Documentation | Uses | Binaries | Source Code | GeoWalls & Domes | Publications | Licensing | Mailing List | Misc. | Links

Partiview is free, open-source software from the National Center for Supercomputing Applications (NCSA) at the University of Illinois Urbana-Champaign. It is an industrial strength, interactive, mono- or stereoscopic viewer for 4-dimensional datasets. It is written in C++/OpenGL and has been compiled to run on Linux, Windows, OS X, and Irix.

The prime author and current maintainer is Stuart Levy (slevy {at}, research programmer at the NCSA. Other contributors to the source code include Steven Marx, from the American Museum of Natural History/Hayden Planetarium (AMNH), who added improvements to the user interface, developed the Mac OS X port, and worked on a fish-eye projection of Partiview. Toshiyuki Takahei of RIKEN wrote a Director plugin and helped Marx port Partiview to OS X. Steve McMillan of Drexel University wrote the "kira" library that partiview uses to draw results of stellar dynamics calculations.

Others who have contributed a great deal to design, management, and documentation issues are Donna Cox (NCSA), Peter Teuben (Maryland), and Brian Abbott (Hayden/AMNH).


To get Partiview to display your data, the data has to be wrapped in some ascii (plain text) files. Here's where you can find details of the wrappers.

Beginners : the Partiview User Guide by Brian Abbott.

Intermediate : the Partiview Reference Manual (HTML) (or PDF) by Peter Teuben and Stuart Levy; or Peter Teuben's collection of partiview and related reference material.

Other miscellaneous documentation from Cosmus.

Some past emails from Stuart about commands for including Maya OBJ files, the warp command to allow the three spatial coordinates to be arbitrary weighted combinations of as many dimensions as you specify (and change the weights without having to restart partiview), compiling partiview with static libraries,...

Also have a look at the archives of the mailing lists for partiview and the Digital Universe (once the archives exist, anyway...)

Back to top of page

Uses of Partiview

There are many things one can do with a fast 3d+time viewer. Here are some examples that have free downloads and documentation.


Digital Universe by Brian Abbott, Carter Emmart, and Ryan Wyatt of Hayden/AMNH. Want an accurate planetarium on your desk? Get this. Press:
Transneptunian asteroids and dwarf planets, including the first clearly interstellar visitor, asteroid 'Oumuamua in 2017. It arrived from roughly the direction of Vega, and you can see that in the image at left.

The file contains a sample table of orbital elements, partiview .speck files suitable for embedding in the AMNH Digital Universe data, and a stand-alone version of the same, with orbits embedded in the Hipparcos stars. It also provides a tool for adding new orbital elements using text copy-and-pasted from Wikipedia articles on transneptunian bodies.

Sloan Digital Sky Survey & WMAP in Stereo by Cosmus (Mark SubbaRao, Dinoj Surendran and Randy Landsberg of Adler Planetarium and the University of Chicago). While not as detailed as the DU, it is prettier - galaxies are represented by photographs instead of points. It was the first partiview dataset to work on a geowall.
Stars around a Black Hole. Animation of stars based on observations by Andrea Ghez and Jessica Lu (UCLA Galactic Center Group). These observations were used to predict precisely the location of a black hole at the center of our Milky Way.
Walking to the Stars. Scale the universe so that you can walk to the nearby stars. Imagine: how large is the Solar System then? Or, look at the stars of the Summer Triangle, Deneb, Vega and Altair: all are roughly equally bright in our sky, but does that mean they're all roughly equally far away?
HIPPARCOS stars with space motions. Thousands of stars have not only their positions, but their full space motions, measured. Run this animation to extrapolate those velocities, at (by default) 10,000 years per second.
John Kielkopf at the University of Louisville Physics Department uses Partiview to help his students explore real data in his Introduction to Astrophysics course.
Viewing the Canis Major Dwarf. Nicolas Martin and Rodrigo Ibata of the Observatoire de Strasbourg used Partiview and Nemo to create pictures and animations of their discovery with other astronomers in Europe and Australia of the nearest galaxy to the Milky Way in November 2003.

