GPU modules target high-performance machine vision

February 01, 2021 //By Jean-Pierre Joosting
Cincoze GPU modules target high-performance machine vision
Modular machine vision system delivers 1920 CUDA cores, 5.3 TFLOPS peak FP32 computing, with AI computing Tensor, and specialized RT core for ray tracing.

Cincoze has expanded its GM-1000 machine vision embedded system with two new Quadro MXM GPU modules. Building on the GM-1000’s powerful processing base, the MXM-RTX3000 and MXM-T1000 provide the additional GPU capacity for rapid adoption of machine vision in smart factories, from simple environmental perception applications such as positioning, measurement, identification, and sorting, to more complex vision-guided automation functions. GPU requirements for each scenario are different, so specifications must match the environment and application. The two new Quadro® MXM GPU modules broaden the GM-1000's available selection to cover a wider range of uses.

The GM-1000, part of the Cincoze GOLD series, is positioned as a high-performance machine vision system featuring high computing performance, high-speed I/O, and industrial-grade reliability. It is the preferred choice for machine vision system integrators and AOI (Automated Optical Inspection) manufacturers.

A unique carrier board can be matched with a selection of Cincoze MXM GPU modules, including the MXM-RTX3000, MXM-T1000, MXM-P2000, and MXM-E9174, providing a precise match for different computing requirements. Upgrade flexibility has always been a proud advantage of Cincoze. The GM-1000 only needs a carrier board and MXM GPU module replacement to facilitate future performance upgrades without replacing the computer host.

The new Quadro® MXM GPU modules, the MXM-RTX3000 and MXM-T1000, use the latest NVIDIA® Quadro® Turing™ GPU architecture based on the latest 12nm process.

In terms of core computing power and performance, the MXM-RTX3000 has 1920 CUDA cores, 5.3 TFLOPS peak FP32 high-end computing power, parallel integer execution, AI computing Tensor core, and specialized RT core for ray tracing. Together these provide fully integrated GPU features to face complex visual image calculations. The MXM-T1000 has 896 CUDA cores, 2.6 TFLOPS peak FP32 computing power, and only 50 W power consumption for power-conscious high-speed computing.


Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.