BRN Discussion Ongoing

[Fact Finder]

And does anyone remember the recent Edge Impulse presentation where the presenter described AKIDA as science fiction well do you also remember he said that they had hocked up Nvidia Jetson with an Indian client to count traffic at an intersection.

Well the following paper makes clear the Brainchip and Prophesee would have been a far better choice:

Neuromorphic Vision Based Multivehicle Detection and Tracking for Intelligent Transportation System

Neuromorphic vision sensor is a new passive sensing modality and a frameless sensor with a number of advantages over traditional cameras. Instead of wastefully sending entire images at fixed frame rate, neuromorphic vision sensor only transmits the local pixel-level changes caused by the...

www.hindawi.com

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Deadpool]

I do not recall this being posted before and the date of release suggests not so guess which space program is using a COTS anomaly detection SNN on space missions:

Small Business Innovation Research/Small Business Tech Transfer
Neuromorphic Spacecraft Fault Monitor, Phase II
Completed Technology Project (2020 - 2022)
Project Introduction
The goal of this work is to develop a low power machine learning anomaly detector. The low power comes from the type of machine learning (Spiking Neural Network (SNN)) and the hardware the neuromorphic anomaly
detector runs on. The ability to detect and react to anomalies in sensor readings on board resource constrained spacecraft is essential, now more than ever, as enormous satellite constellations are launched and humans push out again beyond low Earth orbit to the Moon and beyond. Spacecraft are autonomous systems operating in dynamic environments. When monitored parameters exceed limits or watchdog timers are not reset, spacecraft can automatically enter a 'safe' mode where primary functionality is reduced or stopped completely. During safe mode the primary mission is put on hold while teams on the ground examine dozens to hundreds of parameters and compare them to archived historical data and the spacecraft design to determine the root cause and what corrective action to take. This is a difficult and time consuming task for humans, but can be accomplished faster, in real- time, by machine learning. As humans travel away from Earth, light travel time delays increase, lengthening the time it takes for ground crews to respond to a safe mode event. The few astronauts onboard will have a hard time replacing the brain power and experience of a team of experts on the ground. Therefore, a new approach is needed that augments existing capabilities to help the astronauts in key decision moments. We provide a new machine learning approach that recognizes nominal and faulty behavior, by learning during integration, test, and on-orbit checkout. This knowledge is stored and used for anomaly detection in a low power neuromorphic chip and continuously updated through regular operations. Anomalies are detected and context is provided in real-time, enabling both astronauts onboard, and ground crews on Earth, to take action and avoid potential faults or safe mode events.
Anticipated Benefits
The software developed in Phase II can potentially be used by NASA for anomaly detection onboard the ISS, the planned Lunar Gateway, and future missions to Mars. The NSFM software can also be used by ground crews to augment their ability to monitor spacecraft and astronaut health telemetry once it reaches the ground. The NSFM software can furthermore be used during integration and test to better inform test operators of the functionality of the system during tests in real time.
The software developed in Phase II can potentially be used for anomaly detection onboard any of the new large constellations planned by private companies. It can also be applied to crewed space missions, deep space probes, UUVs, UAVs, and many industrial applications on Earth. The NSFM software developed in Phase II can also be used during Integration and Test of any commercial satellite.

NASA TechPort

NASA's Technology Portfolio Management System (TechPort) is a single, comprehensive resource for locating detailed information about NASA-funded technologies. Those technologies cover a broad range of areas, such as propulsion, nanotechnology, robotics, and human health. You can find useful...

techport.nasa.gov

My opinion only DYOR
FF

AKIDA BALLISTA

Hey @factfinder, reminds me of Arther C Clark's 1968, 2001 A Space Odessey, who wrote about this very proposal
HAL 9000 Heuristically programmed ALgorithmic computer.

