{"id":7089,"date":"2020-06-30T09:35:07","date_gmt":"2020-06-30T07:35:07","guid":{"rendered":"https:\/\/blog.generationrobots.com\/?p=7089"},"modified":"2023-05-05T09:59:02","modified_gmt":"2023-05-05T07:59:02","slug":"comparison-of-nvidia-jetson-nanocomputers","status":"publish","type":"post","link":"https:\/\/www.generationrobots.com\/blog\/en\/comparison-of-nvidia-jetson-nanocomputers\/","title":{"rendered":"Comparison of NVIDIA Jetson nanocomputers"},"content":{"rendered":"\n\n \n

\n NVIDIA\u00ae Jetson\u2122 Systems: Deploy powerful AI in your robotic projects\n <\/h2>\n

\n Compact, able to run powerful real-time calculations, while being energy efficient, NVIDIA Jetson nanocomputers adapt perfectly to the constraints of embedded systems.\n <\/p>\n

\n \n \n NVIDIA\u00ae Jetson\u2122 nanocomputers\n <\/a>\n <\/strong>\n are used to deploy deep learning applications, manage multiple neural networks in parallel, or process high-resolution data from multiple sensors in real-time. These boards combine ultra-high performance with unprecedented energy-efficiency, making them a smart choice for on-board robotics projects (autonomous vehicles, connected city, home automation, health, etc.).\n <\/p>\n

\n The NVIDIA\u00ae Jetson\u2122 boards are a benchmark for autonomous projects, and have been adopted by leading companies in industrial robotics such as FANUC or Komatsu.\n <\/p>\n

\n \n Applications: virtual reality, automotive simulation, architecture, robotics, smart cities, e-health…\n <\/strong>\n <\/p>\n

\n Key differences between NVIDA Jetson supercomputers\n <\/h2>\n

\n NVIDIA Jetson Nano\n <\/h3>\n

\n The\n \n \n NVIDIA Jetson Nano\n <\/a>\n <\/strong>\n development kit puts in your hands an affordable and rich in possibilities AI development solution.\n <\/p>\n

\n The NVIDIA Jetson NANO is the entry-level NVIDIA Jetson board. It offers great performance for prototyping projects in the fields of small autonomous robotics or home IoT. It is also a great educational tool for higher education.\n <\/p>\n

\n NVIDIA Jetson TX2\n <\/h3>\n

\n Get computing performance with the\n \n NVIDIA Jetson TX2\n <\/strong>\n , exceptional accuracy and energy-efficiency, combined with an ultra-light module.\n <\/p>\n

\n \n Deploy Deep Learning applications in compact devices such as drones.\n <\/strong>\n <\/p>\n

\n NVIDIA Jetson Xavier NX\n <\/h3>\n

\n \n \n NVIDIA Jetson Xavier NX\n <\/a>\n <\/strong>\n is the world’s smallest AI supercomputer for Embedded and Edge systems. \nThis NVIDIA Jetson nanocomputer also offers native cloud support. This native cloud support allows you to develop computing power-hungry features and applications on embedded devices and other types of devices.\n <\/p>\n

\n It also allows algorithms to be deployed quickly and on a large scale.\n <\/p>\n

\n \n Targeted application areas include robotics, smart cities, healthcare, industrial IoT, etc.\n <\/strong>\n <\/p>\n

\n Jetson Xavier NX offers performances 10 times higher than its predecessor, the NVIDIA Jetson TX2 board.\n <\/p>\n

\n NVIDIA Jetson Xavier AGX\n <\/h3>\n

\n The\n \n \n NVIDIA Jetson Xavier AGX\n <\/a>\n <\/strong>\n is the world’s first computer specifically designed for autonomous machines. It provides high performance that can support visual odometry, sensor fusion, location, mapping, object detection and trajectory planning algorithms.\n <\/p>\n

\n \n Applications: delivery and logistics robots, factory automation systems, industrial UAVs.\n <\/strong>\n <\/p>\n

\n NVIDIA Jetson AGX Orin\n <\/h3>\n

\n The\n \n \n Jetson AGX Orin\n <\/a>\n <\/strong>\n development kit is 8 times more powerful than its predecessor, the Xavier AGX. With its ultra-compact dimensions, it offers an unprecedented computing speed (up to 275 TOPS), favouring numerous possibilities in edge computing and AI development. Moreover, with its eco-design (between 15 and 60 W), you will stretch the limits of your embedded and autonomous robotics projects.\n <\/p>\n

\n \n Applications: industrial production, logistics, retail, human services, agriculture, connected cities, health and life sciences.\n <\/strong>\n <\/p>\n

