{"id":7054,"date":"2020-06-30T09:35:07","date_gmt":"2020-06-30T07:35:07","guid":{"rendered":"https:\/\/blog.generationrobots.com\/?p=7054"},"modified":"2023-05-05T09:58:21","modified_gmt":"2023-05-05T07:58:21","slug":"comparatif-des-nano-ordinateurs-nvidia-jetson","status":"publish","type":"post","link":"https:\/\/www.generationrobots.com\/blog\/fr\/comparatif-des-nano-ordinateurs-nvidia-jetson\/","title":{"rendered":"Comparatif des nano-ordinateurs NVIDIA Jetson"},"content":{"rendered":"\n\n \n

\n Syst\u00e8mes NVIDIA\u00ae Jetson\u2122 : d\u00e9ployez de puissantes IA dans vos projets robotiques\n <\/h2>\n \n \n

\n Tr\u00e8s compacts, avec une tr\u00e8s grosse puissance de calcul en temps r\u00e9el, tout en \u00e9tant \u00e9conomes en \u00e9nergie, les\n \n \n nano-ordinateurs NVIDIA Jetson\n <\/a>\n <\/strong>\n s\u2019adaptent parfaitement aux contraintes de syst\u00e8mes embarqu\u00e9s.\n <\/p>\n \n \n

\n Les processeurs NVIDIA\u00ae Jetson\u2122 sont utilis\u00e9s pour d\u00e9ployer des applications de\n \n Deep Learning\n <\/strong>\n , g\u00e9rer des\n \n r\u00e9seaux de neurones multiples en parall\u00e8le\n <\/strong>\n ou traiter en\n \n temps r\u00e9el\n <\/strong>\n des donn\u00e9es issues de capteurs multiples en\n \n haute r\u00e9solution\n <\/strong>\n .\n <\/p>\n \n \n

\n Ces modules allient de ultra hautes performances et un\n \n rendement \u00e9nerg\u00e9tique\n <\/strong>\n sans pr\u00e9c\u00e9dent, ce qui en fait des composants de choix pour les projets de robotique embarqu\u00e9e (v\u00e9hicules autonomes, ville connect\u00e9e, domotique, sant\u00e9, etc).\n <\/p>\n \n \n

\n Les modules NVIDIA\u00ae Jetson\u2122 sont une r\u00e9f\u00e9rence pour le\n \n tout-autonome\n <\/strong>\n , et ont d\u2019ailleurs \u00e9t\u00e9 adopt\u00e9s par des entreprises leader en robotique industrielle comme FANUC ou Komatsu.\n <\/p>\n \n \n

\n \n Applications : jeux vid\u00e9o, simulation automobile, architecture, robotique, villes intelligentes, e-sant\u00e9\u2026\n <\/strong>\n <\/p>\n \n \n

\n Principales diff\u00e9rences entre les supercalculateurs NVIDA Jetson\n <\/h2>\n \n \n

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

\n Le kit de d\u00e9veloppement\n \n \n NVIDIA Jetson Nano\n <\/strong>\n <\/a>\n place entre vos mains une solution de d\u00e9veloppement IA abordable et riche de possibilit\u00e9s.\n <\/p>\n \n \n

\n Entr\u00e9e de gamme de la gamme NVIDIA Jetson, la carte NVIDIA Jetson NANO offre cependant des performances plus que louables qui conviendront \u00e0 des projets de\n \n prototypage dans les domaines de la petite robotique autonome ou l\u2019IoT domestique.\n <\/strong>\n <\/p>\n \n \n

\n Nous la recommandons \u00e9galement pour une utilisation dans l\u2019\n \n enseignement sup\u00e9rieur\n <\/strong>\n .\n <\/p>\n \n \n

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

\n B\u00e9n\u00e9ficiez de performances de calcul avec la\n \n NVIDIA Jetson TX2\n <\/strong>\n , d\u2019une pr\u00e9cision et d\u2019un rendement \u00e9nerg\u00e9tique exceptionnels avec un module ultra l\u00e9ger. D\u00e9ployez des applications de Deep Learning dans des produits compacts tels que des\n \n drones\n <\/strong>\n .\n <\/p>\n \n \n

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

\n \n \n NVIDIA Jetson Xavier NX\n <\/strong>\n <\/a>\n se positionne comme le plus petit supercalculateur d’IA au monde pour les syst\u00e8mes Embedded et Edge. Ce nano-ordinateur NVIDIA Jetson offre \u00e9galement une prise en charge native du cloud.\n <\/p>\n \n \n

