Cirrus Logic


Beginnings and development of the Home Audio team

Back in 2004, an American semiconductor company, Cirrus Logic, which specializes in mixed signal audio as well as DSP processing, was looking for a C-compiler for its proprietary 32-bit fixed point DSP engine. It was well known at the time that C-compilers were not readily available for DSP cores, unlike MCU cores, for which there was an abundance of well-developed software and firmware development tools. Given the unique architecture of DSP cores (data and program memory architectures, hardware multipliers, various shift registers and general lack of rich register files) there were very few developers of C-compilers on the open market for DSP cores.

Through various industry connections a link was set up between Cirrus Logic and the RT-RK’s software tools development team and early in 2005 the first contract was signed, and a C-compiler development project started. About 12 months later, the first C-compiler was available. Soon thereafter, a follow-up project related to further optimization of the C-compiler performance was agreed.

However, in parallel with software tools development projects, Cirrus Logic was eager to expand the relationship with the talented RT-RK team in other areas of DSP application development. Over the next 4-5 years, the team had grown from initially less than 10 engineers to more than thirty engineers skilled in various aspects of DSP programming, both applications as well as tools development.

The first teams for DSP programming and tools development

RT-RK’s Home Audio (HA) team working on Cirrus Logic proprietary DSP platforms became an essential extension to the existing Cirrus’ team based in Austin, Texas. An extremely cordial relationship was developed, and many American based engineers became good friends with the engineers in Novi Sad, keeping some of those friendships to this day.

During that time, RT-RK’s engineers developed major audio decoding as well as encoding technologies on Cirrus DSP, primarily related to Dolby and DTS technologies but also including other audio technologies, like Audyssey and Dirac room correction, as well as many proprietary audio processing modules called DSP audio primitives.

Around 2008/2009 Cirrus Logic was looking to further expand the scope of work with RT-RK and a new major functional group around software testing (SQA) was founded and it grew rapidly. Soon thereafter, RT-RK became the sole testing center for the entire Cirrus’ DSP software activity, both software tools and firmware applications.

In 2012 a major shift happened in the audio market with Dolby’s introduction of an object audio format called Dolby Atmos. Soon DTS followed with DTS:X, Fraunhofer with MPEG-H, and Auro with Auro3D. RT-RK’s HA team started development of what was the most complex audio algorithm that consumer audio electronics had seen up to that point. Cirrus Logic and RT-RK’s HA teams worked together and secured some major design wins in Japan with the single chip audio solution, compared to four discrete DSPs that competition needed at that point. That was all happening in 2014.

In 2015, Cirrus Logic decided to focus on portable audio opportunities, de-emphasizing the consumer audio side. At that time, a major technology transfer agreement was signed with Cirrus Logic, with the understanding that RT-RK’s HA team would take over the entire software development process and infrastructure, and Cirrus Logic continue to manufacture the necessary ICs required for the consumer audio market. That was a major inflection point in RT-RK’s HA team development, since we became the independent software company selling audio software products and associated system and testing services. We have been operating in that mode ever since, with the team growing from some twenty engineers at the beginning of this process to about 45 today.

DSP tools team has continued the development of the state-of-the-art Eclipse based CLIDE system (Cirrus Logic IDE) which is at par with other major DSP powerhouses like TI or ADI.

Throughout this process, equally important was development of the system-wide eco-system and necessary utilities that enable easy system level adoption and integration of our DSP solutions into the final product. In particular, the development of sophisticated real-time audio centric operating system and utilities like DSP condenser allow for seamless integration of our DSP solution into final audio system – soundbars and AV receivers. This ease of integration is highly appreciated by all our customers.

The team and labs, 2015

Onto previous NSDC


Onto next NSDC