Science fiction becoming actuality

 

[Fact Finder]

This is part of what Carnegie Mellon submitted to the:

The White House Office of Science and Technology Policy – on behalf of the National Science and Technology Council's Select Committee on Artificial Intelligence and Machine Learning and AI Subcommittee, the National AI Initiative Office, and the Networking and Information Technology Research and Development National Coordination Office – released a Request for Information (RFI) on February 2, 2022, to request input on updating the National Artificial Intelligence Research and Development Strategic Plan. The RFI was published in the Federal Register and the comment period was open from February 2, 2022, through March 4, 2022.
This document contains the 63 responses received from interested parties. In accordance with the RFI instructions, only the first 10 pages of content were considered for each response

“Recommendations of AI Research Focus Areas to Create Solutions to Major Societal Challenges
The National AI Research and Development Strategic plan can catalyze innovations in both fundamental discoveries and applications that address specific societal challenges. Progress towards realizing this potential can be realized by collaborative efforts in the following areas.
77

Foster Interagency Collaboration to Ensure America Leads in Enabling Distributed Artificial Intelligence
The U.S. should lead a bold transformative agenda over the next five years to enable AI to evolve from highly structured and controlled, centralized architectures to more adaptive and pervasively distributed ones that autonomously fuse AI capability among the enterprise, the edge, and across AI systems and sensors embedded on-platform. CMU terms this revolutionary architectural advance as AI Fusion. The vision is built upon plans for a cohesive research advancing capabilities in microelectronics, AI frameworks and algorithms and innovations in federated learning in the AI fabric and abstraction layers.
Building a community research roadmap for distributed AI will address several critical challenges for the growth of AI, challenges that cut across agency-specific missions. The ability to enable distributed AI at the edge will minimize the dependence on aggregating and engineering massive data sets and reduce the need to “move the data to the algorithms” as well as the inherent challenges associated with the need for continuous high-bandwidth connectivity. Research in this area will also greatly enhance the capacity to address privacy and security challenges. It is dependent on, will contribute to and will benefit from the national computing infrastructure initiatives launched by the NAIIO.
Most critically, an AI Fusion research agenda will contribute to the network of AI institutes by enabling a host of applications emerging from increased convergence across AI-enabled cyber and physical systems. This convergence is vital to the viability of applications in commercial, military and national security domains. AI Fusion, for example, will be a critical contribution to the Department of Defense’s (DOD) focus on Multi-Domain Operations. It will also enhance the potential for advances in smart city applications and AI breakthroughs aiding manufacturing, energy, health care, education and agricultural innovations. A focus on AI Fusion should operate synergistically with national initiatives in microelectronics and tie directly with research and innovation efforts aimed at enhancing, protecting and hardening critical U.S. supply chains.
Initiate Research to Engineer AI into Societal Systems
While fundamental advances are needed in AI science, advances in engineering AI into systems of societal importance are vital to realize the full impact on major national missions. Engineering AI into such systems will be essential to transform U.S. manufacturing and enhance infrastructure and energy systems to meet critical national economic and societal goals.
Engineering AI will require the design, development and deployment of new use-inspired AI algorithms and methodologies, targeted to real-world applications and possessing enhanced scalability, robustness, fairness, security, privacy and policy impact. Advancing Engineering AI will also require new hardware and software systems, including cloud, edge and device computing infrastructures that sense and store the vast amounts of data collected in the real world and that enable devices to access and transmit this data from anywhere, to anywhere, in secure and private ways. Foundational research for Engineering AI is needed to enable the deployment of the highest performing and most energy-efficient AI systems. Such systems will
78

require architecting new hardware and computing frameworks; designing faster, more powerful and efficient integrated circuits; and developing sensing modalities to support data collection, storage and processing of the data deluge.
In addition, Carnegie Mellon recognizes that research on Engineering AI must include a focus on creating trust not only from a technical standpoint but from the system of stakeholders interacting with the AI system — be it in education, infrastructure or climate. Users and communities have to trust the system that is allocating resources and making decisions.
Potential applications and use cases for Engineering AI include autonomous infrastructure systems (AIS) that can help create equitable, innovative and economically sustainable communities. AIS technology could, for example, include initiatives integrating food delivery, the tracking of goods while preserving privacy and tools to improve mobility. Engineering AI will be key to the digital transformation of manufacturing in the U.S., including robotics for manufacturing, development of a timely and trustworthy supply chain and additive manufacturing. Engineering AI also has the potential to revolutionize how electricity is produced, distributed and consumed. It can provide insights to improve electricity distribution through demand forecasting, load management and community governance, as well as to innovate new energy storage solutions, control pollutants and advance wind, solar and nuclear energies.”

https://www.ai.gov/rfi/2022/87-FR-5876/NAIRDSP-RFI-2022-combined.pdf

If you have read the above you too may think that there is a lot more to the relationship between Brainchip and Carnegie Mellon than we first thought.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[dippY22]

Was browsing the Computer Science area at Carnegie Mellon U website. Not sure I've seen this posted before, ....doesn't mean it hasn't, I just don't recall.

