ナセル Product Lineup

The Condor AGX-IOX is a robust 3U VPX high-performance single board computer (SBC) designed for autonomous and embedded edge computing systems. It performs instantaneous data analysis and transfer, multi-sensor processing, encoding/decoding, and AI inference. The Condor AGX-IOX is equipped with the NVIDIA Jetson AGX Orin SoM (Industrial) and supports the NVIDIA Ampere GPU and Arm Cortex-A78AE CPU. This product supports 64GB LPDDR5 system memory and achieves a maximum memory bandwidth of 205 GB/s shared between the ARM Cortex CPU, NVIDIA Ampere GPU, and accelerator engines.

The Condor AGX-ORIN64-HPC is a robust 3U VPX high-performance single board computer (SBC) designed for autonomous and embedded edge computing systems. It features the NVIDIA Jetson AGX Orin 64GB (Industrial) System on Module (SoM), which is equipped with an NVIDIA Ampere architecture GPU and a 12-core Arm Cortex-A78AE CPU. The module's 64GB LPDDR5 system memory achieves a maximum memory bandwidth of 205 GB/s, shared among the internal CPU, GPU, and accelerator engines. It is equipped with HEVC (H.265)/AVC (H.264) compliant NVENC and NVDEC engines, supporting up to 4K-UHD encoding resolution. The AGX-ORIN64-HPC is designed to support both the VITA 46/65 and Sensor Open System Architecture (SOSA) technical standards for slot profile 14.6.11.

This is a robust 3U OpenVPX video capture and GPGPU processing card designed for mission-critical C5ISR and high-performance embedded computing (HPEC) applications. The Condor GR6S-AD2000, equipped with the NVIDIA RTX 2000 Ada GPU, is the ideal solution for ultra-low latency video capture and delivery in sensor fusion, object detection, and automatic target recognition (ATR) applications. Based on the NVIDIA Ada Lovelace architecture, it features 3,072 CUDA cores, 24 Tensor cores, and 96 RT cores, delivering exceptional performance in parallel computing, AI workloads, and advanced visualization. The 8GB GDDR6 memory with error correction code (ECC) ensures reliable data integrity and stability even in harsh environments. It supports encoding/decoding of H.265 (HEVC), H.264 (MPEG4/AVC), and AV1. Additionally, it is compatible with NVIDIA GPUDirect RDMA, achieving reduced latency and improved data transfer speeds in real-time processing.

The Condor GR5SL-B5000 is a robust 3U OpenVPX video graphics and GPGPU card equipped with the advanced NVIDIA RTX PRO 5000 Blackwell GPU. It supports ECC-enabled 24 GB GDDR7 graphics memory and PCI Express Gen 5 for the first time, significantly improving data transfer speeds and performance per watt. The GPU features 10,496 CUDA cores, 320 Tensor cores, and 80 RT cores, delivering exceptional performance for AI-accelerated data processing and real-time situational analysis. It also includes a 9th generation NVENC encoder and a 6th generation NVDEC decoder that support 4:2:2, providing advanced video compression and playback capabilities. Additionally, it supports NVIDIA GPUDirect RDMA, greatly reducing data transfer latency. Available in both conduction cooling (VITA 48.2) and airflow-through (VITA 48.8) variations, it minimizes performance degradation at high temperatures.

The Condor GR5SL-B4000 is a robust 3U OpenVPX video graphics and GPGPU card equipped with the advanced NVIDIA RTX PRO 4000 Blackwell GPU. It features ECC-supported 16 GB GDDR7 memory and PCI Express Gen 5, providing groundbreaking performance and efficiency in high-performance computing (HPEC) environments such as edge AI inference. The GPU includes 7,680 CUDA cores, 240 Tensor cores, and 60 RT cores, enabling AI-accelerated data processing. Additionally, it has a built-in 9th generation NVENC/6th generation NVDEC encoder, excelling in video performance. It supports NVIDIA GPUDirect RDMA, reducing data transfer latency. The thermal design accommodates conduction cooling (VITA 48.2) and airflow-through (VITA 48.8), improving thermal efficiency.

The Condor GR5SL-B5000 is a robust 3U OpenVPX video graphics/GPGPU card equipped with the advanced NVIDIA RTX PRO 5000 Blackwell GPU. It features 10,496 CUDA cores, 320 Tensor cores, and 80 RT cores, along with ECC-supported 24 GB GDDR7 memory (256-bit), providing high performance and efficiency for AI inference and edge computing. Connectivity: It integrates a Microchip PFX PCIe Gen 4 switch for high-speed connections and supports GPUDirect RDMA to reduce data transfer latency. Cooling: It offers both conduction cooling (VITA 48.2) and airflow-through (VITA 48.8) variations, with a new design that significantly improves thermal efficiency. This card also includes the 9th generation NVENC (supporting 4:2:2) and 6th generation NVDEC decoder, making it ideal for mission-critical environments.

