Next-Generation DSP Architecture for a Data-Centric World
The MIPS ARC® VPX DSP IP family is optimized for the unique power, performance and area (PPA) requirements of embedded workloads such as IoT sensor fusion, radar and LiDAR processing, engine control, voice/speech recognition, natural language processing and other edge AI applications. The VPX processors are based on an enhanced ARCv2DSP instruction set and operate on 128-bit (VPX2, VPX2FS), 256-bit (VPX3, VPX3FS), 512-bit (VPX5, VPX5FS) and 1024-bit (VPX6, VPX6FS) vector words based on the same very long instruction word (VLIW)/single instruction-multiple data (SIMD) architecture.
The safety-enhanced ARC VPXxFS processor IP integrate hardware safety features including error correction code (ECC) protection for memories and interfaces, safety monitors and lockstep mechanisms that help designers achieve the most stringent levels of ISO 26262 functional safety compliance.
The VPX processors are supported by the MIPS ARC MetaWare Development tools, including a vector length-agnostic software programming model specifically optimized for the VPX hardware architecture. The MetaWare compiler’s auto-vectorization feature transforms sequential code into vector operations for maximum throughput.
What's New
Optimizing Sensor Fusion: The High-Performance MIPS ARC VPX DSP Processor IP
ARC VPX DSPs: Scalable Vector Processing for High-Performance Embedded Applications
Products
ARC VPX2/VPX3/VPX5/VPX6 DSP Processors
Scalable DSP processors for AI, IoT, automotive, and voice processing applications, featuring 128-bit, 256-bit, 512-bit, 1024-bit vector SIMD/VLIW, and floating point & math computation engines.
ARC VPX2FS/VPX3FS/VPX5FS/VPX6FS DSP Processors
Scalable DSPs for safety-critical automotive applications. The processors feature a self-checking safety monitor, are ISO 26262 certified up to ASIL-D and ISO/SAE 21434 CyberSecurity compliant.
ARC Development Tools and Software
- MetaWare Development Toolkit (compilers, debugger, and simulator)
- ARC Access Program (portfolio of 3rd party tools, operating systems, and middleware)
Products and Licensable Options
PPA Efficiency
ARC processors are optimized to deliver the best PPA efficiency in the industry for embedded SoCs.
- Harvard architecture for higher performance through simultaneous instruction and data memory access
- High-speed pipeline designed for maximum power efficiency
- 32-bit RISC engine offers a mixed 16-bit/32-bit instruction set for greater code density
Configurability
ARC processors are highly configurable, allowing designers to optimize the performance, power, and area of each processor instance on their SoC.
- Add or omit hardware features to optimize the core for your target application – no wasted gates
- The ARChitect wizard enables drag-and-drop configuration of the core
Extensibility
ARC Processors EXtension (APEX) technology enables users to customize their processor implementation.
- Add user-defined instructions to accelerate software execution and reduce code size, reducing energy consumption and memory requirements
- Tightly couple memories and peripherals to the processors to eliminate the need for additional bus infrastructure, reducing area and latency and increasing system-level performance
