BRN Discussion Ongoing

[Diogenese]

The system features USB 3.0 and micro-USB ports, HDMI, 4GB LPDDR4 memory, 32GB eMMC with up to 1TB micro-SDXC expansion, dual-band Wi-Fi, and two gigabit Ethernet ports for external camera connections, all within a compact, passively-cooled chassis, powered by 12V DC .

https://www.tomshardware.com/tech-i...p-claim-625-times-less-power-41-times-smaller

This is quite interesting. I know Samsung has been sniffing around Brainchip at past presentations etc. and Brainchip used a Samsung DVS camera in a demonstration back in 2018/19 ish.

I wonder what this neuromorphic “not off the shelf secret sauce” is that they have used in this test chip?


The second thing that caught my attention is the C transformers talked about. I don’t know what C transformers is exactly, but I think it is safe to say it has something to do with Transformers which is one of the key new features added to the Akida 2.0 architecture based on costumer requests.

@Diogenese can you potentially have a better look at this test chip and see if you can see anything of interest?

Hi Tfm,


To quote Dirty Harry: "I'm sorta wondering myself ..."

I've posted about KAIST a couple of times in the last week or so:

• #80,099

From the Tom's hardware article:


Korean researchers power-shame Nvidia with new neural AI chip — claim 625 times less power draw, 41 times smaller | Tom's Hardware (tomshardware.com)

Korean researchers power-shame Nvidia with new neural AI chip — claim 625 times less power draw, 41 times smaller

Claim Samsung-fabbed chip is the first ultra-low power LLM processor.

www.tomshardware.com

A team of scientists from the Korea Advanced Institute of Science and Technology (KAIST) detailed their 'Complementary-Transformer' AI chip during the recent 2024 International Solid-State Circuits Conference (ISSCC). The new C-Transformer chip is claimed to be the world's first ultra-low power AI accelerator chip capable of large language model (LLM) processing.
...

The architecture of the C-Transformer chip is interesting to look at and is characterized by three main functional feature blocks. Firstly, there is a Homogeneous DNN-Transformer / Spiking-transformer Core (HDSC) with a Hybrid Multiplication-Accumulation Unit (HMAU) to efficiently process the dynamically changing distribution energy. Secondly, we have an Output Spike Speculation Unit (OSSU) to reduce the latency and computations of spike domain processing. Thirdly, the researchers implemented an Implicit Weight Generation Unit (IWGU) with Extended Sign Compression (ESC) to reduce External Memory Access (EMA) energy consumption.


I haven't found the KAIST C-transformer patent, so it may still be under wraps. Consequently, I can't find any detail of the C-Transformer.

A couple of their earlier patent applications:


US2023098672A1 ENERGY-EFFICIENT RETRAINING METHOD OF GENERATIVE NEURAL NETWORK FOR DOMAIN-SPECIFIC OPTIMIZATION 20210924

YOO HOI JUN [KR]; KIM SO YEON [KR]

The main claim relates to retaining a generative NN:

1 . An energy-efficient retraining method of a generative neural network for domain-specific optimization, the energy-efficient retraining method comprising:
(a) retraining, by a mobile device, a pretrained generative neural network model with respect to some data of a new user dataset;
(b) comparing, by the mobile device, the pretrained generative neural network model and a generative neural network model retrained for each layer with each other in terms of a relative change rate of weights;
(c) selecting, by the mobile device, specific layers having high relative change rate of weights, among layers of the pretrained generative neural network model, as layers to be retrained; and
(d) performing, by the mobile device, weight update for only the layers selected in step (c).

Claim 4 sheds a bit of light on what may be part of the combined DNN-CNN in that it does forward propagation learning, a technique Akida uses (SNN), and subsequently does back propagation (DNN)
4 . The energy-efficient retraining method according to claim 3, wherein, in step (d), the mobile device maintains original weights without weight update for unselected layers, after selecting the k continuous layers, does not perform even back propagation for unselected layers before a first one of the selected layers, performs forward propagation in only a first epoch of retraining, and reuses a result of the forward propagation of the first epoch in repeated retraining epochs thereafter.

This one is about DNN learning:

US2023072432A1 APPARATUS AND METHOD FOR ACCELERATING DEEP NEURAL NETWORK LEARNING FOR DEEP REINFORCEMENT LEARNING 20210831

a deep neural network (DNN) learning accelerating apparatus for deep reinforcement learning, the apparatus including: a DNN operation core configured to perform DNN learning for the deep reinforcement learning; and a weight training unit configured to train a weight parameter to accelerate the DNN learning and transmit it to the DNN operation core, the weight training unit including: a neural network weight memory storing the weight parameter; a neural network pruning unit configured to store a sparse weight pattern generated as a result of performing the weight pruning based on the weight parameter; and a weight prefetcher configured to select/align only pieces of weight data of which values are not zero (0) from the neural network weight memory using the sparse weight pattern and transmit the pieces of weight data of which the values are not zero to the DNN operation core.

Interestingly, KAIST have also been working on MRAM, which could be used in an analog NN:

US2023410864A1 MRAM CELL WITH PAIR OF MAGNETIC TUNNEL JUNCTIONS HAVING OPPOSITE STATES AND MEMORY DEVICE USING THE SAME 20220620

An MRAM cell includes a switch unit configured to determine opening and closing thereof by a word line voltage and to activate a current path between a bit line and a bit line bar in an opened state thereof, first and second MTJs having opposite states, respectively, and connected in series between the bit line and the bit line bar, to constitute a storage node, and a sensing line configured to be activated in a reading mode of the MRAM cell, thereby creating data reading information based on a voltage between the first and second MTJs, wherein the first and second MTJs have different ones of a low resistance state and a high resistance state, respectively, in accordance with a voltage drop direction between the bit line and the bit line bar, thereby storing data of 0 or 1.


All this is no doubt funded by the Korean government who recently announced they were pulling out all the stops to catch up on AI, so perhaps the KAIST hubris is intended to impress their paymasters.

ChatGPT2 is orders of magnitude smaller than ChatGPT3, and nothing like GPT4 and subsequent versions.

