BRN Discussion Ongoing

[TheFunkMachine]

1. There's a webinar on the RA8 on the 29th November.

2. I spy with my little eye HVAC!!!!!


3. Remember this quote from Roger Wendlken (Renesas) :

“We see an increasing demand for real-time, on-device, intelligence in AI applications powered by our MCUs and the need to make sensors smarter for industrial and IoT devices,” said Roger Wendelken, Senior Vice President in Renesas’ IoT and Infrastructure Business Unit. “We licensed Akida neural processors because of their unique neuromorphic approach to bring hyper-efficient acceleration for today’s mainstream AI models at the edge. With the addition of advanced temporal convolution and vision transformers, we can see how low-power MCUs can revolutionize vision, perception, and predictive applications in a wide variety of markets like industrial and consumer IoT and personalized healthcare, just to name a few.”
Roger Wendelken, Senior Vice President IoT and Infrastructure Business Unit, Renesas

4. Remember this quote from Nalin Balin (Realityai, a Renesas company) :

View attachment 49721


5. And heres a video from 21/9/2023 with Sailesh from Renesas talking about HVAC's processing operations at the edge at 1 minute mark.
View attachment 49712
WATCH THIS VIDEO
https://www.renesas.com/us/en/video/generative-ai-happening-edge-and-endpoint

5. View attachment 49733 View attachment 49732

View attachment 49731

Renesas’ New Ultra-High Performance MCUs are Industry’s First Based on Arm Cortex-M85 Processor

www.bloomberg.com

6. View attachment 49735

View attachment 49708

RA8M1 Microcontrollers​

Renesas' RA8M1 MCUs feature Arm® Cortex®-M85 480 MHz core with Helium technology, up to 2 MB Flash, and 1 MB SRAM​

Renesas' RA8 family with the RA8M1 group integrates the Arm Cortex-M85 core with excellent peripherals, memory, and leading-edge security with TrustZone®. The RA8M1 devices include several low-power features to reduce the overall system power while still providing high performance. These devices combine with advanced process technology to bring together the performance, quality, and reliability that Renesas has always been known for.
The RA8M1 MCU group expands the RA family SAM into the high-performance market. These MCUs lie in the crossover space between MCUs and MPUs and are suited for customer applications that require the higher performance of an MPU but the lower power consumption and ease of use of the MCU. The RA8M1 MCUs, with the Cortex-M85 core running at 480 MHz, integrated with large Flash/SRAM memory (2 MB/1 MB) and a rich set of peripherals, are optimally defined to support AI, motor control, PLC, metering, and other broad-based industrial and IoT applications. Multiple external interfaces, connectivity peripherals, and packages from 100 pins to 224 pins are included to serve the needs of a broad range of high-performance applications.
The Arm Cortex-M85 core, with the Arm v8.1m architecture and a 7-stage superscalar pipeline, is the highest-performance Cortex-M core and introduces Helium, the M-Profile Vector Extension (MVE) that significantly accelerates signal processing and machine learning capabilities, enabling an unprecedented four times the acceleration in ML tasks and three times the acceleration in DSP tasks versus the older Cortex-M7 core. Helium is a single instruction, multiple data (SIMD) extension to the instruction set, which can provide a performance uplift by processing more data per instruction. This is similar to Neon on the Arm Cortex-A processor cores but is more optimized for resource-constrained and lower power microcontrollers and is designed to Cortex-M principles: real-time, area efficient, and integrated with TrustZone. It adds 150 scalar and vector instructions and allows the processing of 128-bit registers. Used optimally, this can eliminate the need for an additional DSP core in the system. Helium brings intelligent processing to the edge, enabling local processing (less traffic to the cloud and lower costs) and lower overall system power, allowing small, low-power embedded systems to address compute needs in applications as diverse as audio processing, drone navigation and control, AR/VR applications, sensor hubs, and image processing.
Features

