COLLADA Refinery

Product information
Name: COLLADA Refinery Purpose: unstated Last updated: March 28, 2007 Current version: unknown Status: unstated OS: unstated Report bugs Maintainer: [??? SCEI US R&D] Contact for technical issues:
Download
This article is part of the COLLADA products directory
Adding to the list of products

COLLADA Refinery is a GUI-based application for creating content pipelines for 3D assets using modular components called conditioners.

((EDITOR: This page needs the following improvement: need URL for SCEI US R&D (in info box) ))

COLLADA Conditioner is a module component in COLLADA Refinery. Each conditioner can take a COLLADA document, perform a specific operation on the document, and pass the resulting document to the next conditioner. A series of conditioners is a pipeline that improves the state of COLLADA documents.

At the final production stage, developers generally transform COLLADA content into their own format and then apply their own conditioning pipeline or asset pipeline" to optimize the data, test the data for problems, merge data with other content, and "compile" the data in platform-specific format for use with a specific game engine or real-time application. The conditioning pipeline is generally a set of command-line utilities that are used with a build (makefile) mechanism to preprocess the content. The entire transformation process is equivalent to the source-code-to-binary-code transformation. The conditioning pipeline component is analogous to an optimizing compiler/linker.

Some, if not all, of the conditioning pipeline can be done in the COLLADA format. Refinery is an example of a COLLADA-based conditioning pipeline application. A user interface provides a way to create and test the pipeline, which is a set of interconnected conditioners. For example, going from polygons to optimized triangle mesh can be done with a triangulator conditioner followed by a triangle mesher. Or optimizing content for Google Earth can be done with the KMZ Clean-up conditioner.

The COLLADA Refinery is open source, so additional individual conditioners can be developed to provide COLLADA users with a powerful, ready-to-use pipeline for their end application. To use the pipeline, a user would assemble a pipeline with the individual conditioners they need and then, as the final step, format the file to their application's binary-specific format, using their own plug-in for Refinery. New release for COLLADA Refinery 2.0.3

New release for COLLADA Refinery 2.0.3

March 29, 2007

• Axisconverter

  add matrix support, remove exit(1) that cause unexpected exit.

• coherencytest

  CHECK_unique_id will run faster, update check_skin that crash on null pointers.

• copyrighter

  remove exit(1) that cause unexpected exit.

• deindexer

  add resolveElement() to make sure getElement() will work, add null check to avoid crash.

• kmzcleanup

  add null checks to avoid crash, remove exit(1) that cause unexpected exit.

• packager

  remove exit(1) that cause unexpected exit.

• Pipeline.java

  add line separators to separate output message between execute()

See also

• COLLADA public forum: Search for "refinery" to find discussions

External links

• SourceForge download page
• Contest: Create new conditioners (through July 2007)

Screenshot

This shows a COLLADA Refinery tool window displaying a pipeline of two conditioners in series. These conditioners first triangulate all the polygons to triangles and then deindex the triangles.