A snip from a VERy GEEKY read....

http://www.nallatech.com/fpga-acceleration-convolutional-neural-networks/

Conclusion
The unique flexibility of FPGA fabric allows the logic precision to be adjusted to the minimum that a particular network design requires. By limiting the bit precision of the CNN –Convolutional Neural Network calculation the number of images that can be processed per second can be significantly increased, improving performance and reducing power.

The non-batching approach of an FPGA implementation allows for object recognition in 9 milliseconds (a single frame period), ideal for situations where low latency is crucial. E.g. object avoidance. This permits images to be categorized at a frame rate greater than 100 Hz.

The intrinsic scalability demonstrated by our FPGA implementation can be utilized to implement complex CNN – Convolutional Neural Networks on increasingly smaller and lower power FPGAs at the expense of some performance. This allows less demanding applications to be implemented on extremely low power FPGA devices, particularly useful for embedded solutions, E.g. Near sensor computing.

Somebody will take that FPGA IP from GLoFo and run various CNN's on it and see what they get.

IF FPGAs can become the solution for on device Inference.....the value of the related IP goes up ( A LOT)

a nice read

http://semimd.com/blog/tag/globalfoundries/

some snips


NXP uses 28nm FD-SOI for its iMX 7 and iMX 8 processors,


Itow said she has talked to more companies that are looking at FD-SOI, and some of them have teams designing products. “So we are seeing more serious activity than before,” Itow said. “I don’t see it being the main Qualcomm process for high-volume products like the applications processors in smartphones. But I do see it being looked at for IoT applications that will come on line in a couple of years. And these things always seem to take longer than you think,” she said.


GlobalFoundries claims it has more than 50 companies in various stages of development on its 22FDX process, which enters risk production early next year, and the company plans a 12nm FDX offering in several years.


That may open up companies “with a lower cost engineering team” in India, China, Taiwan, and elsewhere to “go off in a different direction” and experiment with FD-SOI, Gwennap said.


“If you believe the future is about mobility, about more communications and low power consumption and cost sensitive IoT chips where analog and RF is about 50 percent of the chip, then FD-SOI has a good future.

NXP says...



“NXP's next generation of i.MX multimedia applications processors are leveraging the benefits of FD-SOI to achieve both leadership in power efficiency and scaling performance-on-demand for automotive, industrial and consumer applications,” said Ron Martino, vice president, i.MX applications processor product line at NXP Semiconductors. “GLOBALFOUNDRIES’ 12FDX technology is a great addition to the industry because it provides a next generation node for FD-SOI that will further extend planar device capability to deliver lower risk, wider dynamic range, and compelling cost-performance for smart, connected and secure systems of tomorrow.”

on going to QCOM
will allow us to further enhance our leadership positions, and expand the already strong partnerships with our broad customer base, especially in automotive, consumer and industrial IoT and device level security," said Rick Clemmer, NXP Chief

It goes like this....

< Prev12345

1. 

   jfiebWell-Known Member

   
   
   

   Facebook web page...
   
   https://code.facebook.com/posts/196146247499076/delivering-real-time-ai-in-the-palm-of-your-hand/
   
   
   Delivering real-time AI in the palm of your hand
   We've developed a new deep learning platform on mobile so it can — for the first time — capture, analyze, and process pixels in real time, putting state-of-the-art technology in the palm of your hand. This is a full-fledged deep learning system called Caffe2Go, and the framework is now embedded into our mobile apps. By condensing the size of the AI model used to process images and videos by 100x, we're able to run various deep neural networks with high efficiency on both iOS and Android. Ultimately, we were able to provide AI inference on some mobile phones at less than 1/20th of a second, essentially 50 ms — a human eye blink happens at 1/3rd of a second or 300 ms.
   
   
   . Our Applied Machine Learning group had been working toward building an AI engine that would run on mobile devices. The Camera team had a clear understanding of the user needs. Along with the contribution of many others, these teams produced a best-in-class solution that runs highly optimized neural networks live on mobile devices. We'll explain how we thought about and developed the applicable technologies, starting with Caffe2go.
   
   
   
   Its is a parlor trick for now. Key points to consider...
   