The class agenda in one of the Computer Science / Machine Learning courses ....

 

[Dallas]

 

[Dallas]

Chris Stevens, Industry Veteran Joins BrainChip to Lead Worldwide Sales

LAGUNA HILLS, CA / ACCESSWIRE / September 6, 2022 / BrainChip Holdings Ltd (ASX:BRN)(OTCQX:BRCHF)(ADR:BCHPY), the world's first commercial producer of neuromorphic AI IP, announced that Chris Stevens has joined as Vice President of Worldwide Sales to lead and expand BrainChip sales. Mr. Stevens...

www.accesswire.com

 

[Food4 thought]

This is part of what Carnegie Mellon submitted to the:

The White House Office of Science and Technology Policy – on behalf of the National Science and Technology Council's Select Committee on Artificial Intelligence and Machine Learning and AI Subcommittee, the National AI Initiative Office, and the Networking and Information Technology Research and Development National Coordination Office – released a Request for Information (RFI) on February 2, 2022, to request input on updating the National Artificial Intelligence Research and Development Strategic Plan. The RFI was published in the Federal Register and the comment period was open from February 2, 2022, through March 4, 2022.
This document contains the 63 responses received from interested parties. In accordance with the RFI instructions, only the first 10 pages of content were considered for each response

“Recommendations of AI Research Focus Areas to Create Solutions to Major Societal Challenges
The National AI Research and Development Strategic plan can catalyze innovations in both fundamental discoveries and applications that address specific societal challenges. Progress towards realizing this potential can be realized by collaborative efforts in the following areas.
77

Foster Interagency Collaboration to Ensure America Leads in Enabling Distributed Artificial Intelligence
The U.S. should lead a bold transformative agenda over the next five years to enable AI to evolve from highly structured and controlled, centralized architectures to more adaptive and pervasively distributed ones that autonomously fuse AI capability among the enterprise, the edge, and across AI systems and sensors embedded on-platform. CMU terms this revolutionary architectural advance as AI Fusion. The vision is built upon plans for a cohesive research advancing capabilities in microelectronics, AI frameworks and algorithms and innovations in federated learning in the AI fabric and abstraction layers.
Building a community research roadmap for distributed AI will address several critical challenges for the growth of AI, challenges that cut across agency-specific missions. The ability to enable distributed AI at the edge will minimize the dependence on aggregating and engineering massive data sets and reduce the need to “move the data to the algorithms” as well as the inherent challenges associated with the need for continuous high-bandwidth connectivity. Research in this area will also greatly enhance the capacity to address privacy and security challenges. It is dependent on, will contribute to and will benefit from the national computing infrastructure initiatives launched by the NAIIO.
Most critically, an AI Fusion research agenda will contribute to the network of AI institutes by enabling a host of applications emerging from increased convergence across AI-enabled cyber and physical systems. This convergence is vital to the viability of applications in commercial, military and national security domains. AI Fusion, for example, will be a critical contribution to the Department of Defense’s (DOD) focus on Multi-Domain Operations. It will also enhance the potential for advances in smart city applications and AI breakthroughs aiding manufacturing, energy, health care, education and agricultural innovations. A focus on AI Fusion should operate synergistically with national initiatives in microelectronics and tie directly with research and innovation efforts aimed at enhancing, protecting and hardening critical U.S. supply chains.
Initiate Research to Engineer AI into Societal Systems
While fundamental advances are needed in AI science, advances in engineering AI into systems of societal importance are vital to realize the full impact on major national missions. Engineering AI into such systems will be essential to transform U.S. manufacturing and enhance infrastructure and energy systems to meet critical national economic and societal goals.
Engineering AI will require the design, development and deployment of new use-inspired AI algorithms and methodologies, targeted to real-world applications and possessing enhanced scalability, robustness, fairness, security, privacy and policy impact. Advancing Engineering AI will also require new hardware and software systems, including cloud, edge and device computing infrastructures that sense and store the vast amounts of data collected in the real world and that enable devices to access and transmit this data from anywhere, to anywhere, in secure and private ways. Foundational research for Engineering AI is needed to enable the deployment of the highest performing and most energy-efficient AI systems. Such systems will
78