 

[Stable Genius]

BrainChip’s Akida™ neuromorphic processor has been integrated into several microcontroller units (MCUs) and embedded systems-on-chip (SoCs). Here are some notable instances:

1. SiFive Essential™ Processors with Akida-E:
   • These compact processors are suitable for deeply embedded applications. They combine SiFive Essential™ processors with BrainChip’s Akida-E, providing efficient solutions for low-power, small form factor devices at the edge. These MCUs are ideal for applications like audio processing, filtering, denoising for hearing aids, and hearables1.
2. SiFive X280 Intelligence™ AI Dataflow Processors with Akida-S:
   • For more complex applications such as object detection and robotics, the SiFive X280 Intelligence™ AI Dataflow Processors are tightly integrated with BrainChip’s Akida-S. These processors offer enhanced digital signal processing (DSP) and machine learning (ML) capabilities, making them suitable for a wide range of intelligent edge devices1.

Probably already been posted, anyway here it is again for what it's worth...Tech.

Hi @TECH those quotes make great reading. They are written in the past tense as though they have integrated the tech and SiFive has SOC on the market? The links go back to Brainchip’s web page. Is this info something new I’ve missed.

I appreciate SFive were part of the Gen2 promotional material but didn’t realise it had progressed to products.

Did you have more info as this is huge and I suggest ASX announcement worthy?

 

[Diogenese]

BrainChip’s Akida™ neuromorphic processor has been integrated into several microcontroller units (MCUs) and embedded systems-on-chip (SoCs). Here are some notable instances:

1. SiFive Essential™ Processors with Akida-E:
   • These compact processors are suitable for deeply embedded applications. They combine SiFive Essential™ processors with BrainChip’s Akida-E, providing efficient solutions for low-power, small form factor devices at the edge. These MCUs are ideal for applications like audio processing, filtering, denoising for hearing aids, and hearables1.
2. SiFive X280 Intelligence™ AI Dataflow Processors with Akida-S:
   • For more complex applications such as object detection and robotics, the SiFive X280 Intelligence™ AI Dataflow Processors are tightly integrated with BrainChip’s Akida-S. These processors offer enhanced digital signal processing (DSP) and machine learning (ML) capabilities, making them suitable for a wide range of intelligent edge devices1.

Probably already been posted, anyway here it is again for what it's worth...Tech.

Hi Tech,

Just as I was getting dressed for a formal dinner, that blew me sox off - took the thongs as well.

Akida in silicon with SiFive is big news. Confirmation that Akida has been integrated into SiFive processors is big news. SiFive are into RISC-V, while ARM use RISC-IV.

... or am I reading too much into this. Is it just simulator compatibility.

The SiFive testimonial for Akida 2 reads as though this is a work-in-progress:

"Through our collaboration with BrainChip, we are enabling the combination of SiFive’s RISC-V processor IP portfolio and BrainChip’s 2nd generation Akida neuromorophic IP to provide a power-efficient, high capability solution for AI processing on the Edge. Deeply embedded applications can benefit from the combination of compact SiFive Essential™ processors with BrainChip’s Akida-E, efficient processors; more complex applications including object detection, robotics, and more can take advantage of SiFive X280 Intelligence™ AI Dataflow Processors tightly integrated with BrainChip’s Akida-S or Akida-P neural processors."

In either case, it's great news that the RISC-V ingénue SiFive can co-promote Akida, as they are very well known in the processor world, maybe almost as well known as Brainchip.

 

[Diogenese]

This is a complementary technology for Akida:

https://www.msn.com/en-au/news/othe...E&cvid=54ef76db22fe4cf5963090599b2b4d0c&ei=64

Australian innovation in X-ray technology​

Story by Sky News Australia
• 7h

Akida could do the diagnosis in situ in real time, and maybe control the X-ray machine to get different views.

 

[Boab]

Hi Tech,

Just as I was getting dressed for a formal dinner, that blew me sox off - took the thongs as well.

Akida in silicon with SiFive is big news. Confirmation that Akida has been integrated into SiFive processors is big news. SiFive are into RISC-V, while ARM use RISC-IV.

... or am I reading too much into this. Is it just simulator compatibility.

The SiFive testimonial for Akida 2 reads as though this is a work-in-progress:

"Through our collaboration with BrainChip, we are enabling the combination of SiFive’s RISC-V processor IP portfolio and BrainChip’s 2nd generation Akida neuromorophic IP to provide a power-efficient, high capability solution for AI processing on the Edge. Deeply embedded applications can benefit from the combination of compact SiFive Essential™ processors with BrainChip’s Akida-E, efficient processors; more complex applications including object detection, robotics, and more can take advantage of SiFive X280 Intelligence™ AI Dataflow Processors tightly integrated with BrainChip’s Akida-S or Akida-P neural processors."

In either case, it's great news that the RISC-V ingénue SiFive can co-promote Akida, as they are very well known in the processor world, maybe almost as well known as Brainchip.

I laughed so hard picturing those thongs getting blown off. Seriously funny.
Cheers

 

[Teach22]

BrainChip’s Akida™ neuromorphic processor has been integrated into several microcontroller units (MCUs) and embedded systems-on-chip (SoCs). Here are some notable instances:

1. SiFive Essential™ Processors with Akida-E:
   • These compact processors are suitable for deeply embedded applications. They combine SiFive Essential™ processors with BrainChip’s Akida-E, providing efficient solutions for low-power, small form factor devices at the edge. These MCUs are ideal for applications like audio processing, filtering, denoising for hearing aids, and hearables1.
2. SiFive X280 Intelligence™ AI Dataflow Processors with Akida-S:
   • For more complex applications such as object detection and robotics, the SiFive X280 Intelligence™ AI Dataflow Processors are tightly integrated with BrainChip’s Akida-S. These processors offer enhanced digital signal processing (DSP) and machine learning (ML) capabilities, making them suitable for a wide range of intelligent edge devices1.

Probably already been posted, anyway here it is again for what it's worth...Tech.

Link please….

 

[Esq.111]

YAY! We're third on the list!

If we get 10 percent of 29.2 billion, that's 2.92 billion dollars!!!!


