“The potential for major returns in energy efficiency and performance gained from parallel processing are clear, and in fact critical, to support rapid growth in the telecoms industry. This makes innovations such as Adapteva’s very exciting for Ericsson and for the customers we work with,” –Sebastian Tolstoy, Ericsson VP Business Development, PA Radio, BNET
“We firmly believe in early engagement to enable our customers to deliver innovative products with industry-leading performance and energy-efficiency. Our close collaboration with Adapteva is helping them bring a whole new class of multicore microprocessors to market on the foundry industry’s only production-proven HKMG technology.” —Mike Noonen, Executive VP of Worldwide Marketing, GlobalFoundries
“Visual computing can enable gesture-based gaming, advanced user interfaces, augmented reality, and even improved health and safety. Visual processing, however, requires many more flops than voice processing. Adapteva’s architecture can deliver the performance required for visual computing.” —Linley Gwennap. Principal Analyst and Founder, The Linley Group
“Adapteva has delivered an architecture that supports massive on-chip parallelism with impressive power efficiency. OpenCL provides the perfect foundation for such a processor. We leveraged this API in an SDK that provides programmers with a clear path for code development. Programmers will find their parallel algorithms mapping naturally to the Epiphany architecture. The chip is designed for massively parallel algorithms. Extracting performance from the architecture becomes relatively easy as a result of this design.” –David Ritchie, CEO, Browndeer Technology
“It seems like our industry has not demonstrated much creativity in the past few years, and processing improvements have only come at the expense of increasingly ridiculous power consumption, which has in turn created burdensome cooling problems. We believe that Adapteva’s Epiphany architecture represents a flash of brilliance that reintroduces some much needed creativity into the embedded signal processing world” —Jeff Milrod, CEO, BittWare
“High performance embedded computing is naturally split into high throughput front end processing tasks and high complexity back end processing tasks. FPGAs are great at front end bit shuffling type operations but have proven challenging to use for high complexity algorithms, some requiring floating point scaling. Microprocessors are great for the highly complex processing with 100k lines of code found at the back end. However, none of the traditional products on the market today gives a satisfactory processing solution to the middle-end processing step where both massive throughput, low power, and complex floating point point math is needed. This is where Adapteva’s Epiphany multicore architecture comes in, with just the right amount of flexibility to provide outstanding energy efficiency while not sacrificing the traditional C-programming model we have grown accustomed to.” —Mikael Taveniku, CEO, XCube
“Adapteva’s Epiphany processor demonstrates how a small number of well designed instructions can work efficiently with GCC, making it the next generation of RISC architecture. The Epiphany architecture clearly shows that it’s possible to get outstanding compiler performance without a complex instruction set.” —Jeremy Bennett, CEO, Embecosm
“The future of HPC (and probably all computing) is both many core and heterogeneous. Trying to get there by adapting today’s commodity processors and GPUs has the benefit of better support for legacy applications, but at some stage this approach will break down. Adapteva’s MIMD architecture should scale better for programs with irregular data structures and program flows, and applying their Epiphany architecture to the hard problems in real time data analytics seems especially intriguing. Watch this space.” –John Barr, Industry Analyst
“Andreas and his team have managed a rather fantastic feat: to develop a genuinely useful microprocessor with many task-parallel cores, operating within the power-budget of smart-phones and embedded devices. Accomplishing all this with a team of only a few full-time employees may seem unbelievable, but I’ve seen how he operates his company and the magic is simple: Focus on the simplest solution to any given task, hire a consultant when you need expert advice, and stick to all schedules!” —Magnus Snorrason, Computer Vision Expert, Smart Systems Consulting
“Massive parallelism is the only way to meet both the computing performance demands and the power limitations in future, advanced embedded signal processing applications. Processor architectures need to be built on simple, scalable principles. Powerful processors and straightforward means for moving data around is a key to high throughput in streaming applications. To my understanding, the Adapteva Epiphany architecture has been designed based on such principles. It has been a positive experience for my research group and students to use the Epiphany and try it out on challenging industrial applications. Our students and post-docs have been able to use the architectural features as well as the programming tools efficiently already at their first tries.” –Bertil Svensson, Professor and Director of CERES, Halmstad University