   1. Its Facebook moving into AI and they want it ON THE DEVICE.
   
   2. Regular readers of this thread will see Sensory Inc. in this, ieIT HAS TO BE a Neural Network to get the small footprint. The FIRST BIG DOG to say what Todd M ( Sensory CEO ) has been saying for a Loooong, long, long time.
   INtelligence ON THE DEVICE.
   
   
   IF anyone can build a ubiquitous NNLE ( neural network learning engine ) that can reside on a mobile device -it is Dr Saxe, they have ALL the bits and pieces needed to do so. The latest tidbit of info on the use of FPGA as a very good place to run inference
   is new this yr....someone just has to try it on their benches, I mean running the inference on a FPGA designed for low power so it can be done on the device. With the IP $$, I am happy that Dr Saxe will get to make that Eos S4.
   
   I have written in my journal it goes like this....
   
   FPGA= Inference ( ?)
   
   Facebook wants inference on the mobile device.
   
   The IP around an FPGA purposed for the mobile device just moved up substantially in value?
   
   Does anyone see a hole in this line of thinking?
   
   Thanks in advance.

this helped me the most...
 Changes to Executive Comp

---

jfieb, if I have this whole thing down correctly, Quik is opening up their locked store room that has Quik stock certificate bundles on it. Quik has told each one of the aforementioned executives that only they can stop in on September 8 or later to reach up on that shelf and pull off as many stock certificates as their allotted dollar amount allows. In Brian Faith's example it would be $500,000 worth. And the amount of stock that BF could buy would be determined by the price of Quik on the open market on the purchase date (September 8 or later). I am not sure how all the laws work for insider knowledge/buying but from a market perspective I assume it would be better to buy directly from Quik's store room than to go into the market and try to buy $500,000 worth of stock and keep the price at or below where the strike price was when you began. As to the laws, could this be a safe way to load up on Quik stock while knowing important information that has yet to be disclosed publicly due to NDA's etc.

1. 
   

   Can we partner with these guys just as we have done with Sensory Inc?
   
   Samsung Licenses Valencell’s Ear-Based Biometrics Tech
   Posted on July 20, 2016 by Alex Perala
   “…Valencell has been keen to emphasize the applications of its technology in products such as earbuds, with a recently granted patent pertaining to ‘earbud covers’.”
   
   Samsung has licensed Valencell patents related to the latter’s ear-based biometric wearable technology. The non-exclusive patent licensing deal provides Samsung access to Valencell’s technology concepts, but not to its products and support.
   
   In a statement announcing the deal, Valencell emphasized the importance of licensing and its patent portfolio to its business, noting that it currently has 35 granted patents and over 70 others pending. CEO Dr. Steven LeBoeuf also highlighted the applications of the technology entailed in the licensing agreement, stating,“We’re pleased to work with Samsung Electronics to provide a patent license for our inventions that are critical for accurate heart rate monitoring in hearables.”
   
   By “hearables”, Dr. LeBoeuf uses his company’s terminology for ear-based biometric devices. Blood flow signals detectable through a user’s ears can be used to monitor a number of cardiac measures, and Valencell has been keen to emphasize the applications of its technology in products such as earbuds, with a recently granted patent pertaining to ‘earbud covers’.
   
   With a major consumer electronics company like Samsung now licensing its technology, it could soon reach an even wider swath of end users in the mass market, further popularizing biometrics in consumer electronics.
   
   this snip of text up from the archives to reread....
   
   third-party algorithm folks to deploy their algorithms on our platform and we do see activities in this area
   
   
   For the casual reader. Looking back SPI is a fine example of a third party algo company.
   Biometrics has very specialized algos, and Valencell, like Sensory, has TOP Notch stuff.
   They are a third party algo group of folks.
   

   Last edited: 23 minutes ago

   jfieb, 42 minutes agoEditDeleteReport

   #35Reply
2. 

   jfiebWell-Known Member

   
   

   commentary of the deal...
   
   
   TECHFLASH
   Raleigh wearables upstart scores Samsung deal
   Jul 20, 2016, 2:15pm EDT Updated Jul 22, 2016, 9:49am EDT
   
   Property Spotlight: WestChase I
   See All Bizspace Properties
   
   A small Raleigh wearable technology firm has inked what could be a very big deal with consumer electronics giant Samsung Electronics.
   
