Concurrent Number Cruncher (CNC)

The Concurrent Number Cruncher (CNC) is a high-performance preconditioned conjugate gradient solver on the GPU using the GPGPU AMD-ATI CTM and NVIDIA CUDA APIs. The CNC was developed by Luc Buatois using a general optimized implementation of sparse matrices using Block Compressed Row Storage (BCRS) blocking strategies for various block sizes, and optimized BLAS operations through massive parallelization, vectorization of the processing and register blocking strategies.

Uncertainty Visualizer

Uncertainty Visualizer is a stand-alone application dedicated to uncertainty visualization, developped by Thomas Viard. It features two different methods, which respectively map uncertainty to the intensity of a 'fabric' texture pattern or to the blending ratio between a sharp and a blurred display of the model. Input data should be provided as grid slices given in the GSLIB format.
Uncertainty Visualizer reproduces the basic behavior of the UncertaintyViewer Gocad plugin. The Gocad plugin (accessible to sponsors) has much more features for the uncertainty displays on corner-point reservoir grids.

ParticleEngine

ParticleEngine is a visualization engine dedicated to vector fields developed by Thomas Viard and MSc student Gregoire Piquet. It is based on particles randomly sampled over the domain of interest and displaced according to the local orientation and intensity of the vectors. The vector fields are read from a file written in an extended GSLIB format. The package features two different modes of particle displacement, one on the CPU and the other on the GPU.

Magnetostratigraphic correlation (Cupydon)

Magnetostratigraphic correlation is generally a manual task. The Cupydon software allows you to automatically correlate your mag section to the reference scale. The main benefit is not so much the gain in time but the possibility to look at a large number of likely correlations depending on the length of your section and on the variations of the sedimentation rate.

Free 3D structural models

We provide 9 synthetic structural models. Their purpose is to test and benchmark geomodeling algorithms. You may reused and modify these models for research and educational purposes without limitations, provided that you cite our 2015 Computers&Geosciences paper.