Wags
Regular
There's that bloody 1% again. @Fact Finder get with it man, it will be way more than that, just buckle up and hang on. Love your work.
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Wags
Regular
Bravo, of course AI stresses him out. Maybe, just maybe, he knows deep down he needs AKIDA. Love your work too !I've just been thinking a bit more about Elon's "AI stresses me out" comment, which he apparently made near the end of a more than three-hour presentation to Tesla investors about company plans. In the article (article 1) below it says the following:
View attachment 31010
However there was another article that I posted earlier today which describes how Tesla inadvertently admitted its so-called Full Self Driving technology is not the world’s most advanced driver assistance system. It stated the following:
View attachment 31011
I thought it was pretty contradictory to on the one hand to acknowledge that other rival car-makers are making greater advancements with their AI technology and then on the other hand claim that "AI isn't helping us make cars any time soon". But that's just my two bobs worth.
Article 1
Article 2
Tesla admits its semi-autonomous driving tech is not the world’s most advanced
Tesla has admitted its controversial ‘Full Self-Driving’ autonomous driving technology is less capable than systems developed by its rival car-makers.www.drive.com.au
Tothemoon24
Top 20
Only at matter of time
LinkedIn; 2nd February 10.30 pm
Marcus Schafer : Mercedes
In addition to the electrification of our vehicles, automated driving is also a crucial future of mobility topic for us Mercedes-Benz AG.
I am proud that we are already at the forefront of automated driving with our Level 3 system DRIVE PILOT and our Level 4 system INTELLIGENT PARK PILOT! The current DRIVE PILOT system, which is certified for speeds of up to 60 km/h on German Autobahn, relies on the LiDAR sensor from our friends at #Valeo as the third sensor-modality.
Looking into the future, we are also happy to use the next-generation LiDAR by Luminar, which is why we invited Austin Russell, CEO of Luminar Technologies, along to discuss our partnership.
He introduced us to their new Iris sensor, which is designed to further improve automated driving systems. Iris is a first-of-its-kind, automotive-grade solution with an extremely long range. And, as it is also designed to be scalable across our #MercedesBenz vehicle fleet, I’m already looking forward to taking our cooperation to the next level.
Before that happens, we will be attending a SXSW 2023 session together, which is scheduled for 15 March at 11:30 am. The exciting subject: “Autonomous Driving: More Time to Do What You Love”. If you are on site, be sure to come by!
LinkedIn; 2nd February 10.30 pm
Marcus Schafer : Mercedes
In addition to the electrification of our vehicles, automated driving is also a crucial future of mobility topic for us Mercedes-Benz AG.
I am proud that we are already at the forefront of automated driving with our Level 3 system DRIVE PILOT and our Level 4 system INTELLIGENT PARK PILOT! The current DRIVE PILOT system, which is certified for speeds of up to 60 km/h on German Autobahn, relies on the LiDAR sensor from our friends at #Valeo as the third sensor-modality.
Looking into the future, we are also happy to use the next-generation LiDAR by Luminar, which is why we invited Austin Russell, CEO of Luminar Technologies, along to discuss our partnership.
He introduced us to their new Iris sensor, which is designed to further improve automated driving systems. Iris is a first-of-its-kind, automotive-grade solution with an extremely long range. And, as it is also designed to be scalable across our #MercedesBenz vehicle fleet, I’m already looking forward to taking our cooperation to the next level.
Before that happens, we will be attending a SXSW 2023 session together, which is scheduled for 15 March at 11:30 am. The exciting subject: “Autonomous Driving: More Time to Do What You Love”. If you are on site, be sure to come by!
Attachments
Last edited:
Great find TTM,
Drive Pilot is "current".
Valeo SCALA 3 is in Drive Pilot!?
Luminar LiDaR uses "foveated" LidaR which focuses on point(s) of interest (denser laser point distribution) and relegates less interesting data to "peripheral vision".
Will Mercedes mandate the use of Akida with Luminar?
Tothemoon24
Top 20
Exciting times Dio ,
Fingers crossed
the mandate of Akida is placed on the table , it’s certainly a big chance one would think . This was put to air 1 hr ago , I’ve added photos to my original post
Learning
Learning to the Top 🕵♂️
Could this be an important date?