View attachment 58809

Top 10 Neuromorphic Computing Companies | Better Output Technology​

Tajammul Pangarkar
Updated · Mar 7, 2024
SHARE:
Scoop.market.us is supported by its audience. When you purchase through links on our site, we may earn an affiliate commission. Learn more.

close
Advertiser Disclosure
At Market.us Scoop, we strive to bring you the most accurate and up-to-date information by utilizing a variety of resources, including paid and free sources, primary research, and phone interviews. Our data is available to the public free of charge, and we encourage you to use it to inform your personal or business decisions. If you choose to republish our data on your own website, we simply ask that you provide a proper citation or link back to the respective page on Market.us Scoop. We appreciate your support and look forward to continuing to provide valuable insights for our audience.
Table of Contents

• Neuromorphic Computing Market Overview
• Market Drivers
• Market Size
• List of Major Companies
• Intel
• IBM
• BrainChip
• Qualcomm
• NVIDIA
• Hewlett-Packard
• Samsung
• Accenture
• Cadence-Design
• Knowm

Neuromorphic Computing Market Overview​

Inspired by the brain’s neural networks, Neuromorphic computing revolutionizes computer architecture by emulating biological neurons. Using spiking neurons and synaptic plasticity principles, neuromorphic hardware enables efficient, low-power processing.
This approach finds applications in diverse fields, such as sensor networks for environmental monitoring, robotics for enhanced perception and control, and neuroscience research for studying brain function.
With its potential to deliver highly efficient, brain-like computation, neuromorphic computing holds promise for transforming various industries and advancing our understanding of complex biological systems.

Market Drivers​

The global neuromorphic computing market is growing due to several factors. Firstly, various industries have a rising demand for efficient computing solutions.
Neuromorphic architectures, known for their energy efficiency and parallel processing capabilities, are becoming increasingly popular. The emergence of edge computing and IoT devices also drives the need for low-power, high-performance solutions, further boosting the adoption of neuromorphic technologies.
Furthermore, ongoing advancements in hardware and software, coupled with increased investments, are expanding the market. These trends indicate a promising future for the global neuromorphic computing market’s growth.

Market Size​

The global neuromorphic computing market reached USD 4.2 billion in 2022 and is projected to reach USD 29.2 billion by 2032, with an estimated compound annual growth rate (CAGR) of 22% from 2023 to 2032.

List of Major Companies​

These are the top 10 companies operating in the Neuromorphic Computing Market:

• Intel
• IBM
• BrainChip
• Qualcomm
• NVIDIA
• Hewlett-Packard
• Samsung
• Accenture
• Cadence-Design
• Knowm

Intel​

Company Overview​

Establishment Year	1968
Headquarter	Santa Clara, California, U.S.
Key Management	Pat Gelsinger (CEO)
Revenue (US$ Bn)	$ 54.2 Billion (2023)
Headcount	~ 124,800 (2023)
Website	https://www.intel.com/

About Intel Corporation​

Intel Corporation leads the way in neuromorphic computing, leveraging its expertise in semiconductor manufacturing and technology innovation.
The groundbreaking neuromorphic research chip Loihi is at the forefront of its efforts, mimicking the human brain’s neural networks for efficient data processing. With robotics, healthcare, and IoT applications, Intel collaborates with research institutions and industry partners to explore innovative uses for neuromorphic computing.
Through ongoing research and development, Intel aims to push the boundaries of computing capabilities, positioning itself as a key player in shaping the future of neuromorphic technologies.

Geographical Presence​

Intel Corporation strategically situates itself globally, headquartered in Santa Clara, California. Its primary operations cover North America, including major Oregon, Arizona, and Texas manufacturing sites.
In Europe, Intel maintains hubs in Ireland and Germany, fostering innovation. Expanding into the Asia-Pacific, notable facilities in China and India tap diverse markets and talent.
Operations extend to Japan and other Asia-Pacific nations, driving technological progress. Intel also maintains a presence in Latin America, the Middle East, and Africa, emphasizing its dedication to global innovation and market leadership.

Recent Developments​

• In October 2021, Intel unveiled Loihi 2, its latest neuromorphic computing chip, representing a notable progression in the field. Utilizing Intel 4 process technology, Loihi 2 presents improved capabilities for studying advanced neuromorphic neural networks.
• In March 2020, Intel revealed information about Pohoiki Springs, a neuromorphic computing system featuring 100 million artificial neurons for processing tasks.

IBM​

Company Overview​

Establishment Year	1911
Headquarter	Armonk, New York, United States
Key Management	Arvind Krishna (Chairman & CEO)
Revenue (US$ Bn)	$ 61.8 B (2022)
Headcount	~ 288,300 (2022)
Website	https://www.ibm.com/

About IBM Corporation​

IBM Corporation is a leader in neuromorphic computing, pioneering innovative solutions for various applications. Its TrueNorth architecture is a milestone in this field, offering highly efficient and low-power processing capabilities akin to the human brain.
With applications spanning healthcare, finance, and cybersecurity, IBM collaborates with research institutions and industry partners to explore and develop TrueNorth technology.
Through ongoing research and development efforts, IBM continues to push the boundaries of neuromorphic computing, positioning itself as a driving force in shaping the future of computing architectures.

Geographical Presence​

IBM Corporation maintains a global presence with operations spanning across multiple continents. The company’s headquarters are located in Armonk, New York, USA.
IBM has established offices, research facilities, and manufacturing sites in key regions worldwide, including North America, Europe, Asia Pacific, and Latin America. This extensive geographical footprint enables IBM to effectively serve its diverse customer base and collaborate with partners on a global scale.
Additionally, IBM’s presence in various regions facilitates market expansion and supports its commitment to innovation and technology leadership across the globe.

Recent Developments​

• In February 2024, Wipro introduced a new solution that harnesses IBM’s WatsonX AI and data platform capabilities to boost AI adoption in enterprises.
• In October 2023, IBM launched its inaugural AI-focused chip, NorthPole, claiming it to be 22 times faster than competitors. The chip aims to surpass current computing performance standards and address the growing demand for AI technologies.

