Up: Downloadable Software from the Geometry Center

Qhull Downloads

• Qhull Home Page

  Qhull computes convex hulls, Delaunay triangulations, halfspace intersections about a point, Voronoi diagrams, furthest-site Delaunay triangulations, and furthest-site Voronoi diagrams. It runs in 2-d, 3-d, 4-d, and higher dimensions. It implements the Quickhull algorithm for computing the convex hull. Qhull handles roundoff errors from floating point arithmetic. It can approximate a convex hull.

  Visit Qhull News for news, bug reports, change history, and users.

• Download: Qhull 2003.1 for Windows (1.2 Mbytes)

  Type: console programs for Windows 95, 98, ME, 2000, NT, XP

  Includes executables, documentation, sources files, and a cygwin Makefile. It runs in a DOS window.

• Download: source for Qhull 2003.1 (600K)

  Type: C source code for 32-bit and 64-bit architectures

  Includes documentation, source files, and a simple Makefile

• Download: Qhull 2002.1 for Unix

  Type: C source code for Unix systems (Debian configure)

  Includes documentation, source files, and a configure Makefile. Includes Debian configuration files. Includes downloads of Qhull's current and previous versions.

• Download: Qhull version 2002.1 i686 rpm (643K)

  Type: rpm build for Mandrake 8.2 and RedHat 7.3

  For other Linux systems, use the src rpm [L. Mazet]

• Qhull@Savannah

  Type: CVS repository

  CVS repositiory of Qhull sources, documentation, and Makefiles.

• Spanish mirror site

  Download Qhull's current and previous versions from Servidor WWW de la Escuela de Minas of the Universidad de Oviedo.

• Download: Article about Qhull (210K)

  Type: various formats on CiteSeer

  Barber, C.B., Dobkin, D.P., and Huhdanpaa, H.T., "The Quickhull algorithm for convex hulls," ACM Transactions on Mathematical Software, vol. 22, pp. 469-483, Dec 1996 [abstract].

• Download: Qhull version 1.0 (92K)

  Type: C source code for 32-bit architectures

  Version 1.0 is a fifth the size of version 2.4. It computes convex hulls and Delaunay triangulations. If a precision error occurs, it stops with an error message. It reports an initialization error for inputs made with 0/1 coordinates.

  Version 1.0 compiles on a PC with Borland C++ 4.02 for Win32 and DOS Power Pack. The options for rbox are "bcc32 -WX -w- -O2-e -erbox -lc rbox.c". The options for qhull are the same. [D. Zwick]

• Download: Qhull version 1.0 (Mac) (180K)

  Type: C source code, packaged for Macintosh w/68881

  The Macintosh implementation of Qhull 1.0 reads input from a file and returns the results to a file. The code is compiled for a 68881 floating point coprocessor. The package includes Think C projects for Qhull and the rbox input generator.

Up: Downloadable Software from the Geometry Center

The Geometry Center Home Page

Comments to: qhull@qhull.org