.0
4 : 0.308641975309 99.9228395062
5 : 0.771604938272 99.6141975309
6 : 1.62037037037 98.8425925926
7 : 2.93209876543 97.2222222222
8 : 4.78395061728 94.2901234568
9 : 7.02160493827 89.5061728395
10 : 9.41358024691 82.4845679012
11 : 11.4197530864 73.0709876543
12 : 12.8858024691 61.6512345679
13 : 13.2716049383 48.7654320988
14 : 12.3456790123 35.4938271605
15 : 10.1080246914 23.1481481481
16 : 7.25308641975 13.0401234568
17 : 4.16666666667 5.78703703704
18 : 1.62037037037 1.62037037037
Average = 12.2445987654 Spread = 2.84684444531 Mean deviation = 2.31853947569
Note: Requires Moscow ML to be installed.
Last update: July 16 2021.
Last update: March 22 2021.
Note: Requires ML server pages.
Last update: April 3 2018.
Last update: April 28 2009.
Last update: April 24 2009.
Last update: March 15 2016.