Planet Stuff

Overlay a rectangular map of a world on a simple sphere and - voila! - you have a planet.
Planets for Partiview by Selden Ball. Earth (with clouds), Mars, and Jupiter.
Earth Views by Cosmus. With simple button clicks you can easily switch between different views of earth on a single globe - night vs day, 20 000 years ago vs today.
Planetary Scales by Cosmus. Russian dolls model of Sol and its (first) nine planets, to scale.


Global Networks

PlanetLab Overlays by Dinoj Surendran and Matei Ripeanu (UChicago). Using a 3d viewer allows the height of a link to represent information - in this case bandwidth.



AIRES Simulations by Cosmus. AIRES is a standard package used by physicists to produce the positions and times of particles in the air shower created when a cosmic ray/particle hits the atmosphere. Partiview is used to visualize its output. The resulting shower models, and short movies made with them, are used by for outreach and presentations by members of the Pierre Auger and VERITAS projects.

Machine Learning

Partiview makes textured 3d scatterplots usable. Three-dimensional scatterplots are normally clunky to use, but not with partiview. And having textured glyphs is really cool - sometimes even useful - for data with a natural visual representation. These, and other features, turn out to be very useful for testing the output of machine learning algorithms.
Clustering patterns. How well can a computer recognize handwritten digits written by humans? Partiview views the output of one particular algorithm on five thousand different digits. A similar demo was used by a keynote speaker (Dimitris Achlioptas of Microsoft) at the ECML/KDD 2004 conference.


  • NIPS 2004 demo abstract by Dinoj Surendran & Stuart Levy. The lowdown on what features of partiview are useful for a machine learning audience.
  • Matlab Interface to produce partiview format files given data matrices. Matlab is a standard working environment for researchers in machine learning, speech recognition, and other fields.

Remote Collaboration

Partiview was used with Virtual Director by groups in Illinois and New York to meet online in a virtual environment and simultaneously design the space show "The Search for Life: Are We Alone?" for Hayden Planetarium.

Back to top of page


Binaries for partiview for Windows, Linux, and OS X can be downloaded from the Digital Universe site.

Windows (Dec 2013) (zipped, 1.5MB).

(older) Mac OS X (10.5), Intel x86, .tar.gz (November 2010) (zipped, 0.4Mb).

Mac OS X (10.7-10.12), Intel x86_64, .tar.gz (Aug 2017) (tar.gz, 1.2Mb).

(older) Linux Intel x86 (November 2010) (zipped, 0.6Mb)

Linux Intel x86-64, RedHat 5 (Dec 2013) (tar.gz, 0.9Mb)

Back to top of page

Source code

Partiview includes a data-handling and display library that can be embedded in other systems such as Virtual Director.

To get the source code, check out the git version - literally - with the following command:

  git clone

To compile Partiview you'll also need FLTK 1.1.x or later (current is 1.3.x), from

Here's an example session for compiling partiview on linux, on a system where you aren't root, and where you have to install FLTK yourself. We start with assuming that you're in your home directory /home/elmo

  1. cd /home/elmo
  2. Type the git command above. This creates the directory /home/elmo/partiview
  3. Download some version of the fltk.1.3.x source code from Untar/zip/bz it into a new directory, say /home/elmo/fltk.1.3
  4. cd /home/elmo/fltk.1.3
  5. ./configure
  6. make
  7. cd /home/elmo/partiview/src
  8. ./configure --with-fltk=/home/elmo/fltk.1.3
  9. make
  10. This should result in an eponymous partiview binary in the /home/elmo/partiview/src directory that you can move to wherever you want.

For more details, such as how to compile Partiview on Windows, see the Reference Guide.

Back to top of page

Using Partiview on GeoWalls and Planetarium Domes

Partiview supports several stereo features, including red-cyan, chromadepth, and side-by-side.

