Free vertex software for linux

Divmod Vertex is an implementation of Q2Q, a protocol for opening authenticated connections, even through NAT.

It allows a user to reliably demonstrate their identity (for distributed authentication) and receive real-time data directly from other users, and provides a mechanism for a user to decide whether they want to expose their IP address to a third party before accepting a peer-to-peer connection.

Divmod Vertex is byte-stream oriented and application-agnostic. Any peer-to-peer application can use Q2Q to open connections and deliver messages.

Sauerbraten is an experimental engine based on Cube, can be seen as "Next-Gen Cube", or "Cube 2". Much like cube, the aim of this engine is not to produce the most eyecandy possible, but rather allow map/geometry editing to be done dynamically in-game, and make map editing a lot of fun.

Sauerbraten has an even simpler world model than cube (fewer exceptions, just one kind of building block), is quicker to edit geometry with, yet allows for significantly greater class of shapes. One way to see the transition from Cube to Sauerbraten is to say Cube was a 2-directional heightfield (floor and ceiling), and Sauerbraten is a 6-directional heighfield (heighfields can be modeled in all 6 directions).

The world consists of an octree of deformable cubes. It being an octree has just one important effect: it allows the mapper to work at any scale, from large landscape areas to small architectural details. The octree is largely invisible to the mapper though, he can arbitrarily break up larger cubes or merge them, and the engine takes care of the rest.

The deformable cubes are geometric shapes made out of maximally 8 vertices. In its maximum size, it is a regular cube that fills the entire octree node it sits in, as minimum size all 8 vertices coincide which is to say the cube is "empty". Anywhere in between theres a variety of slanted cubes and wedge shapes that are possible to any degree and orientation, and together with neighbouring cubes can form any shape easily.

The editor is similar to the one in Cube but much easier / direct to use, as there arent all the different kinds of cubes to worry about. Here, you can simply "push and pull" geometry with your mousewheel in 6 directions, depending on the orientation of the surface you are looking at. Modifier keys allow you to influence single vertices on a cube, connected vertices touching multiple cubes, edges, faces and entire cubes. Selections can be made to operate on more primitives at once. Many complex shapes are quick and easy to make. It is definitely the most fun way to model architecture to date.

The internal representation of a deformable cube is very different from a vertex based representation, and is based on "edge spans" which allows the engine to represent any kind of shape uniformly in just 12 bytes. This means it can hold great amounts of geometry in memory and maps on disk are small. To render this kind of geometry, the engine goes through a process of converting the internal representation to vertices using plane intersections based on the edge ranges, culls coinciding faces, merges vertices etc. to arrive at something which renders efficiently by hardware and caches this in chunks based on the octree. This allows Sauerbraten a significantly higher polygon throughput than Cube while maintaining its ability to have dynamic geometry.

The current implementation inherents all the non-world geometry related code from Cube, and as such works without modification: entities, AI, gameplay, network, console/script. For the new code, rendering and editing are close to done, physics and lighting are working quite well, level loading/saving is complete, SP & MP work, but plenty of more advanced stuff like occlusion culling & LOD is missing.

UPDATE: Sauerbraten has now started as a Cube community Open Source Engine (& Game) project. It interesting for developers & mappers, but maybe not players yet (though SP & MP do work). If you are interested in checking it out, or even in contributing, I suggest you visit the forums (see below).

NEWS: may 24 2005 release with lightmaps and working MP/SP gameplay! download from sourceforge below.

Mesh Viewer is an easy to use lightweight application to display triangular meshes from a variety of file formats (see 3D formats).
Mesh Viewer uses the OpenGL API to render the models. The program was born under the need for quickly displaying reconstructed triangulated meshes. The Mesh Viewer based on an idea and an early elementary implementation from Craig Robertson.
The current version was developed by Helmut Cantzler. Triangular meshes can be displayed texture mapped (optional with bilinear filtering), solid or as a skeleton (full or just the front lines).
The surface normals of the triangles can be displayed optionally. Features (from a different data file) like edges and points can be displayed into the mesh. Loaded models can be rotated, translated and scaled (all done with the mouse). The model is lighted by multiple light sources.
Viewpoints can be saved. Screenshots of the model can be taken (as BMP, JPEG, PNG and so on). The program is able to calculate the texture map for the 3D model (experimental!).
It can read:
- PMesh files (used at the Vision group of the University of Edinburgh)
- GTS files (from the Gnu Triangulation Library)
- Geomview files (only format "OFF")
- PLY files (only ASCII format)
- VRML 1 files (no texture mapping)
- VRML 2 files
- Feature file
- List file
File format descriptions:
The PMesh format begins optional with a header. The first header line consists of "#pmesh". The header ends with the first line not starting with a "#". The data starts with nv lines of the format "v float float float" for the vertices. Follows nf lines of the format "p 3 int int int" for the triangles. Each line contains the three indices of the vertices. The index of the vertices starts at one.
The GTS format is describe shortly. The first line contains three unsigned integers separated by spaces. The first integer is the number of vertices, nv, the second is the number of edges, ne and the third is the number of faces, nf. Follows nv lines containing the x, y and z coordinates of the vertices. Follows ne lines containing the two indices (starting from one) of the vertices of each edge. Follows nf lines containing the three ordered indices (also starting from one) of the edges of each face.
The Geomview formal starts with a header consisting of the format line "OFF" and a second line with three integer numbers. The first integer is the number of vertices, nv, the second is the number of polygons, np, and the third number is typically "0". Following are the nv lines of vertices (each consists of three floats). The other part of the file are np lines of polygons. The first number of each line stands for the size of the polygon. Mesh Viewer just reads polygons consisting of 2 or 3 vertices. Following are a number of indices of the vertices depending on the size of the polygon. The index of the vertices starts at zero.
VRML is only supported partly. The Mesh Viewer extracts only vertices and triangles from VRML files and ignores all other shapes. Shapes in VRML 2.0 are rotated, scaled and translated if necessary. The file name for JPEG texture and texture coordinates are read from VRML 2.0 files if existent.
The Feature file is used to store vertices and edges. A feature file can be loaded beside the mesh to display vertices or edges into the mesh. The first header line consists of "#list". The header ends with the first line not starting with a "#". The lines consists of either vertices or edges. A vertex line consists of the format "fv float float float". A edge line starts with "fe" followed by the x, y and z coordinates for the start and the end vertex.
The List file is only a text file listing n mesh files, which a loaded subsequently as different shapes into one mesh. Lists are useful for displaying a segmentation of a mesh. The first header line consists of "#list". The header ends with the first line not starting with a "#". The following n lines consists of the names for the mesh files (one per line).

