NVIDIA’s Jetson TX2 NX delivers the next step in AI performance for entry-level embedded and edge products. It provides up to 2,5 times the performance of Jetson Nano and shares form factor and pin compatibility with Jetson Nano and Jetson Xavier NX.
This small, lower-cost version of the Jetson TX2 is perfect for creating mass-market AI products. Intelligent machine OEMs can scale their product offerings with pin-compatible Jetson modules, while using cloud-native technologies to build, deploy and manage the same software across all of them.
The Jetson TX2 NX system-on-module (SOM) packs hardware accelerators for the entire AI pipeline, including ISP, encoder, decoder, image compositor and more. The NVIDIA JetPack SDK includes libraries and samples for using these accelerators, as well as the latest NVIDIA tools for faster application development and optimisation. All major AI frameworks are supported.
Pre-trained AI models from NVIDIA NGC and the NVIDIA Transfer Learning Toolkit also provide an ideal way to accelerate time-to-market. You can now develop custom AI networks with less effort and containerised deployments make updates flexible and seamless.
Ease of development and speed of deployment – plus a unique combination of form factor, performance and power advantage – make Jetson TX2 NX the ideal mass-market AI product platform to get to market and continuously update over the lifetime of a product.
First NVMe SSD Built with 8th-gen BiCS FLASH EBV Electrolink
Computer/Embedded Technology
KIOXIA recently announced the development and prototype demonstration of its new KIOXIA CM9 Series PCIe 5.0 NVMe SSDs, which incorporates CMOS directly Bonded to Array technology.
Read more...Wi-Fi 7 front-end module RF Design
Telecoms, Datacoms, Wireless, IoT
The Qorvo QPF4609 is an integrated front end module designed for 802.11be systems that has integrated matching, which minimises layout area.
Read more...QuecPi smart development board iCorp Technologies
Computer/Embedded Technology
The QuecPi Alpha smart MOB development board is Quectel’s smart development board based on Qualcomm’s QCS6490 high-performance 64-bit octa-core processor.
Read more...Eskom announces PV registration extension
Computer/Embedded Technology
The South African Photovoltaic Industry Association has welcomed Eskom’s decision to extend its zero-registration fee policy and free smart meter installation for residential Small-Scale Embedded Generation customers until March 2026.
Read more...GNSS chipset for wearables RF Design
Telecoms, Datacoms, Wireless, IoT
The UBX-M10150-CC from u-blox is a GNSS chip that supports GPS, QZSS/SBAS, Galileo, and BeiDou constellations, and is designed for integration into wearable applications.
Read more...Unlock enhanced wireless performance
Computer/Embedded Technology
Duxbury Networking has introduced Cambium Networks’ Deep Virtual Circuit (Deep VC) technology, a free upgrade for the PMP 450 platform that will transform wireless broadband performance across the country.
Read more...X-band radar RF Design
Editor's Choice Telecoms, Datacoms, Wireless, IoT
X-band radar systems, particularly those leveraging beamforming ICs (BFICs), advanced gallium nitride (GaN) and gallium arsenide (GaAs) components, are leading the way in providing the high-performance radar capabilities required for modern defence and surveillance.
Read more...GNSS antenna redefining what’s possible RF Design
Telecoms, Datacoms, Wireless, IoT
u-blox has achieved what was once thought impossible with the launch of the DAN-F10N, the industry’s smallest and most reliable L1, L5 dual-band GNSS antenna module.
Read more...u-blox expands NORA-B2 BLE modules RF Design
Telecoms, Datacoms, Wireless, IoT
The new nRF54L chipset-based wireless modules reduce current consumption and double processing capacity, catering to diverse mass market segments.
Read more...New GNSS passive patch antenna RF Design
Telecoms, Datacoms, Wireless, IoT
The HP24510A from Taoglas is a stacked patch GNSS passive antenna that operates from 1215 to 1610 MHz covering the L1/L2 GNSS spectrum.
printer friendly version