The side-by-side feature allows Partiview to be used on low-cost dual-projector polarization-based stereo systems such as the GeoWall. Documentation on that can be found here.

Partiview does not work on planetarium domes. However, some major makers of digital domes - such as Sky Skan, Evans & Sutherland - have written beta-stage plugins that convert files in Partiview format to the native format of their dome projector software.

Brian Abbott and Ryan Wyatt treated attendees of the 10th Great Lakes Planetarium Conference to a tour of their Partiview-based Digital Universe on a Sky Skan dome in October 2004 at the Detroit Science Museum.

Therefore, if you're an astronomer and want your work to be usable on a digital dome, publishing it in Partiview format is a good start.

Back to top of page


"Interactive 3-D visualization of particle systems with Partiview" by Stuart Levy, 2001. Appeared in the Proceedings (Vol 208) of the International Astronomical Union Symposium on "Astrophysical Supercomputing Using Particles".

Immersive 4D Interactive Visualization of Large-Scale Simulations by Peter Teuben, Piet Hut, Stuart Levy, Jun Makino, Steve McMillan, Simon Portegies Zwart, Mike Shara, Carter Emmart. 2001. Arxiv preprint astro-ph/0101334. Also in ASP Conf. Ser., Vol. 238, Astronomical Data Analysis Software and Systems X, ed. F.R. Harnden, Jr., F.A. Primini, & H.E.Payne (San Francisco: ASP) p 499.

Visualizing High-dimensional datasets with Partiview by Dinoj Surendran and Stuart Levy, 2004. Poster Proceedings of Information Visualization 2004.

Visualising and Analysing Massive Astronomical Datasets with PartiView, Charles Liu (AMNH). A short powerpoint presentation at a Virtual Observatory conference in 2002.

More partiview-related references can be found with Google Scholar.

Back to top of page


Terms and conditions for the use and distribution of the data viewer, Partiview, are in the Illinois Open Source License below. Partiview users are subject to its non-onerous terms.
Copyright 2002-2003  NCSA, University of Illinois 
Urbana-Champaign, All rights reserved.

Developed by:
        Stuart Levy, NCSA Virtual Director Group
        University of Illinois Urbana-Champaign

Permission is hereby granted, free of charge, to any person 
obtaining a copy of this software and associated documentation 
files (the "Software"), to deal with the Software without 
restriction, including without limitation the rights to use, 
copy, modify, merge, publish, distribute, sublicense, and/or 
sell copies of the Software, and to permit persons to whom the 
Software is furnished to do so, subject to the following conditions:

  * Redistributions of source code must retain the above 
    copyright notice, this list of conditions and the 
    following disclaimers.

  * Redistributions in binary form must reproduce the above 
    copyright notice, this list of conditions and the following 
    disclaimers in the documentation and/or other materials 
    provided with the distribution.

  * Neither the names of NCSA, University of Illinois 
    Urbana-Champaign, nor the names of its contributors may 
    be used to endorse or promote products derived from this 
    Software without specific prior written permission.


Back to top of page

Mailing List

Google Groups Subscribe to partiview
Browse Archives at


Partiview is an acceptable entry format for the NSF's Science and Engineering Visualization Challenge - something that is not at all obvious from their rulebook. This is because Partiview files can be run off a CD in Windows without requiring anything to be copied onto the machine it runs on.

Partiview can be called from Powerpoint by embedding it in an object.

Back to top of page


Here are several members of the partiview community mentioned on this page:

At Hayden Planetarium at the American Museum of Natural History:

John Kielkopf, Department of Physics, University of Louisville

Peter Teuben at the Astronomy Department at the University of Maryland.

Steve McMillan of the Physics Department at Drexel University.

Cosmus is an informal group of individuals from Adler Planetarium & Science Museum, and the University of Chicago.

Related Software:

Other Partiview sites:

Other projects from NCSA's Advanced Visualization Lab (AVL) are described here.

Back to top of page

Send brickbats for page design to DS. Dec '04

Comments on content? Email Stuart Levy, slevy{AT}