Racer is a free car simulation project, using real car physics to get a realistic feeling. Cars, tracks, scenes and such can be created with relative simplicity in mind (compared to other driving simulations).
The 3D and other file formats are, or should be, documented. Editors and support programs are also available to get a very flexible and expandable simulator. It uses OpenGL for rendering.
It attempts to do well at the physics section, trying to create life-like cars to emphasize car control and doesnt cut back on realism in the interest of fun. If youve played Grand Prix Legends from Papyrus, youll know what Im talking about.
Racer major features:
- Its totally free! (for non-commercial use)
- Available for multiple platforms; Windows 2000/XP (95/98/ME may work but have some trouble with fonts), Linux and Mac OS X.
- 6 DOF models used (the car can move around freely)
- Uses motion formulae from actual engineering documents from SAE for example.
- Total flexibility; almost everything is customizable through ASCII files.
- Commercial-quality rendering engine (with smoke, skidmarks, sparks, sun, flares, vertex-color lit tracks).
- Support for Matrox Surround Gaming. See the corresponding page on Matrox site.
- Lots of addon cars and tracks available on the web.
- Easy integration of your own cars and tracks that you create in ZModeler, 3D Studio Max(tm), Maya etc.
- At least 15 degrees of freedom for a regular car (6 DOF for the car body, 1 for each wheels vertical motion and 1 for each wheel spinning, and 1 for the engine, several more for the driveline). Depending actually on how many wheels you put on the car.
- Real-time internal clock; no physical dependency on framerate. Controller updates are also done independently of the framerate.
- Not limited to 4 wheels; anything from 2 to 8 wheel vehicles are currently supported (but mostly untested, and some problems with hardcoded differentials for example may exist (v0.5.0)).
- Not much constraints on the track data; surface info is taken from polygon data (VRML tracks), and splines are used to smooth out the track surface (polygons are too harsh for driving on just like that).
- Tools to modify the cars & tracks are freely available on this site (though external utilities are recommended for best results).
- Some used algorithms are explained on this site. Also, references are available. If you are programming a carsim, you might find something useful here.
- Save game present!
Main features:
PHYSICS FEATURES
- Full 6 degree of freedom motion of the car chassis.
- Independent suspension for all wheels.
- Suspension features: springs, dampers, anti-rollbar, rollcenter, anti-pitch.
- Wheel features: camber, wheel hop (the wheels have mass), toe.
- Tires: Pacejka tire model, relaxation length.
GRAPHICS FEATURES
- View frustum culling for increased framerate.
- Shader system for Quake-style rendering.
- Fog definable per track.
- Environment mapping for shiny materials.
- Live track environment mapping.

PETSc is a suite of data structures and routines for the scalable (parallel) solution of scientific applications modeled by partial differential equations.
PETSc library employs the MPI standard for all message-passing communication.
PETSc is easy to use for beginners. Moreover, its careful design allows advanced users to have detailed control over the solution process. PETSc includes a large suite of parallel linear and nonlinear equation solvers that are easily used in application codes written in C, C++, Fortran and now Python.
PETSc provides many of the mechanisms needed within parallel application codes, such as simple parallel matrix and vector assembly routines that allow the overlap of communication and computation. In addition, PETSc includes support for parallel distributed arrays useful for finite difference methods.
Main features:
- Parallel vectors
- scatters (handles communicating ghost point information)
- gathers
- Parallel matrices
- several sparse storage formats
- easy, efficient assembly.
- Scalable parallel preconditioners
- Krylov subspace methods
- Parallel Newton-based nonlinear solvers
- Parallel timestepping (ODE) solvers
- Complete documentation
- Automatic profiling of floating point and memory usage
- Consistent interface
- Intensive error checking
- Portable to UNIX and Windows
- Over one hundred examples
- PETSc is supported and will be actively enhanced for many years.
Enhancements:
- This release adds support for 2D and 3D mesh generation, refinement, partitioning, and distribution.
- It adds support for partitioning based upon arbitrary dimensional entities, e.g. vertices, faces, etc.
- It adds support for FIAT element generation.
- It adds support for higher order elements.
- A Poisson problem example in a separate repository has been added.
- The Hypre interface has been updated to use version 2.0.0.
- The Mumps interface has been updated to use version 4.7.3.
- The fftw interface has been updated to use v3.2alpha2.
- There are numerous bugfixes.

