RV1103
Cortex A7 + MCU
USD Bulk Pricing
Rockchip RV1103 Cortex A7 + MCU 0.5TOPS NPU
∞
Sold
7x24 Hours
Technical Support
1 V 1
Service
Product Highlights:
- 8nm process, quad-core Cortex-A76 @2.4GHz + quad-core Cortex-A55 @1.8GHz
- ARM Mali-G610 MC4 GPU, embedded high performance 2D image acceleration module
- 6.0 TOPs NPU, enable various AI applications
- 8K video codec ,8K@60fps display out
- Rich Display Interface, multi-screen display
Quote
Matching package
Rockchip RV1103
Key features
- Cortex A7 + MCU
- 3rd generation ISP, 4M30 2F HDR/3DNR/WDR
- 0.5TOPS NPU, enable various AI applications
- H.264/H.265 4M30FPS
- 4K@60FPS JPEG snapshot
- Six bit rate control modes (CBR, VBR, FIXQP, AVBR, QPMAP, and CVBR)
- Intelligent encoding mode
The Rockchip RV1103 represents a sophisticated and highly integrated vision processor System-on-Chip (SoC), specifically engineered for Industrial PC (IPC) and Artificial Intelligence (AI) applications. It utilises a single-core ARM Cortex-A7 32-bit processor, equipped with NEON and Floating Point Unit (FPU) support. The architecture encompasses a 32KB instruction cache, a 32KB data cache, and a 128KB unified Level 2 (L2) cache. Its embedded Neural Processing Unit (NPU) supports hybrid operational modes with INT4, INT8, and INT16 precision, delivering computational capabilities of up to 0.5 Trillions of Operations Per Second (TOPs).
The platform demonstrates excellent compatibility with models from prevalent frameworks such as TensorFlow, MXNet, PyTorch, and Caffe, facilitating straightforward model conversion. Additionally, RV1103 is equipped with a next-generation hardware-based Image Signal Processor (ISP) capable of handling resolution up to 4 megapixels. It includes multiple algorithm accelerators for functionalities such as High Dynamic Range (HDR), 3A algorithms, Lens Shading Correction (LSC), 3DNR, 2DNR, sharpening, dehazing, and gamma correction. When integrated with two MIPI CSI or Low-Voltage Differential Signalling (LVDS) interfaces, it supports systems capable of concurrently receiving video data from two camera sensors.
The integrated video encoder supports H.265 and H.264 formats and enables multi-stream encoding, facilitating high-resolution video capture, local storage, and real-time transmission of lower-resolution streams to cloud storage. The system features an AI-enhanced video engine comprising 22 processing units to expedite video processing tasks. Furthermore, it includes embedded 16-bit DDR2 DRAM for high memory bandwidth, as well as built-in Power-On Reset (POR), an audio codec, and MAC PHY/AC PHY components.
DETAIL SPEC
CPU | Cortex A7 + MCU |
Video/Image Digital Processor | 3rd generation ISP, 4M30 2F HDR/3DNR/WDR |
Supports up to 2 sensor inputs | |
NPU | Supports 0.5T computing power |
Encoder | H.264/H.265 4M30FPS |
4K@60FPS JPEG snapshot | |
Six bit rate control modes (CBR, VBR, FIXQP, AVBR, QPMAP, and CVBR) | |
Intelligent encoding mode | |
8-area OSD | |
Ultra-low delay encoding | |
Integration | Built-in 64MB DDR2 |
Built-in 100M Ethernet PHY, Audio codec, POR | |
Package | QFN88 9mm x 9mm |