require architecting new hardware and computing frameworks; designing faster, more powerful and efficient integrated circuits; and developing sensing modalities to support data collection, storage and processing of the data deluge.
In addition, Carnegie Mellon recognizes that research on Engineering AI must include a focus on creating trust not only from a technical standpoint but from the system of stakeholders interacting with the AI system — be it in education, infrastructure or climate. Users and communities have to trust the system that is allocating resources and making decisions.
Potential applications and use cases for Engineering AI include autonomous infrastructure systems (AIS) that can help create equitable, innovative and economically sustainable communities. AIS technology could, for example, include initiatives integrating food delivery, the tracking of goods while preserving privacy and tools to improve mobility. Engineering AI will be key to the digital transformation of manufacturing in the U.S., including robotics for manufacturing, development of a timely and trustworthy supply chain and additive manufacturing. Engineering AI also has the potential to revolutionize how electricity is produced, distributed and consumed. It can provide insights to improve electricity distribution through demand forecasting, load management and community governance, as well as to innovate new energy storage solutions, control pollutants and advance wind, solar and nuclear energies.”

https://www.ai.gov/rfi/2022/87-FR-5876/NAIRDSP-RFI-2022-combined.pdf

If you have read the above you too may think that there is a lot more to the relationship between Brainchip and Carnegie Mellon than we first thought.

My opinion only DYOR
FF

AKIDA BALLISTA

This journey we are on is becoming very very positive, exciting,thrilling and down right amazing ( did someone say KFC)

 

[thelittleshort]

Samsung only just started producing 3nm in June this year and one report I read, said they were having problems producing them profitably..
Not saying they won't get there, I wouldn't have a clue, but they're going to be expensive chips, you would think..

Synopsys, who I think Prophesee ditched for Brainchip have partnered with Samsung on the 3nm process for their chips..

Samsung 3nm GAA Yield Rate at an Abysmal 20%, Aims to Improve Output Through Cooperation With U.S.-Based Firm

Samsung’s attempt to mass produce 3nm GAA chips is going horribly and it aims to improve output by partnering with a U.S. company

wccftech-com.cdn.ampproject.org

 

[Learning]

This is part of what Carnegie Mellon submitted to the:

The White House Office of Science and Technology Policy – on behalf of the National Science and Technology Council's Select Committee on Artificial Intelligence and Machine Learning and AI Subcommittee, the National AI Initiative Office, and the Networking and Information Technology Research and Development National Coordination Office – released a Request for Information (RFI) on February 2, 2022, to request input on updating the National Artificial Intelligence Research and Development Strategic Plan. The RFI was published in the Federal Register and the comment period was open from February 2, 2022, through March 4, 2022.
This document contains the 63 responses received from interested parties. In accordance with the RFI instructions, only the first 10 pages of content were considered for each response

“Recommendations of AI Research Focus Areas to Create Solutions to Major Societal Challenges
The National AI Research and Development Strategic plan can catalyze innovations in both fundamental discoveries and applications that address specific societal challenges. Progress towards realizing this potential can be realized by collaborative efforts in the following areas.
77

