This dice rolling simulator was designed as a digital aid for playing roleplaying games, and named "droll" as an abbreviation for "die roll" or "dice roller". The term "droll" also seemed apt as a reference to its beginnings as an amusing little toy program, though it now offers reasonably sophisticated support for different dice-rolling techniques used in various RPGs. It is a Ruby library with included command line interface (droll) and IRC dicebot (drollbot).

At present, droll provides a number of different die rolling techniques, including simplistic roll-and-sum, various exploding roll schemes, counting dice that meet a threshhold, and rolling several dice then discarding some by threshhold before summing. Threshholds are typically the highest possible number for the die type, or the maximum possible for the sum in some cases, or even the lowest possible number for the die type, though alternate threshholds can be used.

Use NdN as the pattern for normal die rolls (where N stands in for a number), and NxN for exploding die rolls, for instance. Modifiers can be included as well: NdN+N or NdN-N. Die codes, at this time, do not work if they have spaces in them. Certain die codes are rejected (i.e. 0x0). Exploding is limited to an unreasonably high number (1000), to prevent crashing. Specifying alternate threshholds uses a .N syntax (NxN.N). Die values are chosen (pseudo)randomly from numbers between 1 and N (the number following the x or d), or between 0 and N if the N is preceded by a 0 character. An example of the full sophistication of die code parsing is:


This would roll four virtual dice numbered 0-5, and each die would explode on any 4 or 5 result. All die results are added together, and 7 added to the total to yield a final number.

The first number following each 4 or 5 in this example is the result of a die immediately rolled to handle exploding die values.


Installing the Ruby programming language's runtime is a relatively simple operation, and a prerequisite for running droll (because it is a Ruby project). The "standard" reference implementation, often called MRI/YARV (for Matz' Ruby Implementation + Yet Another Ruby VM) is available in the default software management system of most open source Unix-like operating systems as well as Apple MacOS X, though you should make sure your system uses the 1.9 version of ruby instead of the older 1.8 version (see below). For users of Microsoft Windows, the RubyInstaller for Windows makes Ruby easy to install there, as well.

The normal way to install Ruby on most operating systems will also install the gems tool, which is a sort of software management system specific to Ruby tools, and is the way most Ruby libraries and applications are distributed, including droll.

Once you have Ruby installed with rubygems, installing droll should be easy. Just use the gem command:

$ gem install droll

You can also download the gem package from the Bitbucket project and use the gem command to install it:

$ gem install droll-<version>.gem

In this example, replace <version> with the version number in the name of the gemfile you downloaded. For version 1.0.0, for instance, the command might look like this (though as of this writing it is not yet at version 1.0.0):

$ gem install droll-1.0.0.gem

Note that the $ character indicates your shell prompt, and is not part of the command. Depending on your setup, you may need to use sudo, log in as root, or engage in some other additional activities to ensure the gem is installed correctly. If you know of different requirements for installation on other systems, please feel free to submit patches to this README file via one of the approaches detailed in the contributions section at the bottom of this file.

Ruby 1.8

Droll assumes Ruby 1.9.x, and some die code validation (for zero-based die codes, e.g. 1d05) does not work properly with older Ruby versions. Normally, it will not install on a system using a version of Ruby older than 1.9, but this can be overridden if you wish by using the -f option with the gem install command:

$ gem install -f droll


The following sections explain how to use the executable tools that come with the droll library: a droll command line interface and a drollbot IRC dicebot.

droll command line

Using the basic droll program from the command line is pretty simple:

$ droll d20
d20: [15] + 0 = 15
$ droll 2d4+3
2d4+3: [1, 2] + 3 = 6
$ droll 1d10 2d03
1d10: [1] + 0 = 1
2d03: [3, 2] + 0 = 5
$ droll 4x05.4+7
4x05.4+7: [5, 2, 0, 4, 0, 1] + 7 = 19

In each example, the numbers between [brackets], separated by commas, are the numberic results for each of the individual dice rolled. As shown by the third example of using the command line droll program above, multiple die codes can be listed on a single line, with the results of each roll being shown on a separate line of output. In the last example, the first number following each 4 or 5 in this example is the result of a die immediately rolled to handle exploding die values.

drollbot IRC dicebot

The drollbot interface to the functionality of the droll library does some things a little differently. For instance, any input to a channel monitored by drollbot will be checked to see if it starts with a die code that drollbot understands. If so, it will parse that line as a command to produce die roll output. Starting a line with droll or drollbot will not have the same effect. As such, this may occur:

12:38 < apotheon> 1d20+3
12:38 < drollbot> <apotheon> rolls 1d20+3: [10] + 3 = 13

The output from drollbot is a touch more verbose; this is an example of how it works within a standard IRC channel. Private messages may be sent to drollbot, with slightly differently formatted output:

12:40 <apotheon> /msg drollbot 1d20+3
12:40 <drollbot> result of 1d20+3: [13] + 3 = 16

Note that this example is not perfectly representative. The /msg command is sent from an IRC channel where the drollbot instance is logged in, and the result message appears in a PM that is not visible to anyone else in the IRC channel where the dicebot was invoked.

In either case, a comment may be appended to the end of the line. Within an IRC channel:

12:42 < apotheon> 1d20+3 crappy save
12:42 < drollbot> <apotheon> rolls 1d20+3: [17] + 3 = 20 (crappy save)

Drollbot will echo any text following your die code back to you as a parenthetical note. This applies to die codes as well as other text so that, unlike with the command line droll utility, multiple rolls cannot be specified in a single command:

12:43 < apotheon> 2x05 d20+3 2d4 many
12:43 < drollbot> <apotheon> rolls 2x05: [4, 2] + 0 = 6 (d20+3 2d4 many)

To start drollbot, all you need to do is ensure that it is configured with an IRC network to which it should connect, and some channel names it should join, then run the program. An example configuration file is included with the installed gem. After installing the droll gem, you can find the example configuration file in the directory hierarchy where the gem was installed, under the etc subdirectory. The example configuration file is called drollbot.conf.sample. Use the drollbot -c or drollbot --config command to see more information about configuring drollbot.


Droll project files may be redistributed under the terms of the Open Works License. The OWL was chosen with a conscious adherence to copyfree policies. See the Copyfree Initiative site for more details about the copyfree philosophy of licensing.

See the COPYING file in the project repository for more information about copyright and licensing for droll and the code associated with it.


If you wish to contribute to the project, please feel free to do so.

Mercurial (in the form of the hg tool) offers a simple way to produce a patch, using the export command. To produce a patch based on the most recent commits in your local clone of the project, this command should suffice to produce a usable patch for an update from the immediately previous commit:

$ hg export tip
  • Patches may be submitted via the issue tracker, as a comment with attachment in response to whatever issue it fixes.
  • Patches can also be submitted to Chad Perrin via email, using the code at apotheon dot net address.
  • Another way to submit contributions is to clone the project on Bitbucket and send a "pull request" from the clone when changes have been made. The process of submitting contributions via fork and pull request is described in a TechRepublic article: Contribute to Bitbucket Projects Using Forks and Pull Requests.