SOEDGE
BY PINE64
A Powerful AI Compute Module
The SOEDGE is a 3TOPS compute module that can be paired with a dedicated baseboard or the SOEDGE Baseboard for development. It can also be connected to a SBC, such as the ROCKPro64, or a regular X86 PC using PCIe or USB 3.0 using special adapters. The SOEDGE is, at the time of writing, in early stages of development; if you are a developer interested in Neural Networks and open hardware, please contact us directly for a development unit.
Features and Specifications
| SoC | 2 x ARM Cortex A35 Processor @ 2.0 GHz |
| Memory | DDR4 RAM (2GB) |
| Features | A power-efficient ARM 64-Bit Armv8-A architecture AArch32 for full backward compatibility with Armv7 Supports NEON Advanced SIMD (Single Instruction Multiple Data) instruction for acceleration of media and signal processing function Support Large Physical Address Extensions(LPAE) VFPv4 Floating Point Unit 32KB L1 Instruction cache and 32KB L1 Data cache AArch64 for 64-bit support and new architectural features TrustZone security technology Neon Advanced SIMD DSP and SIMD extensions VFPv4 Floating point Hardware virtualization support 128KB L2 cache 128Mb SPI Flash |
| Form factor | SODIMM form-factor |
SODIMM Form-Factor
With the small SODIMM form-factor the SOEDGE is perfect for use as a embedded project
2GB DDR4 RAM
Each SOEDGE modules includes 2GB DDR4 RAM
Rockchip RK1808
The RK1808 uses a power efficient Armv8-A archeitecture with dual ARM Cortex A35 cores at 2 GHz
NPU
The RK1808 SoC has a Verisilicon Vivante NPU
Documentation
| SOEDGE |
| Schematics and certifications |