Foster Interagency Collaboration to Ensure America Leads in Enabling Distributed Artificial Intelligence
The U.S. should lead a bold transformative agenda over the next five years to enable AI to evolve from highly structured and controlled, centralized architectures to more adaptive and pervasively distributed ones that autonomously fuse AI capability among the enterprise, the edge, and across AI systems and sensors embedded on-platform. CMU terms this revolutionary architectural advance as AI Fusion. The vision is built upon plans for a cohesive research advancing capabilities in microelectronics, AI frameworks and algorithms and innovations in federated learning in the AI fabric and abstraction layers.
Building a community research roadmap for distributed AI will address several critical challenges for the growth of AI, challenges that cut across agency-specific missions. The ability to enable distributed AI at the edge will minimize the dependence on aggregating and engineering massive data sets and reduce the need to “move the data to the algorithms” as well as the inherent challenges associated with the need for continuous high-bandwidth connectivity. Research in this area will also greatly enhance the capacity to address privacy and security challenges. It is dependent on, will contribute to and will benefit from the national computing infrastructure initiatives launched by the NAIIO.
Most critically, an AI Fusion research agenda will contribute to the network of AI institutes by enabling a host of applications emerging from increased convergence across AI-enabled cyber and physical systems. This convergence is vital to the viability of applications in commercial, military and national security domains. AI Fusion, for example, will be a critical contribution to the Department of Defense’s (DOD) focus on Multi-Domain Operations. It will also enhance the potential for advances in smart city applications and AI breakthroughs aiding manufacturing, energy, health care, education and agricultural innovations. A focus on AI Fusion should operate synergistically with national initiatives in microelectronics and tie directly with research and innovation efforts aimed at enhancing, protecting and hardening critical U.S. supply chains.
Initiate Research to Engineer AI into Societal Systems
While fundamental advances are needed in AI science, advances in engineering AI into systems of societal importance are vital to realize the full impact on major national missions. Engineering AI into such systems will be essential to transform U.S. manufacturing and enhance infrastructure and energy systems to meet critical national economic and societal goals.
Engineering AI will require the design, development and deployment of new use-inspired AI algorithms and methodologies, targeted to real-world applications and possessing enhanced scalability, robustness, fairness, security, privacy and policy impact. Advancing Engineering AI will also require new hardware and software systems, including cloud, edge and device computing infrastructures that sense and store the vast amounts of data collected in the real world and that enable devices to access and transmit this data from anywhere, to anywhere, in secure and private ways. Foundational research for Engineering AI is needed to enable the deployment of the highest performing and most energy-efficient AI systems. Such systems will
78

require architecting new hardware and computing frameworks; designing faster, more powerful and efficient integrated circuits; and developing sensing modalities to support data collection, storage and processing of the data deluge.
In addition, Carnegie Mellon recognizes that research on Engineering AI must include a focus on creating trust not only from a technical standpoint but from the system of stakeholders interacting with the AI system — be it in education, infrastructure or climate. Users and communities have to trust the system that is allocating resources and making decisions.
Potential applications and use cases for Engineering AI include autonomous infrastructure systems (AIS) that can help create equitable, innovative and economically sustainable communities. AIS technology could, for example, include initiatives integrating food delivery, the tracking of goods while preserving privacy and tools to improve mobility. Engineering AI will be key to the digital transformation of manufacturing in the U.S., including robotics for manufacturing, development of a timely and trustworthy supply chain and additive manufacturing. Engineering AI also has the potential to revolutionize how electricity is produced, distributed and consumed. It can provide insights to improve electricity distribution through demand forecasting, load management and community governance, as well as to innovate new energy storage solutions, control pollutants and advance wind, solar and nuclear energies.”

https://www.ai.gov/rfi/2022/87-FR-5876/NAIRDSP-RFI-2022-combined.pdf

If you have read the above you too may think that there is a lot more to the relationship between Brainchip and Carnegie Mellon than we first thought.

My opinion only DYOR
FF

AKIDA BALLISTA

Hi FF,

Just by reading the extract, I believe Brainchip's Akida neuromorphic architecture will be part of the Carnegie Mellon presentation to the US 'National Artificial Intelligence Research and Development Strategic Plan'

Learning.