• 480 MHz Arm Cortex-M85 core with Helium (MVE for AI/ML)
• Up to 2 MB Flash memory and 1 MB SRAM incl. TCM; 384 KB user SRAM and 128 KB TCM are ECC protected
• 32 KB I/D caches (ECC protected), 12 KB data Flash
• Advanced security with TrustZone, RSIP cryptographic engine, immutable storage, and tamper protection
• Scalable from 100-pin to 224-pin packages
• Octal SPI with decryption-on-the-fly, Ethernet MAC with DMA, CAN-FD, and USB HS/FS (host and device) connectivity options
• CEU camera i/f, 12-bit ADCs, 12-bit DACs, high-speed analog comparators, and 3x sample and hold circuits

• SCI (UART, simple SPI, simple I2C), SPI, I2C, I3C
• Arm Cortex-M85 high-performance core featuring Armv8.1m architecture with Helium for DSP/ML acceleration
• Octal SPI with secure interface to external memory for storage of code and data
• Advanced security for highly secure IoT
• Open Arm Ecosystem, easy-to-use flexible software package, and comprehensive solutions to enable fast development
• S/H enable motor control applications

Applications

• Industrial automation
• IoT gateways/hubs
• Smart home/home automation products
• Thermostats

• Home appliances (refrigerators, ovens, washing machines, and more)
• Security cameras
• Building automation (HVAC, access)

View attachment 49711

WEBINAR

Eval Board​

	IMAGE	MANUFACTURER PART NUMBER	DESCRIPTION	AVAILABLE QUANTITY	VIEW DETAILS
		RTK7EKA8M1S00001BE	EVALUATION KIT FOR RA8M1 MCU GRO	0 - Immediate	View Details

RA8M1 Arm® Cortex®-M85 ICs​

	IMAGE	MANUFACTURER PART NUMBER	DESCRIPTION	AVAILABLE QUANTITY	VIEW DETAILS
		R7FA8M1AHECBD#UC0	MCU RA8M1 ARM CM85 2M FBGA224 -4	310 - Immediate	View Details
		R7FA8M1AHECFB#AA0	MCU RA8M1 ARM CM85 2M LQFP144 -4	232 - Immediate	View Details
		R7FA8M1AHECFB#BA0	MCU RA8M1 ARM CM85 2M LQFP144 -4	0 - Immediate	View Details
		R7FA8M1AHECBD#BC0	MCU RA8M1 ARM CM85 2M FBGA224 -4	0 - Immediate	View Details

Published: 2023-10-31

https://www.digikey.com/en/product-highlight/r/renesas/ra8m1-arm-cortex-m85

Renesas is taping out a chip using the spiking neural network (SNN) technology developed by Brainchip.

Learn more about the collaboration between BrainChip and Renesas as they tap out a chip using the Spiking Neural Network (SNN) technology.

brainchip.com

Just a gentle reminder that the supposed chip MCU by Renasas containing Akida is according to the provided press release supposed to be released to market December 23

It’s supposedly on 22nm architecture
Using RISC V technology to compete with Arm M85 offering if I’m not miss understanding it.

 

[MDhere]

Now that I uninstalled candy crush I have a bit more time on my hands and i was clearingmy emails and tidbits that i save etc and came across the snn canny edge detector - google scholar report by our very own Krishnamurthy Vemuru which refers to LiDar as well. Not sure if @MadMayHam | 合氣道 is still amongst as but he did mention this earlier this year.

Does anyone know if LiDar uses this form of tech and which companies would? I will delete it from my files now but im curious why I put it there in the first place and didn't post about it at the time.

 

[hotty4040]

behind the scenes

You're correct @cosors:

" In the background " = " Not so visible " so to speak...... or even " not visible at all " Just like our NDA's, and EAP's, unfortunately.

hotty...

 

[TheFunkMachine]

In the recent interviews and press release from the CEO they have made a point of mentioning Akida in use for noise cancellation in headphones and hearing aids.

This has made me go down the rabbit hole trying to find some solid connections.

This article talks about the use of Neuromorphic event based singnal processing in cochlea implanted devices for noise reduction.

It’s dated 4 may 2023. Could it be Akida? I don’t know, I can’t see mention of any specific technology used in this specific project so it’s open for interpretation I guess.

Has anyone else found any good connections to Akida for this use case ?

https://www.nature.com/articles/s41928-023-00957-5

 

[TheFunkMachine]

https://www.renesas.com/us/en/about/press-room/renesas-samples-its-first-22-nm-microcontroller

No mention of SNN or secret sauce, however the 22nm MCU expected to be released Q4 2023 does fit with our proposed integration.