\n Ce support natif du cloud permet de d\u00e9velopper des fonctionnalit\u00e9s et des applications gourmandes en puissance de calcul sur des appareils embarqu\u00e9s et autres types de p\u00e9riph\u00e9riques. Cela permet aussi de lancer des algorithmes rapidement et \u00e0 une grande \u00e9chelle.\n <\/p>\n \n \n

\n \n Les domaines d\u2019applications cibl\u00e9s sont la robotique, les villes intelligentes, les soins de sant\u00e9, l\u2019IoT industriel, etc.\n <\/strong>\n <\/p>\n \n \n

\n Jetson Xavier NX acc\u00e9l\u00e8re la pile logicielle de NVIDIA avec des performances jusqu\u2019\u00e0 10 fois sup\u00e9rieures que son pr\u00e9d\u00e9cesseur, la carte NVIDIA Jetson TX2.\n <\/p>\n \n \n

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

\n Premier ordinateur au monde sp\u00e9cifiquement con\u00e7u pour les machines autonomes, la\n \n \n Jetson Xavier AGX\n <\/strong>\n <\/a>\n fournit des performances \u00e9lev\u00e9es \u00e0 m\u00eame de prendre en charge les algorithmes d\u2019odom\u00e9trie visuelle, de fusion de capteurs, de localisation, de cartographie, de d\u00e9tection d\u2019objets et de planification de trajectoire.\n <\/p>\n \n \n

\n \n Applications : robots de livraison et de logistique, syst\u00e8mes d\u2019automatisation d\u2019usine, v\u00e9hicules UAV industriels.\n <\/strong>\n <\/p>\n \n \n

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

\n Le kit de d\u00e9veloppement\n \n \n Jetson AGX Orin\n <\/strong>\n <\/a>\n est 8 fois plus performant que son pr\u00e9d\u00e9cesseur, le Xavier AGX. Avec ses dimensions ultra compactes, il offre une vitesse de calcul in\u00e9dite (jusqu\u2019\u00e0 275 TOPS), favorisant de nombreuses possibilit\u00e9s en termes d\u2019edge computing et de d\u00e9veloppement d\u2019IA. De plus, gr\u00e2ce \u00e0 son \u00e9co-conception (entre 15 et 60 W), il vous permettra de repousser les limites de vos projets embarqu\u00e9s et de robotique autonome.\n <\/p>\n \n \n

\n \n Applications : production industrielle, logistique, vente au d\u00e9tail, services \u00e0 la personne, agriculture, villes connect\u00e9es, sant\u00e9 et sciences de la vie.\n <\/strong>\n <\/p>\n \n \n

\n Tableau comparatif entre les diff\u00e9rentes cartes NVIDIA Jetson\n <\/h2>\n \n \n \n \n \n \n \n \n \n \n \n \n
\n <\/td>\n \n
\n \"Microcontroleur\n <\/div>\n <\/td>\n
\n
\n \"Microcontroleur\n <\/div>\n <\/td>\n
\n
\n \"Microcontroleur\n <\/div>\n <\/td>\n
\n
\n \"Microcontroleur\n <\/div>\n <\/td>\n
\n
\n \"Microcontroleur\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 Performance IA\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 M\u00e9moire\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 Consommation d’\u00e9nergie\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 Prix\n <\/strong>\n <\/td>\n \n 129\u20ac TTC\n <\/td>\n \n 515\u20ac TTC\n <\/td>\n \n 515\u20ac TTC\n <\/td>\n \n 925\u20ac TTC\n <\/td>\n \n 3420\u20ac TTC\n <\/td>\n <\/tr>\n <\/tbody>\n <\/table>\n \n <\/body>\n<\/html>","protected":false},"excerpt":{"rendered":"

Syst\u00e8mes NVIDIA\u00ae Jetson\u2122 : d\u00e9ployez de puissantes IA dans vos projets robotiques Tr\u00e8s compacts, avec une tr\u00e8s grosse puissance de calcul en temps r\u00e9el, tout en \u00e9tant \u00e9conomes en \u00e9nergie, les nano-ordinateurs NVIDIA Jetson s\u2019adaptent parfaitement aux contraintes de syst\u00e8mes embarqu\u00e9s. Les processeurs NVIDIA\u00ae Jetson\u2122 sont utilis\u00e9s pour d\u00e9ployer des applications de Deep Learning ,[…]
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":[10523],"tags":[],"class_list":["post-7054","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-comparatifs-et-tests-produits"],"_links":{"self":[{"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/posts\/7054","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=7054"}],"version-history":[{"count":36,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/posts\/7054\/revisions"}],"predecessor-version":[{"id":14217,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/posts\/7054\/revisions\/14217"}],"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=7054"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/categories?post=7054"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.generationrobots.com\/blog\/wp-json\/wp\/v2\/tags?post=7054"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}