
Similarities
-
Fanless Design: Both models feature a fanless design with a passive heatsink, ensuring reliable operation in harsh environments.
-
NVIDIA Jetson Orin™ NX: Both are powered by the NVIDIA Jetson Orin™ NX module, offering excellent AI performance and energy efficiency.
-
Connectivity: They offer rich connectivity options, including dual Gigabit Ethernet (one with PoE), multiple USB ports, DI/DO, COM, HDMI, and M.2 slots for storage and expansion.
-
Industrial-Grade Design: Designed for industrial environments, both support a wide operating temperature range and offer various mounting options (desk, DIN rail, wall, VESA).
-
Software Support: They support the NVIDIA JetPack SDK, providing a comprehensive suite of tools and libraries for AI development and deployment.
-
Storage: Both come with a 128G M.2 NVMe 2280 SSD
Differences
Feature | reComputer Industrial J4011 | reComputer Industrial J4012 |
AI Performance | 70 TOPS | 100 TOPS |
Memory | 8GB LPDDR5 | 16GB LPDDR5 |
CPU | 6-core Arm® Cortex®-A78AE | 8-core Arm® Cortex®-A78AE v8.2 64-bit |
Advantages
Feature | reComputer Industrial J4011 | reComputer Industrial J4012 |
AI Performance/Use Cases | Suitable for entry-level AI tasks, edge inference, machine vision, and AI video analytics. Ideal for applications where cost-effectiveness is a priority. | Best for demanding AI workloads, generative AI, multi-stream video analytics, and complex AI models. Excellent for applications requiring high performance and large memory capacity. |
Generative AI Capability | Capable of Bringing Generative AI to the Edge | Capable of Bringing Generative AI to the Edge |
Computer Vision | Build Multi-streams AI Video Analytics and Fastest Way to Deploy Generative AI and Computer Vision Models including Ollama, Llama3; computer vision including YOLOv8 | Build Multi-streams AI Video Analytics and Fastest Way to Deploy Generative AI and Computer Vision Models including Ollama, Llama3; computer vision including YOLOv8 |
Detailed Comparison Table
Specification | reComputer Industrial J4011 Fanless Jetson Orin™ NX 8GB | reComputer Industrial J4012 Fanless AI Jetson Orin™ NX 16GB |
Jetson Module | Jetson Orin NX | Jetson Orin NX |
AI Performance | 70 TOPS | 100 TOPS |
GPU | 1024-core NVIDIA Ampere architecture GPU with 32 Tensor Cores | 1024-core NVIDIA Ampere architecture GPU with 32 Tensor Cores |
CPU | 6-core Arm® Cortex®-A78AE | 8-core Arm® Cortex®-A78AE v8.2 64-bit |
Memory | 8GB 128-bit LPDDR5 | 16GB 128-bit LPDDR5 |
Video Decode | 1x 8K30 (H.265), etc. | 1x 8K30 (H.265), etc. |
Video Encode | 1x 4K60 (H.265), etc. | 1x 4K60 (H.265), etc. |
DL Accelerator | 2x NVDLA v2 | 2x NVDLA v2 |
Storage | 1x M.2 Key M PCIe Gen4.0 SSD (128G NVMe) | 1x M.2 Key M PCIe Gen4.0 SSD (128G NVMe) |
Ethernet | 2x RJ45 GbE (1x PoE) | 2x RJ45 GbE (1x PoE) |
Wi-Fi/Bluetooth | Support SMD (Customizable) | Support SMD (Customizable) |
M.2 Key B | 1x (4G/5G Module Optional) | 1x (4G/5G Module Optional) |
USB | 3x USB3.2, 2x USB2.0 Type C | 3x USB3.2, 2x USB2.0 Type C |
DI/DO | 4x DI, 4x DO | 4x DI, 4x DO |
COM | 1x DB9 (RS232/RS422/RS485) | 1x DB9 (RS232/RS422/RS485) |
Display | 1x HDMI 2.0 | 1x HDMI 2.0 |
SIM | 1x Nano SIM | 1x Nano SIM |
Fan | Fanless (Passive Heatsink) | Fanless (Passive Heatsink) |
Power Supply | DC 12V-24V | DC 12V-24V |
Dimensions | 159mm x 155mm x 57mm | 159mm x 155mm x 57mm |
Operating Temperature | -20 ~ 60°C | -20 ~ 60°C |
Mounting | Desk, DIN rail, Wall, VESA | Desk, DIN rail, Wall, VESA |
FAQs
1.What operating system is pre-installed on the reComputer Industrial J4011 and J4012?
Both come with the NVIDIA JetPack SDK, which includes Ubuntu Linux.
2.Can I connect a 4G or 5G module to these devices?
Yes, both have an M.2 Key B slot for optional 4G/5G module installation.
3.Do these computers support Wi-Fi and Bluetooth?
Both support SMD Wi-Fi/Bluetooth through customized production. Contact Seeed Studio for more details.
4.What is the power consumption of these devices?
The power consumption varies depending on the workload but is generally energy-efficient due to the Jetson Orin™ NX module.
5.What type of applications are these reComputers best suited for?
They are well-suited for AI video analytics, machine vision, autonomous mobile robots (AMR), generative AI, and other edge computing applications. The J4012 is better for more demanding tasks due to its higher AI performance and memory capacity.