EE Times latest Silicon 60, our list of emerging startups, adds 17 innovators in version 14.0. (Our version 13.0 list, from April, is here. The newcomers include EDA, memory technology and processor companies along with sensors and haptics, wireless communications, power semiconductors, optoelectronics, audio and security.
Silicon 60 companies are selected based on a mix of
criteria, including: technology, intended market, maturity, financial
position, investment profile and executive leadership. Most importantly, they're emerging companies we believe are important to follow. The version 14.0 newcomers are highlighted in red.
ActLight SA (Lausanne, Switzerland), founded in 2010, by Serguei
Okhonin, is a fabless semiconductor company developing designs and
intellectual property for the combination of light and logic on silicon
chips, including the use of CMOS for optical conversion for use in
energy harvesting. www.act-light.com
Adapteva Inc. (Lexington, Mass.), founded in 2008, Adapteva is a
lean startup that has developed the Epiphany architecture of multicore
processors. Adapteva, led by founder Anders Olofsson, is targeting
defense and mobile consumer applications. The company claims to have
reached break-even point with less than $2 million of investment. www.adapteva.com
Adesto Technologies Corp. (Sunnyvale, Calif.) was founded by CEO
Narbeh Derhacobian in 2006. The company is developing a nonvolatile
memory based on programmable metallization cell technology licensed from
Axon Technologies Corp., a spinoff of Arizona State University. The
company is backed by Arch Venture Partners and Applied Ventures amongst
other venture capital companies. www.adestotech.com
Amantys Ltd. (Cambridge, England)
was established by former ARM executives in 2010 as a power electronics
company. The company aims to use digital control to transform power
electronics in medium- and high-voltage applications. The company is
backed by Moonray Investors and ARM Holdings plc. www.amantys.com
Ambiq Micro Inc. (Westlake Hills, Texas) founded in 2010, is a
fabless chip company developing low power wireless processors based on
ARM architecture and mixed-signal systems. Investors include Cisco
Systems and ARM Holdings. www.ambiqmicro.com
Andes Technology Corp. (Hsinchu, Taiwan), founded in 2005, is a
developer and licensor of 32-bit processor technology and associated SoC
platforms intended for a variety of embedded applications. The
company’s U.S.-trained founders are aiming Andes cores at the borders of
markets owned by established licensors such ARM, MIPS and Tensilica. www.andestech.com
Ausdia Inc. (Sunnyvale, Calif.) is a
timing verification EDA company formed in 2006 which commenced product
development in 2008. The company's TimeVision product is a timing
constraints development tool intended to help with static timing
analysis and sign-off for complex system-on-chip designs www.ausdia.com
Avalanche Technology Inc. (Fremont
Calif.) is a attempting to bring to market a form of magnetic random
access memory for memory and storage applications. Founded in 2006, the
company received $30 million in venture capital funding in mid-2012. www.avalanche-technology.com
Brite Semiconductor (Shanghai) Corp.
(Shanghai, China) was founded in 2008 in Zhangjiang Hi-tech Park. It is
an SoC and ASIC design company that pulls together intellectual
property, foundry, test and packaging technologies to create custom
silicon for its customers. Brite is backed by local foundry
Semiconductor Manufacturing International Corp. and Open-Silicon Inc.
(Milpitas, Calif.). www.britesemi.com
Calxeda Inc. (Austin, Texas) was founded in 2008 under the name
Smooth-Stone and is developing server-on-a-chip processors based on
multiple ARM processor cores. The company has received $48 million in
investments from venture capital funds and strategic investors including
ARM, ATIC and Texas Instruments. www.calxeda.com
Cognivue Corp. (Gatineau, Quebec)
was spun off in 2009 from Korea's MtekVision to focus on so-called
cognitive processing, that is, parallel processing engines optimized for
tasks like image recognition and identification. CogniVue provides
SoCs, software and IP to capture, analyze, and render video and images
for smart cameras. www.cognivue.com
Cosmic Circuits Inc. (Bangalore, India), founded in 2005, is a
licensor of analog and mixed-signal intellectual property circuits
cores. The company provides a broad spectrum of IP ranging from power
management, audio and voice to data converters, clocking circuits and
analog front-end subsystems. www.cosmiccircuits.com
Cyclos Semiconductor Inc. (Berkeley, Calif.) was founded in 2006
as a spin-out from the University of Michigan. Based in Berkeley and Ann
Arbor, Cyclos delivers resonant clock mesh semiconductor IP, design
automation tools, and design consulting services for resonant clock mesh
designs that can reduce the power consumption of clock circuits on ICs.
To me (as an employee of a Japanese company of 15 years) the elephant in the room is that there are no Japanese companies on the list; nor can I remember there ever being one; nor could I name a single Japanese startup, despite have visited about a dozen times.
Peter, what about Riviera Waves?
start-up developing WiFi 802.11 a, b, g and selling it as IP. OK, it's a spin-off (Indian design service) but they run in 100% start-up mode... and it look like they could be successful.
They are based in France, is it an issue? (joke)
However, I don't think we should just use startup as an alternative to privately-held or as a euphemism for "not-yet-commercially-successful"
Of course development timescales may differ by industry segment.
Certainly startups are often stealthy for a couple of years before they are even to start their tilt at the market.
There are some circumstances where engineers form a company....go away and do something else.....and only later start to develop their business. So you end up with X years of existence but only Y years of endeavor.
There are no hard and fast rules but i think it is hard to class a ten-year old company as a startup.
Hi Peter: when is a start up no longer a start up? That's a good question as there are no hard or fast rules. As part of the semiconductor start-up alumni I tend to think of accession from the start up ranks is more about financial maturity than time measures... there's an argument to say that all the while a company is being funded (i.e. on the way to being profitable) then it could be classed as a start-up. Let's be realistic here - semiconductor start-ups are an endangered species for a number of reasons - not least because the old fabless model takes (on average) 10 years to get firmly off the ground... we've seen the evolution of new models in the past 6 or 7 years which seek to reduce investment and time scales and ultimately the risk to investment as a response. On the question of Nujira - I still class them as a start up. They've had an interesting journey which flies in the face of the trend. Whilst many investors are encouraging start ups away from the fabless model, in Nujira's case they were actively encouraged to go fabless! Hence they may be xx years old but their foray into the fabless start up land has been much more recent... so I class them firmly as a start up (oh and they've received money fairly recently and investing that in development).
Thanks for maintaining this list over the years - always of great interest.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.