Tensilica Vision DSP doubles vision and AI performance

May 21, 2019 //By Rich Pell
New DSP IP doubles embedded vision, AI performance
Cadence Design Systems (San Jose, CA) has introduced a new Tensilica Vision DSP delivering up to 1.82 tera operations per second (TOPS) for demanding edge vision and AI processing applications.

An expansion of the company's Tensilica Vision DSP family, the Cadence Tensilica Vision Q7 DSP is claimed to double vision and AI performance for automotive, AR/VR, mobile, and surveillance markets. To address the increasing computational requirements for embedded vision and AI applications, the sixth-generation Vision Q7 DSP provides up to 2X greater AI and floating-point performance in the same area compared to its predecessor, says the company.

The Vision Q7 DSP IP is specifically optimized for simultaneous localization and mapping (SLAM), a technique commonly used in the robotics, drone, mobile and automotive markets to automatically construct or update a map of an unknown environment, and in the AR/VR market for inside-out tracking.

"For edge computing in our target markets, offloading vision applications on a high-performance, low-power, highly flexible DSP is a must," says Lazaar Louis, senior director of product management and marketing for Tensilica IP at Cadence. "Cadence has a long and successful track record spanning six generations of Vision DSPs, and the Vision Q7 DSP was designed to address the needs of our key customers deploying highly complex vision and AI algorithms, including SLAM for perception. The Vision Q7 DSP strengthens our very successful automotive portfolio, bringing leading-edge computation to the 'computer in the car' that can be compliant with safety requirements like ISO 26262.”

Key features include:

  • Very long instruction word (VLIW) SIMD architecture delivers up to 1.7X higher TOPS compared to the Vision Q6 DSP in the same area
  • An enhanced instruction set supporting 8/16/32-bit data types and optional VFPU support for single and half precision enables up to 2X faster performance on SLAM kernels compared to the Vision Q6 and Vision P6 DSPs
  • Delivers up to 2X improvement in floating-point operations per mm 2 (FLOPS/mm2) for both half precision (FP16) and single precision (FP32) compared to the Vision Q6 and Vision P6 DSPs
  • Up to 2X greater AI performance in the same area compared to the

Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.