 

[Fact Finder]

View attachment 22974

Samsung 3nm GAA Yield Rate at an Abysmal 20%, Aims to Improve Output Through Cooperation With U.S.-Based Firm

Samsung’s attempt to mass produce 3nm GAA chips is going horribly and it aims to improve output by partnering with a U.S. company

wccftech-com.cdn.ampproject.org

At severe risk of being accused of being a no-it-all (smart arse)as I have said it for two years now but remember this is an investment and so it’s important to firmly understand the significance of what Brainchip and AKIDA technology is bringing to the table so I have said it once now I will say it 1,000 times:

1. Samsung, Intel, Nvidia, ARM, Bosch, Google, APPLE in other words the entire semi conductor world are chasing smaller semiconductors because Von Neumann compute has reached its limits and the only way to increase performance and reduce power is to go smaller and smaller,

2. Two things happen every single time they go smaller cost goes up exponentially and yield goes down,

3. AKIDA at 28nm still wins on power and performance even when the rest are at 3nm,

4. AKIDA at 28nm is cheap to produce,

5. AKIDA at 28nm at TSMC has a yield rate exceeding 90%,

6. AKIDA at 28nm can be produced even by Intel,

7. AKIDA at 28nm crushes Loihi 2 at 7nm,

8. AKIDA can stay at 28nm and use carbon nanotube connections and up performance without going down in size,

9. AKIDA can give NASA what it needs at 90nm making it even more radiation resilient but imagine what happens to all the other tech like Snapdragon chasing smaller and smaller form down to 3nm if you say that’s great we will have it in 90nm thanks.

AKIDA is without doubt the stuff of Science Fiction.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Learning]

Was browsing the Computer Science area at Carnegie Mellon U website. Not sure I've seen this posted before, ....doesn't mean it hasn't, I just don't recall.

The class agenda in one of the Computer Science / Machine Learning courses ....

Hi dippyY22,

From that class agenda, Carnegie Mellon University had been teaching Brainchip's Akida neuromorphic since start of 2022. WOW.

More future engineers, more use case for Brainchip's Akida into the future...

Learning.

 

[Dang Son]

A great Jereme Nadel interview from Alex the Rocket Scientist, 4mths old , sorry if already seen.

 

[Dang Son]

Not a bright future for Intel as presented by Alex ( the Rock Scientist)

 

[Dr E Brown]

What I like about our two latest recruits are that we have not taken them out of retirement but that they are proven operators, who have chosen to leave companies either competing or working with Brainchip in Syntiant and Amazon. I actually find that quite remarkable given the current status of Brainchip’s commercialisation. It provides me, as an investor, with enormous confidence and a nice warm fuzzy feeling!

 

[Diogenese]

At severe risk of being accused of being a no-it-all (smart arse)as I have said it for two years now but remember this is an investment and so it’s important to firmly understand the significance of what Brainchip and AKIDA technology is bringing to the table so I have said it once now I will say it 1,000 times:

1. Samsung, Intel, Nvidia, ARM, Bosch, Google, APPLE in other words the entire semi conductor world are chasing smaller semiconductors because Von Neumann compute has reached its limits and the only way to increase performance and reduce power is to go smaller and smaller,

2. Two things happen every single time they go smaller cost goes up exponentially and yield goes down,

3. AKIDA at 28nm still wins on power and performance even when the rest are at 3nm,

4. AKIDA at 28nm is cheap to produce,

5. AKIDA at 28nm at TSMC has a yield rate exceeding 90%,

6. AKIDA at 28nm can be produced even by Intel,

7. AKIDA at 28nm crushes Loihi 2 at 7nm,

8. AKIDA can stay at 28nm and use carbon nanotube connections and up performance without going down in size,

9. AKIDA can give NASA what it needs at 90nm making it even more radiation resilient but imagine what happens to all the other tech like Snapdragon chasing smaller and smaller form down to 3nm if you say that’s great we will have it in 90nm thanks.

AKIDA is without doubt the stuff of Science Fiction.

My opinion only DYOR
FF

AKIDA BALLISTA

A few chip makers have adopted an alternative technique to reduce the footprint of ICs: 3D stacking as used in the Sony/Prophesee DVS chip where the pixel layer is "laminated" on top of the processing layer.

One problem with 3D IC stacking is that it makes dissipation of heat more difficult, so Akida is a natural fit.

