usage of GPU

I would like to use scilab6.1.1 to do fast numerical computation with GPU, is there any tool for that?

The sciGPPU was developed for Scilab 5.4.x and it was developed in 2013, will it work properly with scilab6.1.1?
If anyone knows, please let me know.
Best regards.

Scilab 01-01-22, 5:09 p.m. M.FUJIMOTO
Instead of using sciGPGPU you can try 'sciCuda (Scilab Cuda toolbox)' which is developed for Scilab 6.0.x version. Check this link for more information: https://atoms.scilab.org/toolboxes/sciCuda/1.0
03-01-22, 11:10 a.m. Rashmi

Hi, Rashmi Thank you for your helpful information. IF I have some question on sciCuda, can I ask question at the suggested page? M. Fujimoto
04-01-22, 12:35 p.m. M.FUJIMOTO

Yes sure, will try my best to answer your queries.

13-01-22, 10:48 a.m. Rashmi

Login to add comment


GPUs are used in embedded systems, mobile phones, personal computers, workstations, and game consoles. Modern GPUs are very efficient at manipulating computer graphics and image processing.


11-01-22, 1:07 p.m. Maurice

Scilab 6.1.1 does not have GPU support, it will be available in Scilab 6.2 (in a couple of months). Scilab 6.0.2 has GPU support, but it only works with CUDA 7.5, which is a bit old (the current one is CUDA 9.1). Also, the GPU module is experimental and not well tested, so I am afraid that you will experience some problems with it (or even cannot install it at all - your GPU needs to be compatible with CUDA 7.5). Scilab has a Cuda wrapper that allows running Cuda code directly in Scilab. I have used it for my master's thesis along with the best uk essay writing servicesand it worked quite well. Scilab 6.0.2 does not have this wrapper, but you can download the sources from Github and compile it yourself.
28-03-22, 3:10 p.m. Elisa56

It helped me on a current project, thank you! MyMilestoneCard
20-06-22, 5:46 p.m. oscarmoc

Fantastic blog. Your articles were interesting to read. This is an excellent read for me. <slope unblocked>
15-07-22, 1:20 p.m. AngelAngle112

Utilize a variety of high-level numerical analysis tools while diggy taking advantage of NVIDIA GPUs' massive numerical throughput in single and double precision.
17-11-22, 5:21 a.m. candysweet

You can use external tools to compute numbers on GPU and combine them contexto with Scilab. Examples: CUDA - NVIDIA's parallel computing platform for GPU-powered applications.
13-03-23, 8:06 p.m. emmausa

sciGPPU was developed for Scilab 5.4.x and developed in 2013. Up to now it has been updated so many times, we need to fleeing the complex and approach the latest one.
07-04-23, 2:16 p.m. Swenyly

Investigate more GPU computing possibilities for numerical calculations basketbros, such as using Python with TensorFlow, PyTorch, or CUDA-compatible libraries. These libraries are extensively used for quick numerical computations on GPUs and have strong GPU support.
15-05-23, 2:09 p.m. timothysykes

Log-in to answer to this question.