BrainChip​

Company Overview​

Establishment Year	2004
Headquarter	Sydney, NSW, Australia
Key Management	Sean Hehir (CEO)
Revenue (US$ Bn)	$ 2.3 Billion (2023)
Headcount	~ 70 (2023)
Website	https://brainchip.com/

About BrainChip Holdings​

BrainChip Holdings Ltd. leads neuromorphic computing, specializing in ultra-low-power solutions for various applications. Its flagship product, the Akida Neuromorphic System-On-Chip (NSoC), offers efficient processing capabilities tailored for edge computing and IoT devices.
With applications spanning surveillance, autonomous vehicles, and cybersecurity, BrainChip collaborates with industry partners and research institutions to innovate and develop solutions using the Akida platform.
Through continuous research and development, BrainChip is committed to advancing neuromorphic computing technology to meet the evolving needs of diverse markets.

Geographical Presence​

BrainChip Holdings Ltd., a leader in neuromorphic computing solutions, has a solid global presence spanning North America, Europe, and Asia-Pacific.
With a focus on key markets such as the United States, Europe (including the UK, Germany, and France), and Asia-Pacific countries like China, Japan, Australia, and South Korea, BrainChip addresses diverse industry needs in cybersecurity, surveillance, automotive, and IoT.
Through its strategic positioning, BrainChip aims to drive innovation and provide cutting-edge AI technologies worldwide.

Recent Developments​

• In December 2023, BrainChip Holdings and Unigen Corporation partnered to offer a new Unigen Cupcake Edge AI Server version, utilizing BrainChip’s Akida neuromorphic processor for enhanced compactness and performance.
• In January 2023, BrainChip Holdings and MYWAI formed a strategic partnership to provide advanced Edge AI solutions using neuromorphic computing.

Qualcomm​

Company Overview​

Establishment Year	1985
Headquarter	San Diego, California, U.S.
Key Management	Cristiano Amon (CEO)
Revenue (US$ Bn)	$38.5 Billion (2023)
Headcount	~ 50,000 (2023)
Website	https://www.qualcomm.com/

About Qualcomm Incorporated​

Qualcomm Incorporated is pivotal in advancing neuromorphic computing, drawing upon its semiconductor design and mobile technologies expertise.
With its Zeroth platform, Qualcomm offers efficient processing capabilities tailored for various applications, including edge computing and IoT devices. The company collaborates with research institutions and industry partners to explore innovative solutions and drive advancements in the field.
Qualcomm’s commitment to research and development positions it as a leading force in unlocking the potential of neuromorphic computing for diverse industries.

Geographical Presence​

Qualcomm, headquartered in San Diego, California, maintains a global presence with operations in key regions worldwide. It focuses on research and development in the US, notably in 5G technology.
Across the Asia-Pacific, including China, South Korea, Japan, India, and Taiwan, Qualcomm has established offices and partnerships to support local collaborations. It engages in research, development, and customer support in Europe, the Middle East, and Africa, particularly in the UK, Germany, and France.
Additionally, Qualcomm has a presence in select Latin American countries, contributing to mobile connectivity adoption. Qualcomm serves global markets through strategic positioning, drives innovation, and maintains semiconductor leadership.

Recent Development​

• In October 2023, Qualcomm unveiled Snapdragon Seamless, a cross-platform technology allowing Android, Windows, and Snapdragon devices on different operating systems to seamlessly discover and share information, functioning as a unified system.
• In September 2023, EE revealed its inaugural strategic plan to become the UK’s foremost customer-focused brand for its home broadband service. Teaming up with Qualcomm Technologies, Inc., EE intends to launch new in-home hardware featuring Qualcomm Wi-Fi 7 platforms. This effort seeks to provide superior home connectivity to UK consumers, positioning them as early adopters of next-generation Wi-Fi technology.

NVIDIA​

Company Overview​

Establishment Year	1993
Headquarter	Santa Clara, California, U.S.
Key Management	Jensen Huang (President and CEO)
Revenue (US$ Bn)	$60.9 B (2023)
Headcount	~ 29,600 (2023)
Website	http://conservis.ag/

About NVIDIA Corporation​

NVIDIA Corporation, a leader in graphics processing units (GPUs) and artificial intelligence (AI) solutions, is advancing neuromorphic computing. Through dedicated research and development, NVIDIA designs specialized hardware architectures and efficient algorithms inspired by the brain’s neural networks.
Its product portfolio, including GPUs and software frameworks like CUDA and TensorRT, supports neuromorphic computing applications in various domains.
Collaborations with industry partners and research institutions further accelerate innovation, positioning NVIDIA at the forefront of the neuromorphic computing revolution.
With its expertise, NVIDIA is poised to reshape AI and computational neuroscience, driving the development of smarter, more efficient systems across industries.

Geographical Presence​

NVIDIA Corporation, a global leader in GPUs and AI technology, maintains a significant presence worldwide. In Santa Clara, California, NVIDIA strategically establishes offices, research facilities, and partnerships across North America, Europe, the Middle East, Africa, and Asia-Pacific.
With key locations in major cities like London, Munich, Beijing, and Bangalore, NVIDIA serves its global customer base and collaborates with technology companies and research institutions to drive innovation in AI-driven technologies.
This extensive geographical presence underscores NVIDIA’s commitment to advancing GPU technology and AI solutions globally.

Recent Developments​

• In February 2024, AMD introduced a solution utilizing ROCm technology to allow NVIDIA CUDA binaries to run on AMD graphics hardware seamlessly without requiring any modifications.
• In November 2023, NVIDIA introduced CUDA Quantum 0.5, its latest platform tailored for quantum-classical computing application development.

​

Evening Bravo ,

Interesting article , but their numbers are not accurate.

Unless $2.299 billion USD have gone to the private accounts of a handful and the remainder landed in BRAINCHIPS accounts ( under $1 million USD ) for the financial year.

The mind truely boggles at the range and depth which our accounts could be played with , due to NDA's.

Once upon a time I used to be a trusting sole , with the passage of time.... no longer.

I do love your research, though think this article may have been created by a random bullshite generator.

Regards,
Esq.