The Irrlicht Engine is an open source high performance realtime 3D engine written and usable in C++ and also available for .NET languages.
Irrlicht Engine is completely cross-platform, using D3D, OpenGL and its own software renderer, and has all of the state-of-the-art features which can be found in commercial 3d engines.
Weve got a huge active community, and there are lots of projects in development that use the engine.
You can find enhancements for Irrlicht all over the web, like alternative terrain renderers, portal renderers, exporters, world layers, tutorials, editors, language bindings for java, perl, ruby, basic, python, lua, and so on. And best of all: Its completely free.
Main features:
- High performance realtime 3D rendering using Direct3D and OpenGL [more]
- Platform independent. Runs on Windows95, 98, NT, 2000, XP and Linux.[more]
- Huge built-in and extensible material library with vertex and pixel shader support [more].
- Seamless indoor and outdoor mixing through highly customizeable scene mangagment. [more]
- Character animation system with skeletal and morph target animation. [more]
- Particle effects, billboards, light maps, environment mapping, stencil buffer shadows, and lots of other special effects. [more]
- .NET language binding which makes the engine available to all .NET languages like C#, VisualBasic, and Delphi.NET.
- Platform and driver independent fast software renderer included. It features z-buffer, gouraud shading, alpha-blending and transparency, fast 2D drawing.
- Powerful, customizeable and easy to use 2D GUI System with Buttons, Lists, Edit boxes, ..
- 2D drawing functions like alpha blending, color key based blitting, font drawing and mixing 3D with 2D graphics.
- Clean, easy to understand and well documentated API with lots of examples and tutorials.
- Written in pure C++ and totally object orientated.
- Direct import of common mesh file formats: Maya (.obj), 3DStudio (.3ds), COLLADA (.dae), DeleD (.dmf), Milkshape (.ms3d), Quake 3 levels (.bsp), Quake2 models (.md2), Microsoft DirectX (.X)... [more]
- Direct import of Textures: Windows Bitmap (.bmp), Portable Network Graphics (.png), Adobe Photoshop (.psd), JPEG File Interchange Format (.jpg), Truevision Targa (.tga), ZSoft Painbrush (.pcx)... [more]
- Fast and easy collision detection and response.
- Optimized fast 3D math and container template libraries.
- Directly reading from (compressed) archives. (.zip)
- Integrated fast XML parser.
- Unicode support for easy localisation.
- Works with Microsofts VisualStudio6.0?, VisualStudio.NET 7.0-8.0?, Metrowerks Codewarrior, and Bloodshed Dev-C++ with g++3.2-4.0.
- The engine is open source and totally free. You can debug it, fix bugs and even change things you do not like. And you do not have to publish your changes: The engine is licensed under the zlib licence, not the GPL or the LGPL.

METAGRAPH is a small set of macros that help in drawing (un)directed graphs with METAPOST and the boxes package.
METAGRAPH project provides low-impact definitions that automatically produce (labelled) nodes/vertices and (labelled) edges/arcs that connect correctly boxed items.
Moreover, circular boxes with fixed radius are available to produce more aesthetically pleasant graphs.
Usage:
Just include metagraph.mp in your source files. Full documentation is provided with the package.
Enhancements:
- Loops can now be custom-curled using a global parameter.

KShaderDesigner is an Intergrated Development Environment (IDE) for graphic vertex and fragment shaders. It supports the Cg programming language.
- It interfaces the NVIDIA Cg Compiler
- It allows to block Cg parameters from the scene and set them manually for debugging purpose
- Its architecture allows graphic developpers to add new complex OpenGL scenes to the program, and apply shaders on them
- It supports the latest hardware profiles, like VP40 et FP40 for GeForce 6 GPUs.
The released version is only a preview, it is not really usable. I would like testers to try compilation, project loading, and basic OpenGL rendering. An FX-class Nvidia card is needed for fragment shaders. Project creation and saving is not implemented yet.
An example project is also provided, it implements a shadow mapping algorithm as well as bump mapping.
Please report all compilation errors, feedback or whatever by email.
You will need the Nvidia Cg Toolkit (URL given in configure script).
Enhancements:
- Updated the build system to the latest bksys development version, using the new modules system
- The rendering window now redraws even when no project is running
- Added a per-pixel lighting example. This is an unoptimized implementation of the Blinn-Phong lighting model, using per-pixel calculation. Usefull for benchmarking.