The HiFi Audio DSP 24-bit Family Guide

Products » Audio DSP » HiFi 2 and HiFi EP

Proven 24-bit Audio DSPs for SOC Designs

Both the HiFi 2 and HiFi EP Audio DSPs are designed around a proven 24-bit architecture for low gate count and the lowest possible power. HiFi EP adds advanced optimizations for DTS Master Audio, improved voice pre- and post-processing, and an improved cache memory subsystem.

Both HiFi DSPs are pre-packed, ready to drop into SOC designs. They provide programmable support for multiple audio formats with over 80  software codeds and audio and voice enhancement packages pre-ported and ready to use, no software porting required.

Optimized for Audio

Tensilica used its automated processor generation capability when designing the HiFi DSPs to optimize them for digital audio streams.They have been extended and optimized by Tensilica to run all the audio formats. They can approach the performance of fixed-gate implementations, while providing the flexibility, time to market, and risk reduction of fixed instruction set RISC processors while simultaneously reducing silicon area and power consumption. Tensilica can provide the codecs so you don’t have to worry about that development process.

The HiFi Audio DSPs utilize Tensilica’s benchmark-breaking Xtensa LX processor technology with custom instructions (better code density, fewer cycles), flexible datapath widths (higher efficiency per cycle), multiple instructions per cycle (VLIW – higher efficiency per cycle), and simultaneous data operations (SIMD – parallelization). A comprehensive Eclipse-based software development environment is provided that includes everything from a source code editor, debugger, and ISS to an optimized compiler that provides excellent code density.

The HiFi Audio DSPs have been extended and optimized by Tensilica to run all the audio formats. They can approach the performance of fixed-gate implementations, while providing the flexibility, time to market, and risk reduction of fixed instruction set RISC processors while simultaneously reducing silicon area and power consumption. Tensilica can provide the codecs listed so you don't have to worry about that development process.

Tensilica's HiFi 24-bit Audio is available two ways:

1. The HiFi 2 Audio DSP-a click-box options for the Xtensa LX processor, so you can do additional optimizations

2. The HiFi EP Audio DSP-a superset of the HiFi 2 Audio DSP with advanced optimizations for DTS Master Audio, improved voice pre- & post-processing, and improved cache memory subsystem.

Additionally, Tensilica offers HiFi 3, a 32-bit audio DSP with higher performance with lower power for high-performance audio post-processing and voice processing algorithms used in smartphone and home entertainment systems.

HiFi EP Architecture Enhancements

There are three major enhancements in the HiFi EP Audio DSP:

1. HiFi EP includes a 32x24 MAC for higher performance at lower power on the popular DTS Master Audio lossless decoder, resulting in a clock rate reduction of almost 35% compared to HiFi 2.

2. To address the increasingly demanding requirements in mobile and VoIP applications for better immunity to background noise and speakerphone mode quality, new instructions have been added to accelerate voice pre- and post-processing for noise cancellation and beam forming microphones. These instructions also provide better general DSP capabilities

3. The cache memory subsystem is enhanced with an integrated predictive fetch unit to significantly improve performance in SOC designs with large external memory latencies.

Power Savings

Through a combination of significantly lower power per MHz and architectural optimization of the HiFi 2 Audio Engine instruction set, the HiFi Audio Engine delivers dramatic improvements in energy efficiency that lead to increased battery life for portable and wireless applications. Based on TSMC's 40 nm LP process and a minimal HiFi 2 configuration, power including memories is as low as 21 μW/MHz dynamic and 7.5 uW static. Decoding a typical MP3 file at 5.7 MHz requires only 0.13 mW of total power.

Simplified Programming

A key advantage of the HiFi Audio Engine is its simple programming model. Because of the efficiency of the optimized audio and voice instructions, software developers can port audio and voice codecs completely in C while maintaining or surpassing the performance of assembly on other DSP and CPU architectures.

Lower Gate Count

Through careful engineering and by fully exploiting the features of the Xtensa LX architecture, Tensilica was able to implement the Xtensa HiFi 2 Audio Engine ISA in approximately 45K gates. A minimal configuration, including the Xtensa LX processor, is approximately 67K gates.

Performance to Spare

Because of the efficiency of the HiFi Audio architecture, the HiFi Audio DSPs require only a fraction of the available MHz to perform the audio decoding and encoding functions. Targeting high performance applications, the Xtensa HiFi Audio Engine can support speeds greater than 900 MHz in TSMC 45GS, 600 MHz in TSMC 65GP, and 500 MHz in TSMC 40LP, leaving lots of headroom for other control, media and signal processing tasks


Marketing Agency