A few questions for Brian Faith, VP of Worldwide Marketing at Quicklogic:
Q: It’s our round – what are you having?
A: How about a few cubic hectares of some nice, cool H2O? I’m parched from all the briefings we’ve been doing on our new EOS platform, but we’re in such a bad drought here in California, 1I’m afraid to turn on the faucet!
Q: What is your product and what need does it address?
A: We have just introduced our new EOS™ multi-core sensor processing platform, which enables the world’s most advanced sensor-based applications for mobile, wearable and IoT devices at a fraction of the power consumption of traditional micro-controllers.
Q: How is the product unique?
A: The EOS platform is a multi-core SoC that incorporates three dedicated processing engines. These include QuickLogic’s proprietary, patent-pending microDSP-like Flexible Fusion Engine (FFE), an ARM® Cortex® M4F Microcontroller (MCU), and a front-end sensor manager. The FFE and sensor manager handle the bulk of the algorithm processing, which minimizes the duty cycle for the floating point MCU. This approach dramatically lowers aggregate power consumption, and enables next generation sensor-driven applications, such as pedestrian dead reckoning (PDR), indoor navigation, and motion compensated heart rate monitoring.
Q: Does the development of this product have an interesting story/history?
A: QuickLogic has been developing devices for mobile applications for some time now, and we quickly learned that it wasn’t just about computational bandwidth – it was about that bandwidth at very low power consumption. Our conclusion was that delivering that low power required a different architecture – and it required challenging conventional wisdom that using several RISC engines is the way to reduce power. Through our analysis of common sensor fusion algorithms, we realized we could develop a ?DSP-like engine, called the Flexible Fusion Engine (FFE), optimized for those types of mathematical functions And now we’ve tightly coupled that FFE with an industry standard ARM Cortex-M4 to take care of the more general purpose computational tasks. We also added extensibility into the architecture by including our re-programmable logic to give users the ability to customize the platform for their specific needs – including adding a second FFE. We were amazed at the results, seeing power levels of 80 ?Watts / DMIPS for our M4 implementation and roughly 12.5 ?Watts / DMIPS for our FFE.
Q: Who is your ideal customer?
A: Our ideal customer for the EOS sensor processing platform is a developer of mobile, wearable, or IoT devices who needs to deliver advanced sensor-based capabilities – perhaps more than what they are doing today – at much lower power. We want to solve their battery life challenge by giving them more MIPS at a much lower power curve.
Q: Are there future iterations/releases/extensions of this product planned and when?
A: This is our 3rd generation sensor processing platform, and we will continue to add functionality and push the performance/power envelope. In the immediate future our efforts will be focused on our extensive SenseMeTM sensor algorithm library by adding new features and enhancing the ones we already offer. Software support is a critical component of the solution and we know that Android compliance is of the utmost importance – and we will see continued support in this area from QuickLogic.
Q: Anything else you wish to share about the product?
A: We know that customers need an easy-to-use and comprehensive development environment. Since the platform is sensor and algorithm agnostic, it can support third party and customer-developed algorithms through QuickLogic’s industry-standard Eclipse Integrated Development Environment (IDE) plugin. The IDE provides optimized and proven code generation tools as well as a feature-rich debugging environment to ensure quick porting of existing code into both the FFE and the ARM M4F MCU of the EOS S3 platform.