 

[Getupthere]

How Lenovo wants to unlock AI for all at the edge | Kirk Skaugen interview — VentureBeat

It makes a lot of sense to put AI as far out on the edge of the network as possible, as it can solve problems for people like first responders and reduce the amount of useless storage sent to the cloud.

apple.news

 

[Fullmoonfever]

Might help @McHale

Was a Christmas present

BRN Discussion Ongoing

Merry Christmas, happy holidays all Be safe and enjoy the time with loved ones, friends, fam and doing what makes you happy however you celebrate this time. Oct French article on the X320 I needed to work around through VPN, private tab etc to get to read with an English translation...

thestockexchange.com.au

However, in a Secret Santa, some more interesting news / developments

According to a paper just released a couple of days ago (not peer reviewed as yet) it appears some Snr Researchers over at Ericsson have been playing with Akida and "for instance, to demonstrate the feasibility of AI-enabled ZE-IoT, we have developed a prototype of a solar-powered AI-enabled ZE-IoT camera device with neuromorphic computing."

My question would be is this something off their own back or do we have a hand in the background somewhere as well


Towards 6G Zero-Energy Internet of Things:
Standards, Trends, and Recent Results

• December 2023

https://www.researchgate.net/public...of_Things_Standards_Trends_and_Recent_Results

Further to the Ericsson post above.

Just saw Dr John Smee, Sr. Vice President, Engineering, Qualcomm Technologies, Inc. posted his article on LinkedIn 3 mths ago about AI and 6G but what I liked was one of the four comments to the post...they obviously feel there is some merit and opps with neuromorphic

John Smee on LinkedIn: The U.S. R&D Agenda Needed to Unlock the Evolution of AI & 6G

AI and advanced communications are rapidly transforming how we live, work, and interact. From the use of #AI for malware detection in devices, to how vehicles…

www.linkedin.com

John Smee
3mo

AI and advanced communications are rapidly transforming how we live, work, and interact. From the use of #AI for malware detection in devices, to how vehicles process data from cameras and sensors, there is an ever-expanding list of applications that bring AI and wireless communications together. Please check out my article below on the opportunity for the U.S. to lead in creating these emerging technologies. We are thrilled that the federal government has established the National Science Foundation (NSF) Directorate for Technology Innovations and Partnerships (TIP), which will accelerate R&D investments in critical technologies. NSF has a tremendous opportunity to ensure the U.S. remains at the forefront of technological innovation - creating jobs and paving the way for a brighter future.

The U.S. R&D Agenda Needed to Unlock the Evolution of AI & 6G​

John Smee on LinkedIn​

Mallik Tatipamula
CTO @ Ericsson | Ph.D. in Information Science and Technology
3mo

Great article, John. It has been fantastic for us at Ericsson in collaborating with NSF in launching 4 major public-private partnerships including NSF PAWR, NSF RINGS, NSF FuSE and NSF/DoD 5G security (convergent accelerator). I could not agree more on your statement on "TIP focus on use-inspired and translational research efforts" between now and 2030, when 6G is commercial". I believe, while 5G is the convergence of "connectivity, computing and control (AI)", and 6G will extend this convergence even further with sensing resulting in "convergence of connectivity, compute, control (AI) and sensing", which presents massive opportunities for innovations, combined with advancements in adjacent technologies such as meta materials, neuromorphic computing among others

 

[jtardif999]

More Edge AI news from STMicroelectronics:

“STMicroelectronics has announced its second-generation STM32 microprocessors, promising improved performance and acceleration for edge artificial intelligence (AI) and machine learning workloads — and a move to 64-bit processors cores for the first time in the range's history.”

STMicro's STM32 Family Makes the 64-Bit Jump with New Edge AI, 3D-Capable STM32MP Parts

Retaining the STM32 name despite a shift to the 64-bit Arm Cortex-A35 core, STMicro's latest STM32 parts pack in the power.

www.hackster.io

STMicro's STM32 Family Makes the 64-Bit Jump with New Edge AI, 3D-Capable STM32MP Parts​

Retaining the STM32 name despite a shift to the 64-bit Arm Cortex-A35 core, STMicro's latest STM32 parts pack in the power.​

Gareth HalfacreeFollow
1 day ago • HW101 / Internet of Things / Machine Learning & AI

"Our embedded MPUs address the trend that’s pushing more workloads and greater demands to smart devices, often deployed at the IoT [Internet of Things] edge, for faster response and increased efficiency," says STMicro's Stephane Henry of the company's new processors.

STMicro has announced new parts in the STM32 range, and they're the company's most powerful yet. (: STMicroelectronics)

"The new STM32MP2 devices we are announcing today," Henry continues, "extend the performance trajectory, introducing our most powerful processing engine, now adding edge AI, and supported by the STM32 ecosystem to accelerate product development."

The first surprise in the new chip range: a move to 64-bit processor cores, despite retaining the STM32 name. The STM32MP2 chips are the company's first to use the Arm Cortex-A35 core, running at speeds of up to 1.5GHz, alongside a 32-bit Arm Cortex-M33 microcontroller core running at up to 400MHz, a dedicated graphics processing unit (GPU) with 3D acceleration and support for 1080p displays, a video processing unit (VPU), and a neural processing unit (NPU) designed to accelerate on-device edge AI and machine learning work.

The new parts include up to two 64-bit Arm Cortex-A35 cores running at up to 1.5GHz, but keep the "STM32" name. (: STMicroelectronics)

This NPU, STMicro claims, can deliver up to 1.35 tera-operations per second (TOPS) of compute, with workloads able to run on the Cortex-A35, Cortex-M33, GPU, VPU, or NPU cores as-required. It's not present in all models, however: only the STM32MP257, STM32MP255, and STM32MP23x chips include the AI co-processor and 3D GPU; the remainder of the range stick with one or two Cortex-A35 cores and a single Cortex-M33.

STMicro is making much of the parts' industrial qualifications: the company says the chips can run 24 hours a day for 10 years, with a junction temperature rated at -40°C to 125°C (-40°F to 257°F). The new parts also come with the promise of Security Evaluation Standard for IoT Platforms (SESIP) Level 3 certification, readying them for the US Cyber Trust Mark announced back in July last year — along with the company's recently-unveiled 32-bit STM32WBA55 and STM32WBA54 chips.