100 years celebratetion.
They will talk about there 'past, present and the future'.
Neuromorphic computing is the future, maybe Valeo can talk about it. (JMHO)
Learning 🏖
Fullmoonfever
Top 20
Just on Socionext and probs posted prev or not, but anyway, this site (translated) has some pretty pics that I hadn't seen.I pretty sure SOCIONEXT announced they have a few products with akida IP in them (the CES 2023 announcement below)
Book a Meeting
Brainchip invites you to join us at CES 2023 January 5-8, 2023, in Las Vegas, NV to experience our advanced, hyper-efficient, event-based AI technology in action, bringing the future to edge AI devices.
Akida’s unique AI approach is built on applied neuromorphic principles that enables inference and learning, untethered from the cloud in energy-constrained edge devices. It efficiently accelerates today’s models and is ready for future trends. Most importantly, it supports frameworks and flows that developers are very familiar with, so tuning for Akida platforms is seamless and Brainchip partners like Renesas are integrating Akida technology into their platforms.
At CES a few of our partners are demonstrating this technology:
Socionext will demonstrate it’s highly integrated ISO26262-certified SC1721/ SC1722/ SC1723 Series Graphics Display Controllers feature built-in safety mechanisms that enable error-free, safety-critical content to meet the safety standards required by today’s multi-display cockpit applications.
- Socionext (North Hall, booth 10564): Automotive solutions for remote display, RADAR and embedded sensing and cabin monitoring.
Socionext has created a variety of smart RADAR solutions, including 24GHz and 60GHz, and has developed a roadmap showcasing future technologies features and capabilities to support multiple in-cabin uses, including seat occupancy monitoring, child presence detection, and theft prevention.
Prophesee will demonstrate the alignment of their event-based vision systems with the unique, intelligent edge-learning capabilities of Akida. This enables learning of gestures on device without storing data on device or on cloud - an important aspect of security and privacy expected.
- Prophesee (The Venetian Hotel): Edge learning on device for complex visuals.
Stay tuned for further updates on our partners CES schedule and announcements.
Our demo imaged is the vibration analysis and wonder if that is where Socionext involved for one?
在CES与Socionext擦肩而过？没关系，一文了解 Socionext最新解决方案-icspec
Socionext 在CES2023上有哪些精彩亮相？快跟小编一起来看看吧
Passing Socionext at CES? It doesn't matter, read about Socionext's latest solutions
Source: Socionext Release time: 2023-02-06
The world's largest technology event, the International Consumer Electronics Show (CES 2023), successfully concluded in Las Vegas on January 8, US time. More than 3,200 technology companies from 10 countries and regions participated, and the number of on-site participants exceeded 100,000.The Socionext booth is located in the Mobility Zone area, displaying the company's advanced technologies and solutions in the fields of automotive autonomous driving, automotive smart cockpits, and millimeter-wave radar applications, demonstrating Socionext's continuous investment in innovative research and development to meet the changing needs of the market.
New car display controller
Socionext demonstrated the company's new generation of vehicle-mounted display controllers to the audience. This product is equipped with APIX®3 SerDes repeater, which supports the connection of multiple intelligent display controllers in a daisy chain. A single chain can connect up to four displays. This chip is designed and developed in strict accordance with ISO 26262 functional safety, and the single chip meets the requirements of ASIL-B level. Socionext's latest vehicle display controller can perfectly meet the latest changes in vehicle architecture, as well as the development trend of widescreen and multi-screen vehicle displays.
Indigo3-X Cluster Demonstrator
Low-power millimeter-wave radar and its car cockpit monitoring solution
Socionext has developed a variety of millimeter-wave radar sensors that can be applied to various scenarios such as smart homes and small AIoT devices. The company's newly developed 60GHz millimeter-wave radar sensor is highly integrated, with built-in antennas, wireless circuits, A/D converters, FIFO memory, SPI or I2C interfaces, and a high-function fixed sensor that can flexibly change the duty cycle and control power consumption. Sequencer, as well as high-performance ranging and angle measuring signal processing circuits, automatic start and other functions, can easily obtain and use 3D position information without high-frequency and signal processing-related expertise, suitable for multi-person detection and non-contact gesture operation Etc., it can be applied to various scenarios in the car cockpit including seat occupancy monitoring, child retention detection and anti-theft.
