The introduction of the Edison cylinder phonograph 130 years ago marked the beginning of the ability to distribute entertainment on physical media. Needless to say, this is coming to an end; and whereas networked, portable distribution and playback of music is becoming extremely convenient to consumers and content providers, it is putting a burden on designers of consumer audio products. Faced with an entirely new level of complexity, engineers are struggling to get back in the driving seat of development and focus on their core competencies and differentiating factors.

The Blu-ray disc most likely marks the end of the road for transportation of media content via boat, plane or car. One may wonder why it took so long. But the distribution of media, as any data, is quickly shifting to the global network infrastructure. Digital audio, requiring less bandwidth, has gotten a head start. According to the Consumer Electronics Association of America[1], the US resale volume of stationary CD players will drop about 60 percent to 65,000 units by 2014, whereas sales of devices that can play music from a portable media player is estimated at 330 million.

The Times They Are ‘a-Changing’
As the majority of audio content is delivered to the end user over the Internet, music is either streamed from online subscription through on-demand services such as Pandora in the US or Spotify in Europe. Alternatively, it is purchased from online music stores such as iTunes and stored locally on a computer or portable network player in a compressed form.

This paradigm shift in usage model imposes a challenge on designers of tomorrow’s audio equipment. Still, audio devices need to amplify and process the signal for audio reproduction – however, they must also connect to the source via a wired or wireless connection and often decompress content on-the-fly. Also, as the signal source lies outside the system, designers take the opportunity to integrate connectivity, speakers, and amplifiers in the same cabinet. In the new paradigm, the source-to-sound signal path involves all or some of the following steps:

1. Connectivity (Ethernet, USB host and/or device, SDIO, WiFi, Bluetooth)
2. Decompression (eg MP3, AAC, WMA)
3. Digital signal processing (equalization, volume control, room correction, etc.)
4. Digital-to-analog conversion
5. Amplification
6. Acoustic reproduction

Few Masters of All Trades
Over the past few years the electronics industry has seen an explosive growth in companies that provide high performance, cost-effective connectivity solutions (step 1). Likewise there is a plethora of companies that address steps 3 and 4 really well. Since the introduction of the CD (compact disc) in the 1980s, step 4 has become common knowledge among all audio manufacturers, and companies that make amplifiers (step 5) and loudspeakers (step 6) have been around “forever”. It seems that the further down the signal path, the more mature the technology.

Here, lies the core of the problem imposed by today’s paradigm. Few companies are experts on computer networking, wireless baseband processing, RF, embedded processing, digital signal processing, analog signal processing and filtering, amplifier and power supply design and acoustics – all areas of know-how prone to end up enclosed in a single audio application cabinet.

A manufacturer of guitar amplifiers, who decides to include a USB port to allow the guitar player to jam along with his favorite MP3 tunes stored on a USB thumb drive, is likely to discover this requires designers to step out of their level of expertise. This is also the case with a loudspeaker manufacturer that decides to combine their core competence in acoustics with an amplifier and iPod docking capabilities.

No matter what part of the signal path companies define as their comfort zone, they are likely to find the level of complexity overwhelming and seek a cost effective turn-key audio application platform that provides:

o Versatility
o No-compromise performance
o Design convenience

An End-to-End Platform
During the recent years, Atmel’s 32-bit AVR32UC3 microcontroller has been embraced by the audio industry. A recent collaboration between Atmel and the Danish company Bang&Olufsen ICE power, an industry leader in Class D amplifier technology, has led to the engineering of a reference design that aims at providing relief to engineers who struggle getting the pieces of the digital audio puzzle assembled. The turnkey solution forms an application basis spanning a wide range in cost and performance - from clock radios to high-end audiophile sound systems.

Versatility – Any Input, Any Output
In order to provide scalability across a wide range of product ranges, the reference design supports nearly any input source and any playback option. The AT32UC3 microcontroller features interfaces to SDIO as well as USB host and device, and Atmel provides all the SW necessary to hook the user application to these. Through an SPI port, the front end of the design can be hooked up to wireless interfaces such as WiFi, Bluetooth or Ethernet. In other words - almost any conceivable signal source, and provides the capability to decode MP3, AAC and WMA compressed audio to boot.

The MS3 single-chip audio system contains a configurable fixed-function DSP engine and a class D amplifier capable of delivering 2x1.5W of output power, sufficient for directly driving a pair of small speakers such as the ones found in laptop computers, cell phones, navigation systems or even small iPod docking stations. The MS3 also features a headphone amplifier and can also be used as a premium quality audio DAC driving a power amplifier for bigger speakers. The design interfaces directly to Bang&Olufsen ICE Power’s ASX series of class D power stages providing from 25 W to 1000 W of audio performance per channel.

Engineers only need to identify the signal source and the loud speakers, configure, connect and get on with building differentiating values and get their cool products out the door fast. Spiraling complexity is not the only challenge imposed by the new paradigm – throw shortened design cycles and a desperate need for differentiation into the mix, and everybody quickly understand that solid, ready-to-run platforms are the way to propel digital audio designs into the future.

[1] CEA Final Detailed Forecast, July 2010

For more information, please see and