 

[Rise from the ashes]

CCTV Camera Market Is Poised To Reach Approximately USD 31,190 Million, Registering a 13.2% CAGR by 2030 - Report by Market Research Future (MRFR)

Use for Surveillance & Monitoring Activities Drives CCTV Market Demand ...

www.globenewswire.com

That's a market I believe we will dominate in the coming years imo.
Sometimes I think of stuff that Akida can do and how some future products will incorporate those abilities that Akida has brought to the world(they don't know it yet)
For example bank CCTV could be trained to send and alert/alarm if it picks up someone wearing a Balaklava outside/inside or someone carrying a firearm or home use: training to recognise certain sounds like a gate opening or trained to send an alert when a particular person or person's have been detected. Trained to spot fires before they get out of hand. Training to recognise sound of glass smashing or the sound of a gun going off or trained to alert when certain voice signatures are detected. I could go on and on.
The amount of useless alerts that get sent these days is ridiculous.
All of the above and much much more at low power at the far edge.

 

[thelittleshort]

At severe risk of being accused of being a no-it-all (smart arse)as I have said it for two years now but remember this is an investment and so it’s important to firmly understand the significance of what Brainchip and AKIDA technology is bringing to the table so I have said it once now I will say it 1,000 times:

I love it when you talk dirty

 

[Fact Finder]

A few chip makers such have adopted an alternative technique to reduce the footprint of ICs: 3D stacking as used in the Sony/Prophesee DVS chip where the pixel layer is "laminated" on top of the processing layer.

One problem with 3D IC stacking is that it makes dissipation of heat more difficult, so Akida is a natural fit.

Here is a short paper on this very point. It should not take much argument to convince a Sony or a Panasonic or some other company that AKIDA is a far superior option:

https://www.researchgate.net/profile/K-Iniewski/publication/224060102_Yield_and_Cost_Modeling_for_3D_Chip_Stack_Technologies/links/0fcfd510895577485a000000/Yield-and-Cost-Modeling-for-3D-Chip-Stack-Technologies.pdf

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Diogenese]

Here are a couple of Sony 3D IC camera patent applications:

This one has a pixel array layer (204) and a logic layer (202).

WO2021117642A1 EBS/TOF/RGB CAMERA FOR SMART SURVEILLANCE AND INTRUDER DETECTION

Priority: 20191213

A sensor system with a plurality of sensors or sensor functions is provided. The sensors can include an event detection sensor, a time of flight sensor, and imaging sensor. The different sensors can be implemented on the same or different substrates. Accordingly, sensors with pixels having different or shared functions can be included in the sensor system. In operation, an event detection signal from an event detection sensor causes the operation of a time of flight sensor to be initiated. In response to the detection of an object within a critical range by the time of flight sensor, the imaging sensor is activated. The image sensing and event detection pixels can be provided as part of different arrays of pixels, or can be included within a common array of pixels.

Fig. 2 is a view illustrating a lamination structure example of an image sensor 200 in accordance with at least some embodiments of the present disclosure. As illustrated in Fig. 2, the image sensor 200 can have a structure in which a light-receiving chip 201 and a logic chip 202 are vertically laminated. A side of the light receiving chip 201 opposite the logic chip 202 is a light receiving surface 204. In joining of the light-receiving chip 201 and the logic chip 202, for example, so-called direct joining in which joining surfaces of the chips are planarized, and the chips are laminated with an inter-electron force can be used. However, there is no limitation thereto, and for example, so-called Cu-Cu joining in which copper (Cu) electrode pads formed on joining surfaces are bonded, bump joining, and the like can also be used.


This one has an image layer and 2 logic layers:

WO2022044553A1 SEMICONDUCTOR DEVICE AND METHOD FOR MANUFACTURING SEMICONDUCTOR DEVICE

Priority: 20200831

Provided is a semiconductor device in which a plurality of semiconductor chips 160, 140, 130) are stacked, the semiconductor device having enhanced functionality while reducing manufacturing cost.　The semiconductor chips include a light-receiving chip (130), a redistribution wiring-side semiconductor chip (160), an intermediate semiconductor chip (140), a through-electrode (171), and redistribution wiring (141). The light-receiving chip receives incoming light. A wiring layer is formed on a predetermined wiring surface of the redistribution wiring-side semiconductor chip. One of a pair of joint surfaces of the intermediate semiconductor chip is joined to the light-receiving chip, and the other is joined to the redistribution wiring-side semiconductor chip. The through-electrode extends through a semiconductor substrate of the intermediate semiconductor chip. The redistribution wiring is wired on a wiring surface and connects the through-electrode and the wiring layer.