Autonomous driving AI solution
Socionext has partnered with artificial intelligence provider BrainChip to develop a smart sensor data solution based on BrainChip's Akida® processor IP. BrainChip's flexible AI processing fabric IP delivers neuromorphic, event-based computing, achieving ultimate performance while minimizing silicon footprint and power consumption. Sensor data can be analyzed in real time through distributed high-performance and low-power edge inference, resulting in faster response time and lower energy consumption.
Neuromorphic AI Processor
Automotive grade custom chip solution
At this CES exhibition, Socionext also demonstrated its powerful design and development capabilities for automotive-grade chips. With strong front-end expertise, superb 7nm/5nm design capabilities, and rich experience in automotive-grade chip design, Socionext has successfully provided leading-edge automotive-grade chips for customers in China and around the world, playing a leading role in industry technology .Socionext can provide customers with automotive-grade quality software and hardware technologies, including SoC architecture, system integration, Spec in the ASIC development process, software engineers and technologies from low-level to high-level, and automotive-grade quality assurance. Socionext has rich development experience Support products conform to ISO26262/ISO21434 functional safety system.
Socionext’s Automotive Custom SoC Solutions and Services
Socionext also works closely with manufacturing subcontractors to carry out various measures such as quality management, and strives to provide high-quality, high-reliability products. Through design strategy (DFM/DFT), test strategy, reliability test (AEC-Q100), production process control (automotive service package), functional safety system (ISO26262), documentation (PPAP), customer complaints (8D report) and other links A design that ensures protection against bad chips and reduces manufacturing defects.
SocionextAutomotive SoC Support Flow
CES2023
CES 2023 is a big "show" field, which allows people to see the power of technological innovation to lead to the future and change the world. All along, Socionext has provided customized SoC services to global customers through its unique "Solution SoC" business model and world-leading technologies, helping customers create differentiated products and services . In the future, the company will continue to strengthen cooperation with more partners and strive to achieve sustainable development.
"Think Solution SoC. Think Socionext." In 2023, we will accelerate Chuangxin to the future together.
cosors
👀
I don't know if this could be of interest for one or the other in this group because of in general about Renesas and GlobaFoundries.
The info is a few weeks old and I postet it for the other group:
from Hebrew:
The technology was developed at Dialog, which was sold to Renesas 12 months ago. GlobalFoundries is positioned as a spin-off of the Israeli Weebit Nano, which develops resistive memory technology for SoC components.
GlobalFoundries reported that it has acquired all the intellectual property and production infrastructure of the Japanese company Renesas Electronics, in the field of Resistive Non-Volatile Memory (NVM) in Conductive Bridging Random Access Memory - CBRAM technology. The attractiveness of the technology in terms of GlobalFoundries is that it is a memory technology that is very economical in power, which is relatively easy to integrate into embedded devices for the industrial market and IoT accessories.
In the company's estimation, the move will strengthen it in the field of manufacturing services for SoC components, as the transaction provides customers with additional technology that allows them to optimize and differentiate their products. GlobalFoundries Chief Business Officer Mike Hogan said the deal is part of a move to position the company as a major supplier of non-volatile memories. "CBRAM technology brings a new paradigm to all battery powered accessories. In some cases, it extends the time between battery charges from hours to weeks and in some cases even to years."
CBRAM technology was originally developed by Dialog Semiconductor. In 2020, GlobalFoundries signed a licensing agreement with Dialog that allows it to use the technology, and in the time that has passed since then it has certified the manufacturing process of the memories using the company's 22FDX process (22 nm). In February 2021, Renesas acquired Dialog for approximately $5.9 billion to strengthen its position in the IoT market. Dialog is particularly strong in the field of low-power connectivity components, while Ransas is mainly focused in the field of microcontrollers (MCU) and MCU-based solutions.