Any smart cookies out there that can confirm or disprove this?

 

[IloveLamp]

Probably worth tuning into

A Spike in Efficiency? - neuromorphic computing and its potential for the automotive industry

www.fzd.tu-darmstadt.de

Sign Up | LinkedIn

500 million+ members | Manage your professional identity. Build and engage with your professional network. Access knowledge, insights and opportunities.

www.linkedin.com

 

[TheFunkMachine]

https://brainchip.com/akida-foundations/

I am very impressed with Brainchips new home page. It is starting to feel like a complete product offering from evaluation to deployment. It is very clear process even for someone like myself who knows next to nothing about Chip design.

It feels like a complete business with a clear business model that emphasise on point of difference and flexibility of design integration!

They have come a looooong way in terms of marketing and exposure, and with a world class product like AKIDA TM IP product line I predict IP sales in the pipeline.

I have thought this for a few years, but it truely do feel like we are on the cusp of breaking into the market and products now.

I am keen to hear more about Renasas product release as it is expected by December this year! Will we hear anything in regards to this or will it be a watch the financials situation?

 

[MDhere]

https://www.renesas.com/us/en/about/press-room/renesas-samples-its-first-22-nm-microcontroller

No mention of SNN or secret sauce, however the 22nm MCU expected to be released Q4 2023 does fit with our proposed integration.

Any smart cookies out there that can confirm or disprove this?

I concur with you as it was the M85 -
The Cortex-M85 delivers the highest levels of performance in the ARM Cortex-M family, with Renesas Electronics as the lead developer, demonstrating AI applications running on fhte core at Embedded World (EW2023) this week. Renesas has also licensed the Akida core.

By successfully demonstrating Akida’s capabilities with the Cortex-M85 in a fully functioning environment, BrainChip paves the way for a new generation of intelligent edge devices that are capable of delivering unprecedented levels of performance and functionality. These include smart home devices, robotics, and drone control, secured system controllers and sensor hubs.

 

[Pom down under]

Worth a listen and especially for me as it’s only a few minutes long

 

[Pom down under]

https://www.megachips.co.jp/english/irinfo/files/pdf/22-24.pdf

 

[Deena]

Worth a listen and especially for me as it’s only a few minutes long

Wow! This is very thought provoking. And Brainchip will likely be front and centre in these revolutionary changes.
Deena

 

[MDhere]

Worth a listen and especially for me as it’s only a few minutes long

Thanks for sharing Pom

 

[Pom down under]

Thanks for sharing Pom

Just call me Rocket lol

 

[Adam]

Careful FF . First spelling mistake in two years . Loose .

Freddy?

 

[Tothemoon24]

Look at this bloke couldn’t even afford a table cloth back in the day

 

[Tothemoon24]

Great read from Tata


Boom

Spiking neural networks

www.tcs.com

 

[Ian]

Great read from Tata

Spiking neural networks

www.tcs.com

 

[Frangipani]

Hi @Frangipani

I am well versed with both Fund WA and VitalTrace.

VitalTrace was a client of my accounting firm from circa 2018 to 2022. I personally prepared their tax compliance for the 2020 and 2021 financial years. I won't comment on Fund WA.

A few supplier searches may or may not have taken place within VitalTraces accounting system over those few years and word is it always came up unlucky

Their founders are Michael Challenor (CTO) and Arjun Kaushik (CEO). They met at a hackathon or incubator event from memory. Arjun is a medical doctor and Michael is a (from memory) chemical engineer and has a PhD in nanotechnology and robotics. Whilst a seemingly random duo, it worked out well for them with Arjun understanding the medical need for innovation in the obstetrics area of medicine and Michael being able to roll up the sleeves and work on a proof of concept to get things started.

They ended up raising a fair wack of money and have a very good presence in the WA MedTech scene.

