Maple can take advantage of CUDA-enabled graphics cards to leverage the tremendous computational power of those cards, dramatically speeding up key computations.
- The CUDA package allows Maple to use the graphics processing unit (GPU) of your NVIDIA Compute Unified Device Architecture (CUDA)-enabled hardware to accelerate key linear algebra routines, when such a card is available on your computer.
- Numeric computations done using CUDA acceleration can run an order of magnitude faster.
- The key computations that take advantage of the CUDA technology are fundamental, often repeated steps in virtually all matrix computations, so substantial speedups are available for all your large-scale linear algebra problems.
- Both single (float[4]) and double-precision (float[8]) operations are supported, depending on the capabilities of the GPU hardware in use.