The Renesas company has been working on a process of renewed organization and focus in recent years, with the previous major move being made in 2019, when it purchased the American IDT company for $6.3 billion. The purpose of these two transactions is to enable Renesas to provide solutions based on three central legs: analog, carrier and embedded processing. This strategy made the memory activity that was in the dialogue unnecessary for Renesas. In the first nine months of 2022, GlobalFounders' sales totaled approximately $6.1 billion. The company expects that sales in 2022 as a whole will amount to approximately $8.2 billion, compared to sales of approximately $6.6 billion in 2021. Ransas' sales last year totaled approximately $11.4 billion.
The latest move simultaneously strengthens and weakens the Weebit Nano company of Hod Hasharon, which recently completed a series of certification tests for a demonstration chip of an embedded system that includes its resistive memory, and entered into negotiations with chip manufacturing service providers (Foundries) in order to implement the technology on the production lines. Since its establishment in 2015, Weebit Nano has been engaged in the development of resistive non-volatile memory (NVM) of the ReRAM type in collaboration with the French research institute CEA-Leti .
This memory technology is characterized by extremely low power consumption alongside resistance to high temperatures and harsh environmental conditions. One of its outstanding advantages lies in the fact that the memory cells can be miniaturized and produced even in geometries as small as 28 nm and 22 nm - in contrast to FLASH memories, which are difficult to produce in a geometry smaller than 40 nm. The company plans to illustrate the result through the production of the first series of 22 nm components already in the coming weeks.
Published in the categories: News , Semiconductors
Posted in tags: Globalfoundries , Ransas"
https://techtime.co.il/2023/02/12/globalfoundries-8/
_________
If there are spelling errors then it is due to the translation from Hebrew. Some words are sometimes translated this way or that way, e.g. Renesas sometimes becomes Ransas or similar.
The info is a few weeks old and I postet it for the other group:
from Hebrew:
"Globalfoundries acquires the resistive memory division of Renesas
February 12, 2023The technology was developed at Dialog, which was sold to Renesas 12 months ago. GlobalFoundries is positioned as a spin-off of the Israeli Weebit Nano, which develops resistive memory technology for SoC components.
GlobalFoundries reported that it has acquired all the intellectual property and production infrastructure of the Japanese company Renesas Electronics, in the field of Resistive Non-Volatile Memory (NVM) in Conductive Bridging Random Access Memory - CBRAM technology. The attractiveness of the technology in terms of GlobalFoundries is that it is a memory technology that is very economical in power, which is relatively easy to integrate into embedded devices for the industrial market and IoT accessories.
In the company's estimation, the move will strengthen it in the field of manufacturing services for SoC components, as the transaction provides customers with additional technology that allows them to optimize and differentiate their products. GlobalFoundries Chief Business Officer Mike Hogan said the deal is part of a move to position the company as a major supplier of non-volatile memories. "CBRAM technology brings a new paradigm to all battery powered accessories. In some cases, it extends the time between battery charges from hours to weeks and in some cases even to years."
CBRAM technology was originally developed by Dialog Semiconductor. In 2020, GlobalFoundries signed a licensing agreement with Dialog that allows it to use the technology, and in the time that has passed since then it has certified the manufacturing process of the memories using the company's 22FDX process (22 nm). In February 2021, Renesas acquired Dialog for approximately $5.9 billion to strengthen its position in the IoT market. Dialog is particularly strong in the field of low-power connectivity components, while Ransas is mainly focused in the field of microcontrollers (MCU) and MCU-based solutions.
The Renesas company has been working on a process of renewed organization and focus in recent years, with the previous major move being made in 2019, when it purchased the American IDT company for $6.3 billion. The purpose of these two transactions is to enable Renesas to provide solutions based on three central legs: analog, carrier and embedded processing. This strategy made the memory activity that was in the dialogue unnecessary for Renesas. In the first nine months of 2022, GlobalFounders' sales totaled approximately $6.1 billion. The company expects that sales in 2022 as a whole will amount to approximately $8.2 billion, compared to sales of approximately $6.6 billion in 2021. Ransas' sales last year totaled approximately $11.4 billion.