   The deal allows Samsung to access Raleigh-based Valencell’s patent portfolio for use in hearable devices, such as the upcoming Samsung Gear IconX cord-free fitness earbuds.
   
   Enlarge
   Valencell CEO Michael Dering, left, and Steven LeBoeuf, president.
   
   TBJ FILE PHOTO
   
   Unlike other Valencell announcements – such as product licensing deals that put its technology in headphones from brands such as LG Electronics and Scosche – this deal is about the intellectual property.
   
   “It’s a really big deal,” Valencell co-founder and president Steve LeBoeuf said Wednesday in an interview. “You can easily argue that Samsung is the most sophisticated consumer electronics brand when it comes to IP.”
   
   He says the fact that they’re licensing patents and not a module designed for a specific product is a testament to Valencell’s innovation. And it’s innovation the firm has been increasingly bullish about protecting, having filed two recent patent infringement lawsuits against electronics mammoths Fitbit and Apple.
   
   Read: Why Valencell's patent fight against Fitbit and Apple could last 'years'
   
   As LeBoeuf has been talking to Samsung for “several years” about a partnership, it’s likely that the deal was in the works before those suits were filed, though LeBoeuf wouldn’t comment on that Wednesday.
   
   Valencell’s patent portfolio, which includes patent protections that give hearable devices the ability to measure blood flow signals that can be translated into biometric data such as heart rate, includes 35 patents granted and more than 70 additional patents pending. The deal with Samsung does not include access to Valencell products, its support personnel or its North Hills wearables testing lab.
   
   Financial terms were not disclosed. LeBoeuf says that, to his knowledge, it's the first patent portfolio licensing announcement in the health and fitness sensor space.
   
   In the coming months, LeBoeuf says to expect additional announcements about the Raleigh technology’s presence in other name-brand electronic devices, including new products coming directly from the sensor-maker.
   
   The Samsung deal is among Valencell’s more high-profile partnerships, but it’s the second collaboration announced this summer. In June, Valencell announced it had partnered with Gieseck & Devrient, a technology provider for the financial sector, and Z-Smart, a wearable device firm in China. The deal puts Valencell’s sensor technology in a new payments-focused smart watch targeting the banking industry.
   
   Valencell execs say the company has more partnerships than what’s listed on its website. Non-disclosure agreements prevent them from publicizing every relationship – and from talking about every product that utilizes its biometric sensors.
   
   
   
   WHen I read about these guys I wanted to buy a part of their biz. Its not possible.
   
   
   In the big pic this is good for sensor hub people as these algos are more complex and would tax an AP and the MCUs> THe Eos is made for stuff like this.
   
   I wonder what the SoC is for the Samsung IconX?
   

   jfieb, 27 minutes agoEditDeleteReport

   #36Reply
3. 

   jfiebWell-Known Member

   
   

   this snip is the one to reread...
   
   QUIK said this a while ago...
   This snip was interesting…verrry interesting?
   
   
   And I will actually give a plug for the fact that we have this nice integrated development environment now that software people are accustomed to. They can actually enablethird-party algorithm folks to deploy their algorithms on our platform and we do see activities in this area. So, it’s a long way of saying that we are pushing in all those areas. The end product may be some combination of us and third-parties to do that, particularly in some of these highly specialized areas.
   
   
   Hope we read about something for Valencel IP and the Eos?
   
   So far any company with something to offer in the adjacent possible of sensor fusion has shown that the coral reef is rich, aware, hungry, .....nice job Valencel!
   
   For the casual reader, note that digressive threads get more interesting as time goes by. It IS a CORAL reef after all. Only 2% of the ocean floor, but 25% of the life resides there.
   
   here is a LOT of good stuff on this thread so if you have time spend a few minutes here..........
   
   If SEC is a matchmaker for hardware and IP it will be something to track along....it is said that these 2 (V & S) have been in talks for a looooong time.
   
   
   SEC should just take them out? As far as patent wars go... the patents of Sensory and of Valencel are of great significance going forward......