The RGL package is a visualization device system for R, using OpenGL or WebGL as the rendering backend. An OpenGL rgl device at its core is a real-time 3D engine written in C++. It provides an interactive viewpoint navigation facility (mouse + wheel support) and an R programming interface. WebGL, on the other hand, is rendered in a web browser; rgl produces the input file, and the browser shows the images.
A pkgdown website is here:
https://dmurdoch.github.io/rgl/
The unreleased development version website is here:
https://dmurdoch.github.io/rgl/dev/
See this
vignette for
details on producing your own pkgdown website that includes rgl
graphics.
The currently active development site is here:
Most users will want to install the latest CRAN release. For Windows, macOS and some Linux platforms, installation can be easy, as CRAN distributes binary versions:
# Install latest release from CRAN
install.packages("rgl")
To install the latest development version from Github, you’ll need to do a source install. Those aren’t easy! Try
# Install development version from Github
remotes::install_github("dmurdoch/rgl")
If that fails, read the instructions below.
Currently installs are tested on older R versions back to R 3.5.x, but
this version of rgl may work back as far as R 3.3.0.
The software is released under the GNU Public License. See COPYING for details.
- portable R package using OpenGL (if available) on macOS, Win32 and X11
- can produce 3D graphics in web pages using WebGL
- R programming interface
- interactive viewpoint navigation
- automatic data focus
- geometry primitives: points, lines, triangles, quads, texts, point sprites
- high-level geometry: surface, spheres
- up to 8 light sources
- alpha-blending (transparency)
- side-dependent fill-mode rendering (dots, wired and filled)
- texture-mapping with mipmapping and environment mapping support
- environmental effects: fogging, background sphere
- bounding box with axis ticks marks
- undo operation: shapes and light-sources are managed on type stacks, where the top-most objects can be popped, or any item specified by an identifier can be removed
macOS Windows 7/10 Unix-derivatives
R recommended tools (gcc toolchain) On Windows, Rtools40 (or earlier versions for pre-R-4.0.0)
For OpenGL display:
Windowing System (unix/x11 or Windows)
OpenGL Library
OpenGL Utility Library (GLU)
For WebGL display:
A browser with WebGL enabled. See https://get.webgl.org.
Debian and variants including Ubuntu:
aptitude install libgl1-mesa-dev libglu1-mesa-dev
Fedora:
yum install mesa-libGL-devel mesa-libGLU-devel libpng-devel
macOS:
Install XQuartz.
rgl should work with XQuartz 2.7.11 or newer, but it will probably
need rebuilding if the XQuartz version changes. XQuartz normally needs
re-installation whenever the macOS version changes.
Windows:
Windows normally includes OpenGL support, but to get the appropriate include files etc., you will need the appropriate version of Rtools matched to your R version.
The libpng library version 1.2.9 or newer is needed for pixmap import/export support.
The freetype library is needed for resizable anti-aliased fonts.
Binary builds of rgl are available for some platforms on CRAN.
For source builds, install the prerequisites as described above, download the tarball and at the command line run
R CMD INSTALL rgl_1.3.15.tar.gz
(with the appropriate version of the tarball). The build uses an
autoconf configure script; to see the options, expand the tarball and
run ./configure --help.
Alternatively, in R run
install.packages("rgl")
to install from CRAN, or
remotes::install_github("dmurdoch/rgl")
to install the development version from Github.
To build on MacOS using one of the ARM64 chips (currently M1, M2 or M3),
follow the instructions on https://mac.r-project.org/tools/ to install
the tools and libraries into /opt/R/arm64. It is important that
/opt/R/arm64/bin appear in your PATH before /usr/local/bin if the
latter directory has been used for x86_64 installs. If you don’t do
this, or have some other error in setting things up, you’ll get a
warning during rgl installation saying that some configure test
failed, and rgl will be installed without OpenGL support.
Some versions of RStudio (including 2024.04.2+764) have a bug that
modifies your PATH on startup and again after every package
installation, putting /usr/local/bin at the head of the PATH. If you
are building rgl in such a system you need to remove files from
/usr/local/bin if there’s a file with the same name in
/opt/R/arm64/bin. Hopefully this bug will be fixed soon!
As of version 0.104.1, it is possible to build the package without OpenGL support on Unix-alikes (including macOS) with the configure option –disable-opengl For example,
R CMD INSTALL --configure-args="--disable-opengl" rgl_1.3.15.tar.gz
On Windows, OpenGL support cannot currently be disabled.
library(rgl)
browseVignettes("rgl")
demo(rgl)
Daniel Adler dadler@uni-goettingen.de
Duncan Murdoch murdoch@stats.uwo.ca
Oleg Nenadic onenadi@uni-goettingen.de
Simon Urbanek simon.urbanek@math.uni-augsburg.de
Ming Chen mchen34@uwo.ca
Albrecht Gebhardt albrecht.gebhardt@uni-klu.ac.at
Ben Bolker bolker@zoo.ufl.edu
Gabor Csardi csardi@rmki.kfki.hu
Adam Strzelecki ono@java.pl
Alexander Senger senger@physik.hu-berlin.de
The R Core Team for some code from R.
Dirk Eddelbuettel edd@debian.org
The authors of Shiny for their private RNG code.
The authors of knitr for their graphics inclusion code. Jeroen Ooms
for Rtools40 and FreeType help.
Yohann Demont for Shiny code, suggestions, and testing.
Joshua Ulrich for a lot of help with the Github migration. Xavier
Fernandez i Marin for help debugging the build.
George Helffrich for draping code.
Ivan Krylov for window_group code in X11.
Michael Sumner for as.mesh3d.default enhancement.
Tomas Kalibera for winutf8 and other help.
David Hugh-Jones for documentation improvements.
Trevor Davis for a snapshot3d patch.
Mike Stein for pointer-handling code.
Jonathon Love for the uname patch.
The Mapbox team for the triangulation code.