The Israeli competitor of Globalfoundries
The latest move simultaneously strengthens and weakens the Weebit Nano company of Hod Hasharon, which recently completed a series of certification tests for a demonstration chip of an embedded system that includes its resistive memory, and entered into negotiations with chip manufacturing service providers (Foundries) in order to implement the technology on the production lines. Since its establishment in 2015, Weebit Nano has been engaged in the development of resistive non-volatile memory (NVM) of the ReRAM type in collaboration with the French research institute CEA-Leti .
This memory technology is characterized by extremely low power consumption alongside resistance to high temperatures and harsh environmental conditions. One of its outstanding advantages lies in the fact that the memory cells can be miniaturized and produced even in geometries as small as 28 nm and 22 nm - in contrast to FLASH memories, which are difficult to produce in a geometry smaller than 40 nm. The company plans to illustrate the result through the production of the first series of 22 nm components already in the coming weeks.
Published in the categories: News , Semiconductors
Posted in tags: Globalfoundries , Ransas"
https://techtime.co.il/2023/02/12/globalfoundries-8/
_________
If there are spelling errors then it is due to the translation from Hebrew. Some words are sometimes translated this way or that way, e.g. Renesas sometimes becomes Ransas or similar.
Last edited:
cosors
👀
Sorry if already posted.
NEWS FOR LAWYERS AND SOURCING PROFESSIONALS
ChatGPT and subsequent artificial intelligence (AI) programs have been in the headlines recently. Not as common is the discussion of the cost associated with developing and operating such AI tools or if such AI is right for every job.
It is estimated that ChatGPT can cost millions of dollars per day to operate. Given the potentially large price tag, consumers may ask how users can harness the benefit of AI without the high operating cost and what the best technology is in applications where precise decision-making and outcomes are desired. Some believe that the answer to both of these questions is neuromorphic computing.
What Is Neuromorphic Computing?
Neuromorphic computing is designed to mimic the human brain, operating in a manner that allows the technology to solve problems in ways our brains would. The chips that power neuromorphic computing are specially designed with the same structure as our brain’s neurons and synapses, allowing it to make decisions and judgments in a way that typical computer systems or algorithms cannot.
Neuromorphic computing is intended to be more efficient, powerful, and cost-effective than other AI technologies. Although still in development and not widely deployed, it is being evaluated in various settings, including cloud computing, robotics, and autonomous vehicle technology.
The End of the Algorithm in AI?
Rather than processing all the data to follow an algorithm to an answer, the goal of neuromorphic computing is to decipher the necessary information to determine the correct solution. Leveraging this would allow companies and consumers to implement technology into everyday life wherever a human touch is required—rather than utilizing answers based solely on an algorithm.
AI is effective at providing large amounts of computing power, responding to queries that may take a human or even a standard computer a long time to answer. Neuromorphic computing, on the other hand, takes a more active approach, giving the correct response or action to a scenario.
Key Takeaway
As technology and society integrate on a deeper level, there will be an increased demand on our computers and technology to interact with us as a human would with speech, movement, and reason. Neuromorphic computing’s deployment is no easy feat, and we will be on the lookout for how companies bring humanity into future computers and technologies.
Topics: Robotics and Artificial Intelligence, Technology and Tech Transactions"
https://www.morganlewis.com/blogs/s...ogy-for-activities-that-require-a-human-touch
"BLOG POST
TECH & SOURCING @ MORGAN LEWIS
TECHNOLOGY, OUTSOURCING, AND COMMERCIAL TRANSACTIONSNEWS FOR LAWYERS AND SOURCING PROFESSIONALS
Neuromorphic Computing: A More Efficient Technology for Activities That Require a Human Touch
March 01, 2023ChatGPT and subsequent artificial intelligence (AI) programs have been in the headlines recently. Not as common is the discussion of the cost associated with developing and operating such AI tools or if such AI is right for every job.
It is estimated that ChatGPT can cost millions of dollars per day to operate. Given the potentially large price tag, consumers may ask how users can harness the benefit of AI without the high operating cost and what the best technology is in applications where precise decision-making and outcomes are desired. Some believe that the answer to both of these questions is neuromorphic computing.
What Is Neuromorphic Computing?