The Condor NVB500xX is a robust XMC video graphics and GPGPU card. GPU: Equipped with the advanced NVIDIA RTX PRO 500 Blackwell GPU. Memory: Supports 6GB of ECC-enabled GDDR7 graphics memory and a 128-bit memory interface. Features: NVIDIA Blackwell, the first architecture to support GDDR7, achieves significant improvements in bandwidth, enhanced signal integrity, and reductions in energy consumption and power usage. Applications: These features make it ideal for demanding high-performance computing environments. Output: Supports three DisplayPort++/DVI video outputs. This card is a high-performance embedded system solution that leverages next-generation memory technology and is equipped with the NVIDIA RTX PRO 500 Blackwell GPU. It has 1,792 CUDA cores, excelling in AI processing and real-time analytics. It enables high-speed data transfer with PCIe Gen 5 and GPUDirect RDMA, compatible with VPX systems.

The Condor NVB2000xX is a robust XMC video graphics and GPGPU card built on the advanced NVIDIA RTX PRO 2000 Blackwell GPU, which supports error correction code (ECC) and 8GB of GDDR7 graphics memory with a 128-bit memory interface. As the first architecture to support GDDR7, NVIDIA Blackwell has made significant advancements in bandwidth speed, improved signal integrity, and reductions in energy usage and power consumption, making it an ideal choice for demanding high-performance computing environments. It also supports three DisplayPort++/DVI video outputs. The Condor NVB2000xX is a high-performance card featuring the NVIDIA RTX PRO 2000 Blackwell GPU, equipped with 3,328 CUDA cores, demonstrating exceptional capabilities for AI processing and real-time analytics. It achieves high-speed data transfer with PCIe Gen 5 and GPUDirect RDMA, making it perfect for mission-critical VPX systems.

The Condor GR5S-B4000 is a rugged video graphics/GPGPU card in the 3U OpenVPX standard, equipped with the cutting-edge NVIDIA RTX PRO 4000 Blackwell GPU. Performance: It features 7,680 CUDA cores and 240 Tensor cores, delivering exceptional performance in edge AI inference and real-time situational analysis. Memory: It is equipped with 16 GB of ECC-supported GDDR7 memory, ensuring reliability for mission-critical applications. Connectivity: High-speed data transfer is achieved through PCIe Gen 4 switching and GPUDirect RDMA. Cooling: It supports both conduction cooling and airflow-through variations, maintaining stable operation even in harsh environments due to its new design. This card is the ideal solution for fields that demand extreme reliability and high computational power.

The Condor GR5S-AD2000 is a robust 3U OpenVPX standard video graphics/GPGPU processing card equipped with an NVIDIA RTX 2000 GPU (Ada Lovelace). Performance: It features 3,072 CUDA cores, 96 Tensor cores, and 24 RT cores, providing high efficiency and responsiveness for edge computing and data-intensive tasks. Memory: It is equipped with 8 GB of GDDR6 graphics memory with ECC (Error Correction Code), ensuring high reliability. Connectivity: It supports PCI Express Gen 4 connections (x4, x8, x16 lanes) and supports GPUDirect RDMA, significantly reducing data transfer latency. Cooling: Both air-cooled and conduction-cooled variations are available, offering robustness that meets the VITA 47.3 design specifications.

The Condor VC300xX is an ultra-high-definition (UHD) video capture solution in the XMC form factor designed for high-performance video processing. This device transfers raw video data to the host system with ultra-low latency via PCIe, enabling real-time viewing, processing, analysis, and display. This allows for seamless integration into workflows that require advanced video processing. One of its key features is its low power consumption of less than 20W, making it ideal for compact platforms sensitive to SWaP (Size, Weight, and Power) such as UAVs (unmanned aerial vehicles). This product is available in ruggedized variants for air cooling and conduction cooling, and can be integrated into systems aligned with VITA and SOSA standards. A comprehensive SDK is provided to simplify capture data management. It also supports VITA 46.9 VPX I/O signal mapping (X12d, X8d, X16s).

The Condor GR5S-AD5000 is an extremely robust 3U OpenVPX standard video graphics and GPGPU processing card equipped with an NVIDIA RTX 5000 GPU (Ada Lovelace architecture). This card features a powerful core configuration with 9,728 CUDA cores, 304 Tensor cores, and 76 RT cores, and it combines high reliability and computational performance with 16 GB of GDDR6 memory with ECC (Error Correction Code). It adopts PCI Express Gen 4 for its interface, and the support for GPUDirect RDMA significantly reduces data transfer latency. Additionally, it is developed in alignment with the SOSA standard, ensuring interoperability and scalability in military and aerospace applications. The cooling method offers both conduction cooling (CC) and airflow-through (AFT) variations to meet stringent thermal requirements. This card provides an unparalleled solution for data-intensive tasks, delivering exceptional efficiency and responsiveness.

The Condor GR2S-AD5000 is built on the NVIDIA Ada Lovelace architecture and supports hardware acceleration for the latest video codecs such as AV1, H.265 (HEVC), and H.264, as well as NVIDIA GPUDirect RDMA. The interface features a 3U VPX form factor equipped with a PCI Express Gen4 switch, supporting x4, x8, and x16 lane connections, and is compatible with PCIe branching and multi-card expansion, enabling flexible system integration and high-bandwidth data transfer. This product adopts a thermal frame structure designed to efficiently dissipate heat generated during operation, ensuring reliable performance for continuous computing work in harsh environments. It is available in both air-cooled and conduction-cooled configurations for robust systems.