You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Dean Herbert 5fdfee1358
Fix missed semicolon in query (#100)
3 months ago
.vscode Remove unnecessary settings file 3 months ago
bin Remove binary file 1 year ago
dependencies Remove precompiled header and restructure project accordingly 1 year ago
include/pp Allow running without rewriting existing PP 9 months ago
scripts Add new table exports to data dumps 1 year ago
src Fix missed semicolon in query (#100) 3 months ago
.appveyor.yml Fix AppVeyor build 5 months ago
.editorconfig Add .editorconfig 1 year ago
.gitattributes Fix incorrect diff annotations within .gitattributes 1 year ago
.gitignore Add vscode launch/build configs 3 months ago
.gitmodules Remove precompiled header and restructure project accordingly 1 year ago
.travis.yml Require gcc-4.9 1 year ago
CMakeLists.txt Remove precompiled header and restructure project accordingly 1 year ago
LICENCE Initial open source commit. 3 years ago Update 5 months ago

osu!performance Travis Build Status Appveyor Build status dev chat

This is the program computing “performance points” (pp), which are used as the official player ranking metric in osu!.


All that is required for building osu!performance is a C++11-compatible compiler. Begin by cloning this repository and all its submodules using the following command:

$ git clone --recursive

If you accidentally omitted the --recursive flag when cloning this repository you can initialize the submodules like so:

$ git submodule update --init --recursive

osu!performance runs on Windows, macOS, and Linux. The build environment is set up using CMake as follows.


Open the command line and navigate to the root folder of this repository.

osu-performance> mkdir build
osu-performance> cd build
osu-performance\build> cmake ..

Now the build folder should contain a Visual Studio project for building the program. Visual Studio 2017 and a 64-bit build are recommended (cmake -G "Visual Studio 15 2017 Win64" ..).

macOS / Linux

On macOS / Linux you need to install the MariaDB MySQL connector and cURL packages. Afterwards, in a terminal of your choice, do

osu-performance$ mkdir build
osu-performance$ cd build
osu-performance/build$ cmake ..
osu-performance/build$ make -j

Sample Data

Database dumps with sample data can be found at This data includes the top 10,000 users along with a random 10,000 user sample across all users, along with all required auxiliary tables to test this system. Please note that this data is released for development purposes only (full licence details available here).

You can import these dumps to mysql (after first extracting them) by running cat *.sql | mysql. Note that all existing data in tables will be dropped and replaced. Make sure to import the latest available data dumps as older snapshots may be incompatible with the latest version of osu!performance.


First, set up a MySQL server and import the provided data from above which is most relevant to your use case. Next, edit bin/config.json with your favourite text editor and configure mysql.master to point to your MySQL server.

After compilation, an executable named osu-performance is placed in the bin folder. You can use it via the command line as follows:

./osu-performance COMMAND {OPTIONS}

where command controls which scores are the target of the computation. The following commands are valid:

  • all: Compute pp of all users
  • new: Continually poll for new scores and compute pp of these
  • scores: Compute pp of specific scores
  • users: Compute pp of specific users

The gamemode to compute pp for can be selected via the -m option, which may take the value osu, taiko, catch, or mania.

Information about further options can be queried via

./osu-performance -h

and further options specific to the chosen command can be queried via

./osu-performance COMMAND -h

Configuration options beyond these parameters, such as various API hooks, can be adjusted in bin/config.json.


osu!performance is licensed under AGPL version 3 or later. Please see the licence file for more information. tl;dr if you want to use any code, design or artwork from this project, attribute it and make your project open source under the same licence.

Note that the sample data is covered by a separate licence.