Neuromorphic computing is designed to mimic the human brain, operating in a manner that allows the technology to solve problems in ways our brains would. The chips that power neuromorphic computing are specially designed with the same structure as our brain’s neurons and synapses, allowing it to make decisions and judgments in a way that typical computer systems or algorithms cannot.
Neuromorphic computing is intended to be more efficient, powerful, and cost-effective than other AI technologies. Although still in development and not widely deployed, it is being evaluated in various settings, including cloud computing, robotics, and autonomous vehicle technology.
The End of the Algorithm in AI?
Rather than processing all the data to follow an algorithm to an answer, the goal of neuromorphic computing is to decipher the necessary information to determine the correct solution. Leveraging this would allow companies and consumers to implement technology into everyday life wherever a human touch is required—rather than utilizing answers based solely on an algorithm.
AI is effective at providing large amounts of computing power, responding to queries that may take a human or even a standard computer a long time to answer. Neuromorphic computing, on the other hand, takes a more active approach, giving the correct response or action to a scenario.
Key Takeaway
As technology and society integrate on a deeper level, there will be an increased demand on our computers and technology to interact with us as a human would with speech, movement, and reason. Neuromorphic computing’s deployment is no easy feat, and we will be on the lookout for how companies bring humanity into future computers and technologies.
Topics: Robotics and Artificial Intelligence, Technology and Tech Transactions"
https://www.morganlewis.com/blogs/s...ogy-for-activities-that-require-a-human-touch
thelittleshort
Top Bloke
Repost by Panasonic Automotive employee
From memory there have been Panasonic posts, but not for automotive applications?
Panasonic Automotive Cockpit Integrated Solutions link
Panasonic Automotive Cabin Interior Innovation Solutions link
Panasonic Automotive Systems Company of America (PASA) is an industry-leading global supplier to Automotive Original Equipment Manufacturers (OEM’s) for infotainment systems and advanced connected car solutions. Our clients include Ford, GM, Chrysler, Daimler, Fiat, Tesla, Honda, Toyota, Nissan and many others. In serving our customers, we don't just strive to put tomorrow's technology on the road – we pride ourselves on recruiting the brightest and best to do it.
From memory there have been Panasonic posts, but not for automotive applications?
Panasonic Automotive Cockpit Integrated Solutions link
Panasonic Automotive Cabin Interior Innovation Solutions link
Panasonic Automotive Systems Company of America (PASA) is an industry-leading global supplier to Automotive Original Equipment Manufacturers (OEM’s) for infotainment systems and advanced connected car solutions. Our clients include Ford, GM, Chrysler, Daimler, Fiat, Tesla, Honda, Toyota, Nissan and many others. In serving our customers, we don't just strive to put tomorrow's technology on the road – we pride ourselves on recruiting the brightest and best to do it.
BaconLover
Founding Member
Ian
Founding Member
BaconLover
Founding Member
supersonic001
Regular
This will be a great webinar with Todd Vierra! | Rob Telson
This will be a great webinar with Todd Vierra! Don't miss it. #edgeai #edgecomputing #tinyml
Fact Finder
Top 20
From the 18 page Mercedes Benz EQXX pdf which is part of @Pmel ‘s
post above:
“Neuromorphic computing – a car that thinks like you
Another key efficiency feature of the VISION EQXX that takes its cue from nature is the way it thinks. It uses an innovative form of information processing called neuromorphic computing. The hardware runs so-called spiking neural networks. Information is coded in discrete spikes and energy is only consumed when a spike occurs, which reduces energy consumption by orders of magnitude.
Working with California-based artificial intelligence experts BrainChip, Mercedes-Benz engineers developed systems based on BrainChip’s Akida hardware and software. The example in the VISION EQXX is the “Hey Mercedes” hot-word detection. Structured along neuromorphic principles, it is five to ten times more efficient than conventional voice control.
Although neuromorphic computing is still in its infancy, systems like these will be available on the market in just a few years. When applied on scale throughout a vehicle, they have the potential to radically reduce the energy needed to run the latest AI technologies.”
Locked away in a pdf document and made available for ongoing public consumption. Brainchip gives all the appearance of being part of the furniture at Mercedes Benz.