Some, but not all, models in the range include 3D-capable GPUs and a 1.35 TOPS neural network coprocessor. (: STMicroelectronics)

Other features of the parts include up to three gigabit Ethernet ports through the use of a two-port switch, support for Time Sensitive Networking (TSN), PCI Express Gen. 2, USB 3.0, and up to three CAN-FD buses, RGB, LVDS, and MIPI Display Serial Interface (DSI) video outputs with H.264 video acceleration on the GPU-equipped models, and a MIPI Camera Serial Interface 2 (CSI-2) input with image signal processor.

While STMicro is unveiling the parts now, however, it is not quite ready to take orders: the parts are due to enter volume production in June this year, with a development board to follow in due course.
More information is available on the STMicro website.

microcontroller
internet of things
machine learning
artificial intelligence
computer vision

Gareth HalfacreeFollow
Freelance journalist, technical author, hacker, tinkerer, erstwhile sysadmin. For hire: freelance@halfacree.co.uk.

---

https://www.st.com/content/st_com/en/campaigns/microprocessor-stm32mp2.html

I particularly like this paragraph:
“This NPU, STMicro claims, can deliver up to 1.35 tera-operations per second (TOPS) of compute, with workloads able to run on the Cortex-A35, Cortex-M33, GPU, VPU, or NPU cores as-required.”
My interpretation of this is it’s saying that workloads can also run on the NPU independently. That to me says this could be us since NPUs usually just support the running of workloads on associated CPUs. AFAIK there are no other NPUs capable of running independently. This could be Akida IP. Maybe @Diogenese could take a look?

 

[Diogenese]

I particularly like this paragraph:
“This NPU, STMicro claims, can deliver up to 1.35 tera-operations per second (TOPS) of compute, with workloads able to run on the Cortex-A35, Cortex-M33, GPU, VPU, or NPU cores as-required.”
My interpretation of this is it’s saying that workloads can also run on the NPU independently. That to me says this could be us since NPUs usually just support the running of workloads on associated CPUs. AFAIK there are no other NPUs capable of running independently. This could be Akida IP. Maybe @Diogenese could take a look?

No mention of spikes or our special sauce:

https://www.hackster.io/news/stmicr...edge-ai-3d-capable-stm32mp-parts-2a470d8eb2cb

STMicroelectronics powers up the intelligent edge with second-generation STM32 microprocessors, bringing performance boost and industrial resilience - ST News

STMicroelectronics powers up the intelligent edge with second-generation STM32 microprocessors, bringing performance boost and industrial resilience - ST News

STMicroelectronics (NYSE: STM), a global semiconductor leader serving customers across the spectrum of electronics applications, is releasing new devices...

newsroom.st.com

These are ST’s first MPUs to contain a 64-bit central processing unit (CPU), the Arm Cortex®-A35, which runs up to 1.5GHz to raise the main processing capability compared to the first-generation STM32MP1 devices.

The CPU is the center of a true heterogeneous processing engine that also contains a Cortex-M33 core. In addition, there is a graphics processor (GPU), neural processor (NPU), and a video processor (VPU). AI workloads can run on the CPU, GPU, or NPU depending on processor loading and application demands, for optimal performance and energy efficiency. In fact, the high efficiency of these MPUs permits system designs to go without active cooling, gaining advantages such as smaller size, silent operation, greater reliability, and reduced power consumption.

 

[TopCat]

From the mining industry….

Harnessing the benefits of Edge AI in mining

Artificial intelligence (AI) is a modern reality that is transforming industrial operations and outcomes. Due to recent advances in edge AI technologies

www.australianmining.com.au

“In a nutshell, it is the way of the future, and businesses that take advantage of edge AI now will have a competitive advantage. Those who stall on the adoption of these technologies run the risk of losing their ability to compete effectively.”

 

[Frangipani]

To me, adding inbetween the lines, the comment looks a bit like this..

Especially with the part..
"one of our female engineers"..


"Because it's International Women's day, we decided to let one of our "female" engineers (yes we really do have them) "drive" our precious EQXX prototype (we have photos to prove, she's actually sitting in the driver's seat pretending to drive!)".
"You go girl!"

Hi DingoBorat,

although Mercedes Benz’s tribute to International Women’s Day on March 8th may have come across as a little awkward, I don’t think they meant to sound condescending at all.

As for “one of our female engineers”: in German, we can simply add the suffix -in to many professions in order to express that a person is a female, similar to the much rarer English suffix -ess as in seamstress, waitress, stewardess or Empress.

So the technically male form Ingenieur (engineer) becomes Ingenieurin, Informatiker (computer scientist) becomes Informatikerin, Pilot (you guessed it - pilot) becomes Pilotin, Arzt (medical doctor) becomes Ärztin (some cases like this one involve a vowel change as well), you name it…

(I will just mention the current heated debate about gender-neutral language in German as a side note - personally, I habitually list both genders when talking about a mixed group of male and female persons such as “Lehrerinnen und Lehrer” (teachers), whereas other people consider the male plural to be inclusive enough. Since the 1980s, new forms have emerged such as Lehrer/innen, Lehrer_innen or LehrerInnen to encompass both male and female forms in writing and even more recently also Lehrer*innen (which is pronounced with a so-called glottal stop, that sounds like the tiny gap in uh-oh) to refer to all genders, including people who identify as non-binary etc. This has led to very heated debates in recent years about the use of gender-neutral language when talking about people. Note that this does not concern the three grammatical genders of German nouns in general, though, so no relief here for beginning learners of German ).

So when the Mercedes press department wants to refer to women in engineering in German-language publications, they will use exclusively feminine nouns such as Ingenieurin (singular) or Ingenieurinnen (plural), which do not have an equivalent in English, though. That’s why in this English-language post, the author felt the need to add the adjective “female”, which, however, seems to reinforce the stereotype that engineers are normally male and the female ones are the exotic and thus odd ones out. While the male-female ratio amongst automotive engineers admittedly happens to be pretty lopsided, the German nouns Ingenieur/Ingenieurin are more neutral in that respect (even though one could argue that even from a grammatical point of view, the male form is the norm, and the female form is marked as different).

Unfortunately that doesn’t equate to greater gender equality in German-speaking countries, though: women in male-dominated fields sadly still get belittled by some ignorant people, even though they have the exact same degree, and there is also the gender pay gap problem in general - so, much to be desired! In fact, some women even prefer to refer to themselves as “Ingenieur” rather than “Ingenieurin”, because they feel that this way others will respect them more, although to me personally it sounds very odd.