\n Comparison table between the different NVIDIA Jetson boards\n <\/h2>\n \n \n \n \n \n \n \n \n \n \n
\n <\/td>\n \n
\n \"NVIDIA\n <\/div>\n <\/td>\n
\n
\n \"NVIDIA\n <\/div>\n <\/td>\n
\n
\n \"Nvidia\n <\/div>\n <\/td>\n
\n
\n \"\n <\/div>\n <\/td>\n
\n
\n \"Nvidia\n <\/div>\n <\/td>\n <\/tr>\n
\n <\/td>\n \n \n \n JETSON Nano\n <\/a>\n <\/strong>\n <\/td>\n \n \n \n JETSON TX2\n <\/a>\n <\/strong>\n <\/td>\n \n \n \n JETSON Xavier NX\n <\/a>\n <\/strong>\n <\/td>\n \n \n \n JETSON Xavier AGX\n <\/a>\n <\/strong>\n <\/td>\n \n \n \n JETSON AGX Orin\n <\/a>\n <\/strong>\n <\/td>\n <\/tr>\n
\n \n AI Performance\n <\/strong>\n <\/td>\n \n 0.5 TFLOPS (FP16)\n <\/td>\n \n 1.3 TFLOPS (FP16)\n <\/td>\n \n 6 TFLOPS (FP16)\n \n 21 TOPS (INT8)\n <\/td>\n \n 5.5-11 TFLOPS (FP16)\n \n 20-32 TOPS (INT8)\n <\/td>\n \n 275 TOPs\n <\/td>\n <\/tr>\n
\n \n GPU\n <\/strong>\n <\/td>\n \n 128-core NVIDIA Maxwell\u2122 GPU\n <\/td>\n \n 256-core NVIDIA Pascal\u2122 GPU architecture with 256 NVIDIA CUDA cores\n <\/td>\n \n NVIDIA Volta architecture with 384 NVIDIA CUDA\u00ae cores and 48 Tensor cores\n <\/td>\n \n 512-Core Volta GPU with Tensor Cores\n <\/td>\n \n NVIDIA Ampere architecture with 1792 NVIDIA\u00ae CUDA\u00ae cores and 56 Tensor Cores\n <\/td>\n <\/tr>\n
\n \n CPU\n <\/strong>\n <\/td>\n \n Quad-core ARM A57 @ 1.43 GHz\n <\/td>\n \n Dual-Core NVIDIA Denver 2 64-Bit CPU\n \n Quad-Core ARM\u00ae Cortex\u00ae-A57 MPCore\n <\/td>\n \n 6-core NVIDIA Carmel ARM\u00aev8.2 64-bit CPU 6 MB L2 + 4 MB L3\n <\/td>\n \n 8-Core ARM v8.2 64-Bit CPU, 8 MB L2 + 4 MB L3\n <\/td>\n \n 8-core ARM Cortex-A78AE v8.2 2MB L2 + 4 MB L3\n <\/td>\n <\/tr>\n
\n \n Memory\n <\/strong>\n <\/td>\n \n 4 GB 64-bit LPDDR4 25.6 GB\/s\n <\/td>\n \n 8 GB 128-bit LPDDR4\n \n 1866 MHz – 59.7 GB\/s\n <\/td>\n \n 8 GB 128-bit LPDDR4x @ 51.2GB\/s\n <\/td>\n \n 32 GB 256-Bit LPDDR4x | 137 GB\/s\n <\/td>\n \n 32 GB 256-Bit LPDDR5 | 204,8 GB\/s\n <\/td>\n <\/tr>\n
\n \n Power Consumption\n <\/strong>\n <\/td>\n \n 5-10W\n <\/td>\n \n 7.5-15W\n <\/td>\n \n 10-15W\n <\/td>\n \n 10-30W\n <\/td>\n \n 15W-20W-50W\n <\/td>\n <\/tr>\n
\n \n Price\n <\/strong>\n <\/td>\n \n 129\u20ac (INCL. VAT)\n <\/td>\n \n 515\u20ac (INCL. VAT)\n <\/td>\n \n 515\u20ac (INCL. VAT)\n <\/td>\n \n 925\u20ac (INCL. VAT)\n <\/td>\n \n 3420\u20ac (INCL. VAT)\n <\/td>\n <\/tr>\n <\/tbody>\n <\/table>\n \n <\/body>\n<\/html>","protected":false},"excerpt":{"rendered":"

NVIDIA\u00ae Jetson\u2122 Systems: Deploy powerful AI in your robotic projects Compact, able to run powerful real-time calculations, while being energy efficient, NVIDIA Jetson nanocomputers adapt perfectly to the constraints of embedded systems. NVIDIA\u00ae Jetson\u2122 nanocomputers are used to deploy deep learning applications, manage multiple neural networks in parallel, or process high-resolution data from multiple sensors[…]
Read this article >><\/a><\/p>\n","protected":false},"author":188,"featured_media":7084,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[10545],"tags":[],"class_list":["post-7089","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-comparisons-and-tests-products"],"_links":{"self":[{"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/posts\/7089","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/users\/188"}],"replies":[{"embeddable":true,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/comments?post=7089"}],"version-history":[{"count":10,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/posts\/7089\/revisions"}],"predecessor-version":[{"id":14218,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/posts\/7089\/revisions\/14218"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/media\/7084"}],"wp:attachment":[{"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/media?parent=7089"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/categories?post=7089"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/tags?post=7089"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}