My opinion only DYOR
FF
AKIDA BALLISTA
post above:
“Neuromorphic computing – a car that thinks like you
Another key efficiency feature of the VISION EQXX that takes its cue from nature is the way it thinks. It uses an innovative form of information processing called neuromorphic computing. The hardware runs so-called spiking neural networks. Information is coded in discrete spikes and energy is only consumed when a spike occurs, which reduces energy consumption by orders of magnitude.
Working with California-based artificial intelligence experts BrainChip, Mercedes-Benz engineers developed systems based on BrainChip’s Akida hardware and software. The example in the VISION EQXX is the “Hey Mercedes” hot-word detection. Structured along neuromorphic principles, it is five to ten times more efficient than conventional voice control.
Although neuromorphic computing is still in its infancy, systems like these will be available on the market in just a few years. When applied on scale throughout a vehicle, they have the potential to radically reduce the energy needed to run the latest AI technologies.”
Locked away in a pdf document and made available for ongoing public consumption. Brainchip gives all the appearance of being part of the furniture at Mercedes Benz.
My opinion only DYOR
FF
AKIDA BALLISTA
Fact Finder
Top 20
https://www.design-reuse.com/redir3/35893/352793/8IDCcehr87FC7QuZSKvfeNO7vLEwt
The above link is taken from todays Design & Reuse daily email update under the heading Artificial Intelligence News. It is free to subscribe and Brainchip information and releases regularly feature now.
This entry appears with one other under Artificial Intelligence which is about using Ai and CHATGpt to maximise the design for incorporating Ai in your chips.
My opinion only DYOR
FF
AKIDA BALLISTA
The above link is taken from todays Design & Reuse daily email update under the heading Artificial Intelligence News. It is free to subscribe and Brainchip information and releases regularly feature now.
This entry appears with one other under Artificial Intelligence which is about using Ai and CHATGpt to maximise the design for incorporating Ai in your chips.
My opinion only DYOR
FF
AKIDA BALLISTA
We know Quadric. They are part owned by MegaChips, who have a deal with us. So if Quadric can start using Akida IP this will be huge.OPPO article mentions production H2 CY2023 for launch next year. Most likely February/March when they launch new models. And they will launch their own SoC.
So if any new smartphones next year will have Akida IP they will have to be manufactured in H2 CY2023.
Vivo X90 launched 3 February, 2023 was first phone to use Snapdragon 8 Gen2 SoC. New Vivo model next year could be the first phone to have Prophesee tech. One to keep an eye on for specs.
OPPO to Launch its Own SoC in 2024: Here’s What We Know
TECHNOLOGY
By Sidharth Joseph Last updated Feb 22, 2023
View attachment 30995
Chinese smartphone manufacturer, OPPO is planning to bring the company’s very own self developed SoC in 2024. This in-house chipset that the company will be introducing will make the brand much more independent and will give an extra advantage in terms of performance and pricing with its competitors.
Reports on the Chinese microblogging platform, Weibo reveals that OPPO has already started its work on its very own smartphone chipset which the company would release in 2024. The chipset is going to use 4nm manufacturing process and is expected to be made by TSMC. It is also expected that the chipset will be compatible with 5G smartphones.
It was from a MediaTek executive that we had first received the information about OPPO’s self made SoC and later an insider of OPPO had also revealed that the company has already started its works. Reports also reveal that the company has invested about 1.4 billion in the research and development of its chipset.
OPPO’s first custom made chipset, MariSilicon X was released in 2021 and was a 6nm imaging chipset. The company also has a connector chipset as well called MariSilicon Y. All these indicate that the brand in the future is trying not to depend more on leading SoC manufacturers like MediaTek and Qualcomm but to rely more on their own self developed chipsets.
Leading smartphone manufacturers like Samsung, Apple and Google already have their own chipsets and with OPPO also following the same trend, it will definitely enable the company to achieve independence and come to the forefront in the smartphone market.
https://www.thetechoutlook.com/news/technology/oppo-to-launch-its-own-soc-in-2024-heres-what-we-know/#:~:text=Chinese%20smartphone%20manufacturer%2C%20OPPO%20is,and%20pricing%20with%20its%20competitors.