Anyway, back to the original post and the lady in the driver’s seat: the ADAS development engineer’s name is Katharina Kupferschmid (check out Markus Schäfer’s LinkedIn comment). She has been with Mercedes for almost 20 years, and I bet she was not just posing for that picture in Saudi Arabia, merely pretending to drive, but was absolutely and confidently test driving the EQXX Vision herself.

What makes me so sure about that? Well, Mercedes not only seems to have trusted her not to crash their precious concept car, but had even handpicked her to chauffeur German Chancellor Olaf Scholz around a 2.8 km test track in an EQE SUV, when he visited the Sindelfingen Mercedes plant on March 4. However, according to German news reports and Scholz’s social media accounts (where he also mentions discussing the importance of AI with Mercedes staff, by the way), the Chancellor actually asked to sit behind the steering wheel himself, telling his three passengers (including Mercedes CEO Ola Källenius, who sat in the back seat) he would really enjoy doing so, because he rarely gets to drive himself any more these days, as he is usually being chauffeured in his armoured official state car (which happens to be a Mercedes, too - a Mercedes S 680 Guard, to be precise). And you know what they say: When You Wish Upon A Mercedes Star, Your Dreams Come True…

(Photos of the Chancellor in the driver‘s seat of course led to a flurry of sarcastic comments by posters critical of his government coalition’s politics, but I suppose there would have been just as many taunting comments had he been caught in the passenger seat, being chauffeured by a female driver… )

See Katharina Kupferschmid’s activity on LinkedIn

Sign in or join now to see posts like this one and more.

www.linkedin.com

While I believe that Mercedes are genuinely proud of their female staff and I appreciate them shining the light on women that can inspire others around the world to take the steering wheel of life into their own hands, the photos taken in Saudi Arabia and the accompanying text seem to ignore the fact that women there still do not enjoy equal rights compared to men and have only been allowed to drive a car since 2018! (It’s hard to imagine this official Mercedes LinkedIn post could intentionally be taking a dig at the Saudi government, although I would love to read that between the lines).

IMO Mercedes did a better job in 2019, when they honoured International Women’s Day with a short film on Bertha Benz (1849-1944), the automotive pioneer, business partner and wife of Carl Benz. Not many people seem to be aware that she was instrumental in the foundation of Mercedes-Benz. The 4 min short film Bertha Benz - The Journey That Changed Everything, shot like a classic Western, tells the story of the world’s first long distance drive in a car from Mannheim to Pforzheim (106 km) in 1888. The only gripe I have about it is that they shot the film in English rather than German, which is totally unauthentic. Why?! They could have easily subtitled it for an international audience… Anyway, I thought it was worth watching.


You can also find a an excellent summary of that legendary trip on Wikipedia:

Bertha Benz - Wikipedia

en.m.wikipedia.org

 

[RobjHunt]

Looking forward to the days shinanigins.

 

[BrainShit]

Hi Tech,

Just as I was getting dressed for a formal dinner, that blew me sox off - took the thongs as well.

Akida in silicon with SiFive is big news. Confirmation that Akida has been integrated into SiFive processors is big news. SiFive are into RISC-V, while ARM use RISC-IV.

... or am I reading too much into this. Is it just simulator compatibility.

The SiFive testimonial for Akida 2 reads as though this is a work-in-progress:

"Through our collaboration with BrainChip, we are enabling the combination of SiFive’s RISC-V processor IP portfolio and BrainChip’s 2nd generation Akida neuromorophic IP to provide a power-efficient, high capability solution for AI processing on the Edge. Deeply embedded applications can benefit from the combination of compact SiFive Essential™ processors with BrainChip’s Akida-E, efficient processors; more complex applications including object detection, robotics, and more can take advantage of SiFive X280 Intelligence™ AI Dataflow Processors tightly integrated with BrainChip’s Akida-S or Akida-P neural processors."

In either case, it's great news that the RISC-V ingénue SiFive can co-promote Akida, as they are very well known in the processor world, maybe almost as well known as Brainchip.

I don't think that Aikda is already integrated into SiFive processors and in silicon for mass marked... but I belive, beside the X280, that the X390 can also contain Akida as an accelerator, if desired by a customer.

Why?
"The SiFive Intelligence X390 builds upon the success of its predecessor, the SiFive Intelligence X280, in combining AI and ML applications with hardware accelerators for mobile, infrastructure, and automotive applications."

X280 = "Through our collaboration with BrainChip, we are enabling the combination of SiFive’s RISC-V processor IP portfolio and BrainChip’s 2nd generation Akida neuromorophic IP to provide a power-efficient, high capability solution for AI processing on the Edge. Deeply embedded applications can benefit from the combination of compact SiFive Essential™ processors with BrainChip’s Akida-E, efficient processors; more complex applications including object detection, robotics, and more can take advantage of SiFive X280 Intelligence™ AI Dataflow Processors tightly integrated with BrainChip’s Akida-S or Akida-P neural processors."

SiFive CEO Patrick Little said in a statement, “The flexibility of SiFive’s RISC-V solutions allows companies to address the unique computing requirements of these segments and capitalize on the momentum around generative AI, where we have seen double-digit design wins, and for other cutting-edge applications.”

He said the company has 350 design wins and customers include Intel, Amazon, Qualcomm, Samsung, Google, NASA and more. SiFive started in the embedded market and moved up the food chain to high-performance cores.

This means that the Akida can be optionally selected via ARM (Risc) and SiFive (Risc-V).
If they already available in silicon for mass marked? I doubt that... without any announcement from SiFive or BC.

Link: https://venturebeat.com/games/sifive-unveils-two-new-high-performance-risc-v-processors/
Link: https://www.allaboutcircuits.com/news/sifive-rolls-out-risc-v-cores-aimed-at-generative-ai-and-ml/
Link: https://www.sifive.com/risc-v-core-ip

 

[DingoBorat]

Hi DingoBorat,

although Mercedes Benz’s tribute to International Women’s Day on March 8th may have come across as a little awkward, I don’t think they meant to sound condescending at all.