[150 = resin bonding layer]

One interesting thing about the 3-layer IC is that the optical layer (130) and the first logic layer (140) could be Prophesee pixel array and Prophesee event detection logic.

I wonder what the third semiconductor layer (160) could be used for?

Note that the 3-layer IC originated 8 months after the 2-layer IC.

 

[Fact Finder]

Here are a couple of Sony 3D camera patent applications:

This one has a pixel array layer (204) and a logic layer (202).

WO2021117642A1 EBS/TOF/RGB CAMERA FOR SMART SURVEILLANCE AND INTRUDER DETECTION

Priority: 20191213


View attachment 22982


A sensor system with a plurality of sensors or sensor functions is provided. The sensors can include an event detection sensor, a time of flight sensor, and imaging sensor. The different sensors can be implemented on the same or different substrates. Accordingly, sensors with pixels having different or shared functions can be included in the sensor system. In operation, an event detection signal from an event detection sensor causes the operation of a time of flight sensor to be initiated. In response to the detection of an object within a critical range by the time of flight sensor, the imaging sensor is activated. The image sensing and event detection pixels can be provided as part of different arrays of pixels, or can be included within a common array of pixels.

Fig. 2 is a view illustrating a lamination structure example of an image sensor 200 in accordance with at least some embodiments of the present disclosure. As illustrated in Fig. 2, the image sensor 200 can have a structure in which a light-receiving chip 201 and a logic chip 202 are vertically laminated. A side of the light receiving chip 201 opposite the logic chip 202 is a light receiving surface 204. In joining of the light-receiving chip 201 and the logic chip 202, for example, so-called direct joining in which joining surfaces of the chips are planarized, and the chips are laminated with an inter-electron force can be used. However, there is no limitation thereto, and for example, so-called Cu-Cu joining in which copper (Cu) electrode pads formed on joining surfaces are bonded, bump joining, and the like can also be used.


This one has an image layer and 2 logic layers:

WO2022044553A1 SEMICONDUCTOR DEVICE AND METHOD FOR MANUFACTURING SEMICONDUCTOR DEVICE

Priority: 20200831
View attachment 22981

Provided is a semiconductor device in which a plurality of semiconductor chips 160, 140, 130) are stacked, the semiconductor device having enhanced functionality while reducing manufacturing cost.　The semiconductor chips include a light-receiving chip (130), a redistribution wiring-side semiconductor chip (160), an intermediate semiconductor chip (140), a through-electrode (171), and redistribution wiring (141). The light-receiving chip receives incoming light. A wiring layer is formed on a predetermined wiring surface of the redistribution wiring-side semiconductor chip. One of a pair of joint surfaces of the intermediate semiconductor chip is joined to the light-receiving chip, and the other is joined to the redistribution wiring-side semiconductor chip. The through-electrode extends through a semiconductor substrate of the intermediate semiconductor chip. The redistribution wiring is wired on a wiring surface and connects the through-electrode and the wiring layer.

[150 = resin bonding layer]

One interesting thing about the 3-layer IC is that the optical layer (130) and the first logic layer (140) could be Prophesee pixel array and Prophesee event detection logic.

I wonder what the third semiconductor layer (160) could be used for?

Note that the 3-layer IC originated 8 months after the 2-layer IC.

Hi @Diogenese
Just my lay opinion of course based on the publicly available podcast with Luca Verre CEO of Prophesee who stated that until AKIDA every other method of processing they had tried led to compromise in the performance of their event based sensor that they had spent more than a decade developing will at some point convince a company like Sony to insert AKIDA technology at (160) or Sony will be at risk that Prophesee will convince Canon or Google or Apple or Samsung or Nokia. Even Uncle Elon if he does decide to have his very own ePhone would be a stand out to take up the best that Prophesee can be.

Actually Blackberry with its deep tentacles into the military would probably be a unique use target market for the best Prophesee can be sensor AKIDA combo. The significance of every soldier in the field being able to detect events occurring on the battlefield regardless of lighting conditions would clearly be invaluable. In 2019 there were approximately 27.5 million soldiers in the World.

My opinion only DYOR
FF

AKIDA BALLISTA