HiFi 2 & HiFi EP Audio DSP Product Brief
330HIFi Audio DSP Product Brief
388VDO Video DSP Product Brief
Everything You Wanted to Know About Blu-ray Audio, but were afraid to hear
Put Low-Power, Low-Overhead, High-Fidelity Digital Sound in Your Next ASIC or SOC
A Designer's Guide to HD Video Pre- and Post-Processing
How to Add Low-Power, Multi-Codec Digital Video and Audio to Your Next ASIC or SOC Design
Need audio in your next SOC design? You have three choices:
1. The HiFi 2 Audio DSP - a click-box option for the Xtensa LX3 processor, so you can do additional optimizations,
2. The 330HiFi pre-configured core that incorporates the HiFi 2 Audio DSP. Just drop it into your SOC design.
3. The HiFi EP Audio DSP - a superset of the HiFi 2 Audio Engine with advanced optimizations for DTS Master Audio, improved voice pre- & post-processing, and improved cache memory subsystem
Over 60 popular audio codecs have been pre-ported to the HiFi Audio DSP, which makes it a “drop-in" block for any SOC application requiring high quality, 24-bit audio capability.
Specialized audio instructions designed into the HiFi Audio DSPs by Tensilica increase code density (reducing memory requirements) and reduce Mhz requirements (lowering power). Techniques used to define these custom instructions include Single Instruction, Multiple Data (SIMD - parallelization), VLIW (multiple operations per cycle), and flexible data path widths (optimum data widths to reduce power).
All of these advanced processor techniques are pre-built into the HiFi hardware, and the codecs supplied by Tensilica utilize this hardware to produce the most efficient programmable audio engine available for licensing.
The following audio packages are part of the Xtensa HiFi Audio DSPs:
The following audio packages are available for the HiFi 2 Audio Engine:
The following voice packages are available for the HiFi 2 Audio Engine:
Because the HiFi DSPs are programmable, multiple standards can run on the same hardware, allowing the same device to play or record digital audio in different standards.
A key advantage of the HiFi Audio DSPs is its simple Xtensa processor-based 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.
HiFi EP includes a novel and unique 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.
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.
The cache memory subsystem is enhanced with an integrated predictive prefetch unit to significantly improve performance in SOC designs with large external memory latencies.
In the chart below, you can see how the optimizing instructions in HiFi EP really pay off in a Blu-ray Disc application.
| Audio Function | Worst Case Performance on HiFi 2 | Worst Case Performance on HiFi EP |
| Primary Audio: DTS Master Audio 24.5 Mbps 5.1 192 kHz | 175 MHz | 115 MHz |
| Secondary Audio: DTS-LBR (DTS Express) 256 kbps 5.1 48 kHz decoding | 63 MHz | 63 MHz |
| Upsample, mixing, post-processing | 48 MHz | 48 MHz |
| Downsample, Downmix, S/PDIF Output: DTS/AC-3 5.1 48 kHz encoding | 76 Mhz | 76 Mhz |
| Subtotal (ideal memory) | 362 MHz | 302 MHz |
| Allowance for cache/bus latency | 100 cycles memory delay | 100 cycles memory delay |
| Total | 541 Mhz | 386 MHz |