Debayan Roy (Gadgetsdata)
@Gadgetsdata
Oppo's Self-Developed smartphone SoC is expected to arrive in next year, 2024.
• It is a 4nm SoC, supports 5G.
• The design process of this SoC is completed and it is ready to be sent into manufacturing in the 2nd Half of 2023.
• It might be manufactured by TSMC.
Via: Weibo
Another company to keep an eye on is MediaTek.
As the industry leader in developing powerful, highly integrated and efficient system-on-chip products, MediaTek is enabling the future of AI by creating an ecosystem of Edge-AI hardware processing paired with comprehensive software tools across its product range - smartphones to smart homes, wearables, IoT and connected cars.
MediaTek NeuroPilot
We’re meeting the Edge AI challenge head-on with MediaTek NeuroPilot. Through heterogeneous computing capabilities such as CPUs, GPUs and APUs (AI processing units) we embed into our system-on-chip products, we are providing high performance and power efficiency for AI features and applications. Developers can target these specific processing units within the system-on-chip or, they can let MediaTek NeuroPoint SDK intelligently handle the processing allocation for them.
Learn how it works >
Artificial Intelligence | Edge-AI Technology
MediaTek leads the pack with its Edge-AI technology solutions with respect to Artificial Intelligence. Try out some powerful, highly integrated SoC products.www.mediatek.com
Many companies becoming active in edge AI. Hopefully, BRN's Akida will become a high volume building block for a majority of devices in future.
Osram recently partnered with MegaChips' other AI partner Quadric. Will be using Quadric's Chimera general purpose neural processor. Appears this application may be better suited to Quadric in leau of Akida. Or they don't know about Akida yet.
Quadric, Ams Osram to develop smart sensing solutions for edge-based applications
Jan 11, 2023 | Abhishek Jadhav
Quadric, an edge AI chip provider, has partnered with Ams Osram, an optical solutions provider, to create a smart sensing solution for edge computing applications. The partnership will combine Ams Osram’s Mira family of CMOS sensors and Quadric’s Chimera general-purpose neural processors. Leveraging the power of both their respective strengths, Quadric says it has developed a low-power smart sensing module that combines image capturing and machine learning capabilities.
The Ams Osram Mira220 CMOS image sensor is designed for use in 2D and 3D consumer and industrial machine vision applications. With the intention of further enhancing its sensor and releasing versions with higher resolutions, Quadric is integrating future models of its CMOS sensors with a range of computing power from 1 to 16 TOPs in its Chimera processor lineup. Both companies showcased this sensing module at CES 2023.
The company says that the Mira CMOS image sensor family provides maximum resolution in a compact form factor while minimizing power consumption. By incorporating new modules within the Mira family, the company says it can provide a broad selection of resolutions that cater to different applications needing high performance and energy efficiency.
“The combination of Ams Osram’s sensors and Quadric’s processing into a single low-power module opens up vast new possibilities for the deployment of smart vision sensing,” said Joost Seijnaeve, the vice president and general manager of CMOS Image Sensors for Ams Osram.
The Quadric Chimera general-purpose neural processor has a unified hardware and software architecture optimized for on-device AI computing. This architecture enables the processor to execute matrix and vector operations, as well as scalar code within a single execution pipeline. The processor’s design combines a neural processor, digital signal processor and real-time CPU into an individual programmable core. This means that it is exceptionally proficient in dealing with multiple tasks, the company says.
“Quadric is excited to be joining forces with Ams Osram,” said Veer Kheterpal, the CEO at Quadric. “Empowering device makers with the capability for a fully-programmable smart sensing device at incredibly low power levels will open a vast new tranche of deployments of machine learning in edge devices.”
In March of 2022, Quadric raised $21 million in a Series B funding round led by mobility supplier Denso’s NSITEXE subsidiary. The company stated the funding would, among other things, accelerate advancements in its next-generation processor architecture.
Quadric, Ams Osram to develop smart sensing solutions for edge-based applications | Edge Industry Review
The partnership will combine Ams Osram’s Mira family of CMOS sensors and Quadric’s Chimera general-purpose neural processors.www.edgeir.com
Similar threads