As for “one of our female engineers”: in German, we can simply add the suffix -in to many professions in order to express that a person is a female, similar to the much rarer English suffix -ess as in seamstress, waitress, stewardess or Empress.

So the technically male form Ingenieur (engineer) becomes Ingenieurin, Informatiker (computer scientist) becomes Informatikerin, Pilot (you guessed it - pilot) becomes Pilotin, Arzt (medical doctor) becomes Ärztin (some cases like this one involve a vowel change as well), you name it…

(I will just mention the current heated debate about gender-neutral language in German as a side note - personally, I habitually list both genders when talking about a mixed group of male and female persons such as “Lehrerinnen und Lehrer” (teachers), whereas other people consider the male plural to be inclusive enough. Since the 1980s, new forms have emerged such as Lehrer/innen, Lehrer_innen or LehrerInnen to encompass both male and female forms in writing and even more recently also Lehrer*innen (which is pronounced with a so-called glottal stop, that sounds like the tiny gap in uh-oh) to refer to all genders, including people who identify as non-binary etc. This has led to very heated debates in recent years about the use of gender-neutral language when talking about people. Note that this does not concern the three grammatical genders of German nouns in general, though, so no relief here for beginning learners of German ).

So when the Mercedes press department wants to refer to women in engineering in German-language publications, they will use exclusively feminine nouns such as Ingenieurin (singular) or Ingenieurinnen (plural), which do not have an equivalent in English, though. That’s why in this English-language post, the author felt the need to add the adjective “female”, which, however, seems to reinforce the stereotype that engineers are normally male and the female ones are the exotic and thus odd ones out. While the male-female ratio amongst automotive engineers admittedly happens to be pretty lopsided, the German nouns Ingenieur/Ingenieurin are more neutral in that respect (even though one could argue that even from a grammatical point of view, the male form is the norm, and the female form is marked as different).

Unfortunately that doesn’t equate to greater gender equality in German-speaking countries, though: women in male-dominated fields sadly still get belittled by some ignorant people, even though they have the exact same degree, and there is also the gender pay gap problem in general - so, much to be desired! In fact, some women even prefer to refer to themselves as “Ingenieur” rather than “Ingenieurin”, because they feel that this way others will respect them more, although to me personally it sounds very odd.

Anyway, back to the original post and the lady in the driver’s seat: the ADAS development engineer’s name is Katharina Kupferschmid (check out Markus Schäfer’s LinkedIn comment). She has been with Mercedes for almost 20 years, and I bet she was not just posing for that picture in Saudi Arabia, merely pretending to drive, but was absolutely and confidently test driving the EQXX Vision herself.

View attachment 58836



View attachment 58829


What makes me so sure about that? Well, Mercedes not only seems to have trusted her not to crash their precious concept car, but had even handpicked her to chauffeur German Chancellor Olaf Scholz around a 2.8 km test track in an EQE SUV, when he visited the Sindelfingen Mercedes plant on March 4. However, according to German news reports and Scholz’s social media accounts (where he also mentions discussing the importance of AI with Mercedes staff, by the way), the Chancellor actually asked to sit behind the steering wheel himself, telling his three passengers (including Mercedes CEO Ola Källenius, who sat in the back seat) he would really enjoy doing so, because he rarely gets to drive himself any more these days, as he is usually being chauffeured in his armoured official state car (which happens to be a Mercedes, too - a Mercedes S 680 Guard, to be precise). And you know what they say: When You Wish Upon A Mercedes Star, Your Dreams Come True…

(Photos of the Chancellor in the driver‘s seat of course led to a flurry of sarcastic comments by posters critical of his government coalition’s politics, but I suppose there would have been just as many taunting comments had he been caught in the passenger seat, being chauffeured by a female driver… )

See Katharina Kupferschmid’s activity on LinkedIn

Sign in or join now to see posts like this one and more.

www.linkedin.com

View attachment 58826


While I believe that Mercedes are genuinely proud of their female staff and I appreciate them shining the light on women that can inspire others around the world to take the steering wheel of life into their own hands, the photos taken in Saudi Arabia and the accompanying text seem to ignore the fact that women there still do not enjoy equal rights compared to men and have only been allowed to drive a car since 2018! (It’s hard to imagine this official Mercedes LinkedIn post could intentionally be taking a dig at the Saudi government, although I would love to read that between the lines).

IMO Mercedes did a better job in 2019, when they honoured International Women’s Day with a short film on Bertha Benz (1849-1944), the automotive pioneer, business partner and wife of Carl Benz. Not many people seem to be aware that she was instrumental in the foundation of Mercedes-Benz. The 4 min short film Bertha Benz - The Journey That Changed Everything, shot like a classic Western, tells the story of the world’s first long distance drive in a car from Mannheim to Pforzheim (106 km) in 1888. The only gripe I have about it is that they shot the film in English rather than German, which is totally unauthentic. Why?! They could have easily subtitled it for an international audience… Anyway, I thought it was worth watching.


You can also find a an excellent summary of that legendary trip on Wikipedia:

Bertha Benz - Wikipedia

en.m.wikipedia.org

All good Frangipani and I agree the short film is much better, although the short BMW tribute to Sabine Schmitz, shared by Corsors, absolutely trumps it.

I'm sure the comment wasn't meant to be condescending, but it just had too many aspects that could be "seen" in that way, the biggest one really, was that it "seemed" International Women's Day, was a "Special Occasion" to let a woman drive it..

Political correctness has reached absurd levels, in my opinion, where the only thing really needed, is respect for all.

I'm surprised that things like equal pay, for the same qualifications, or job, are not the norm these days.

Beyond legislation, this comes down to the employer's, who by their very nature, even in large organisations, are just petty.

 

[RobjHunt]

What‘s your early take on the day Esqy?

 

[RobjHunt]

What‘s your early take on the day Esqy?

Just asking for a friend

 

[mrgds]

Just asking for a friend

MY EARLY CALL

 

[Esq.111]

Morning RobjHunt ,

Wild guess........ the five day chart on one hour duration for volume looks promising . culminating in a possible rise in volume mid to late afternoon today.

Noticed the last few days there have , and continue to be several sell orders at $1.50 to $3.00 ( very small ) .

Going for a mildly positive day , though if we get a positive announcement then look out.


Regards,
Esq