Graph computing is not only related to digital content creation but also closely related to next-generation Internet applications, including meta-computing. Graphics computing requires sufficient computing power as the underlying support. Moore Threads cooperates with Taichi Graphics to accelerate parallel computing of physics and graphics and empower graphics developers and content creators with cutting-edge graphics computing technology in an easy-to-use, convenient, and powerful way. By.
Taichi, an open-source Taichi programming language developed by the founding team of Taichi Graphics, is a parallel programming language designed for high-performance computing. It is directly embedded in Python, which is very easy to learn and write and can greatly improve the productivity of developers. The Taichi general-purpose programming language can successfully run real-time multi-physics simulation effects on Moore Threads GPUs through the Vulkan API SPIR-V instruction system. This effect is called elasticoplasticity dynamics in continuum mechanics. Through the MLS-MPM (Moving Least Squares-Material Point Method) method, the effects of elastic deformation, plastic deformation, hardening, and collision are piled up in simulation calculations.
MLS-MPM relies on data being transferred and processed back and forth between particles and meshes. The algorithm mainly has four steps: 1. Calculate the force on each particle; 2. Propagate the force from the particle to the grid; 3. Grid processing; 4. Collect the velocity from the grid to the particle. The whole process requires the use of Singular Value Decomposition (SVD), particle parallel computing, grid parallel computing, etc., which are all run on Moore Threads GPU through the Vulkan API-driven environment.
The GPU is an important hardware acceleration device in the field of computational physics simulation. It can shorten the physical simulation problem that used to take 12 weeks to complete on the CPU to get the calculation result within a few hours.
The cooperation between Moore Threads and Taichi Graphics is based on the Moore Threads multi-functional GPU chip based on its MUSA unified system architecture. With its capabilities in graphics rendering, AI computing acceleration, and physical simulation computing, it empowers more graphics developers. and content creators, to achieve one-click invocation of the physics engine to automatically generate different physical effects, which will eventually be widely used in 3D digital content creation, improve production efficiency, and allow cutting-edge graphics computing technology to benefit more people.
In the future, Moore Threads and Taichi Graphics will continue to deepen their cooperation to explore computing acceleration strategies using GPUs in various numerical and general parallel computing fields, including real-time physical simulation, numerical computing, AI and machine learning, visual effects, etc. Calculate broad market opportunities.
Taichi Graphics leads the development of the world’s most active open-source parallel computing infrastructure “Taichi” programming language and is committed to the implementation of cutting-edge technologies such as computer graphics, compilers, and cloud computing. The Taichi programming language has more than 18,000 stars on GitHub and is jointly developed by nearly 200 contributors around the world. It has been implemented by well-known Internet companies such as Kuaishou and Tencent and is used by millions of users and nearly 100 universities and research institutes. Based on the Tai Chi programming language, Tai Chi Graphics is building a cloud platform called “Tai Chi Kaiwu” for 3D digital content creators, striving to help every designer, artist, and creative person create 3D. The company has received US$50 million in Series A financing from top venture capital institutions such as Sequoia Capital, Source Code Capital, GGV Capital, and BAI Capital.
Moore Threads is a high-tech integrated circuit company that specializes in GPU chip design. Its primary focus is the development and design of full-featured GPU chips and related products that can provide powerful computing acceleration capabilities to Chinese technology partners. Founded in October 2020, the company is dedicated to developing a new generation of GPUs for meta-computing applications, constructing a comprehensive computing platform that integrates visual computing, 3D graphics computing, scientific computing, and artificial intelligence computing, and establishing an ecosystem based on cloud-native GPU computing to help drive the digital economy’s development.
Moore Threads is distributed by Unixcloud Technology (Shenzhen) Co., Ltd. Unixcloud has a long history of product development and manufacturing. Moore Threads’ official authorized distributor is UnixCloud. It focuses on the field of edge computing in response to the computing power requirements of artificial intelligence development, and it offers the industry edge computing products and solutions that meet a variety of AI requirements. UnixCloud launched the 10G network card business at the same time, offering four-port and two-port 10G optical fiber network cards based on the Mucse-based network controller N10.