I got along with Michael Challenor quite well. After they moved on to another accounting firm we caught up for a coffee and I was little more direct in asking about Brainchip. Michael hadn't heard of them, I explained the technology and told them it aligns extremely well with what they are trying to develop. He agreed and said he would look into it but also noted their main goal for now is to simply get their first device to market. They already had their funding locked away and plans in place for the development cycle, pivoting was out of the question given how far into development they were. This conversation was back in early 2022 so I didn't know it at the time but this attitude and existing capital commitment problem is an issue brainchip is currently dealing with. In other words, Brainchip's technology sounds great and is something they could consider for their next product, but they need to finish what they've started first.

I did offer to make an introduction to Brainchip for Michael and I think I might have sent an email to Adam Osseiran introducing the two to each other but I'm a little hazy on whether that was something I said I'd do but not sure if I actually did.

I've been watching their story closely because I feel invested given I worked with them for years, to my knowledge, their first product still hasn't been commercialised so I'm of the opinion akida isn't involved yet. You are correct though, the harmony is there and akida definitely has a place in their technology.

I think the development of their sensor is taking longer than they had planned. By coincidence, my Fiances step brother works for a company in Sydney that designs chips. We realised in discussion that we both had a link to their company through our work. He wasn't on the team working on VitalTrace's product design but did say the guys he works with has said what they were trying to build was not going to be easy.

Another Perth based Med Tech that has an eerily similar story to VitalTrace is a company called VeinTech. They are developing a product called the VeinWave which helps to find veins and therefore reduce failed canulation rates which are woeful and cost the hospital system hundreds of millions annually in consumable and nurses/doctors time and therefore wages etc. I know Nick Buckley (CEO) from VeinTech through my professional network, having arranged for him to speak to a tech group I was involved with a couple of years ago. The VeinWave could also benefit from some akida magic, and similarly, is likely to be more of a "maybe next time".

I might reach out to Nick and Michael and see if they are keen for a catch up soon.

Cheers

Hi Sera2g,

wow, thank you so much for all that valuable background info!

Just wanted to highlight the following part of your post (especially for those readers who are not interested in dot-joining and may therefore have skipped my post and your reply to it), relating to a factor a lot of us retail shareholders with no relevant industry background keep underestimating:

“They already had their funding locked away and plans in place for the development cycle, pivoting was out of the question given how far into development they were. This conversation was back in early 2022 so I didn't know it at the time but this attitude and existing capital commitment problem is an issue brainchip is currently dealing with. In other words, Brainchip's technology sounds great and is something they could consider for their next product, but they need to finish what they've started first.”

Indeed, VitalTrace are almost two thirds into their ongoing research collaboration with Monash and UWA, and will start another collaboration on this project with UTS in 2024, but it is still good to know they will definitely have us on their radar then for DelivAssure 2.0 or any future products…

View attachment 49881

 

[Easytiger]

FYI , Just to be clear, Re Sean and others, being able to purchase shares. At the recent Perth gathering , Sean acknowledged the difficulty of being allowed to purchase shares , since Black Out periods would be so extensive,as a result of their current privileged inside information.

Simples

Not sure how this stacks up with the company response to the ASX speeding ticket.

 

[MDhere]

Ok when sorting out my emails I found a section where I had been trying to find a report from wevolver after I heard a brainchip podcast in late July / early Aug mention it but I could never get my hands on it as I assumed it was only avail by subscription. Well bowl me over, today I was able to finally get my hands on it - Here is a snippet of it! -

2023 Edge AI Technology Report. Chapter X: Future of Edge AI​

What’s Next for Artificial Intelligence and Edge Computing?​

​

Event-based Processing & Learning: BrainChip’s Neuromorphic AI Solution​ BrainChip is one of the pioneers of bringing neuromorphic computing to the edge. While traditional neuromorphic approaches have used analog designs to mimic the neuron and synapse, BrainChip has taken a novel approach on three counts. Firstly, their design is a fully digital design that is portable and reliable. Secondly, not only do they support spiking neural nets, but they have applied event-based execution to traditional convolutional networks, thereby rendering neuromorphic computing mainstream today. This allows current CNN/RNN models to run much more efficiently and drives far more capable performance on extremely low-footprint, low-power devices at the sensor. Thirdly, delivering on-device learning allows for personalization, customization, and other learning untethered from the cloud. Brainchip’s Akida neural processor is offered as IP and is configurable from energy-harvesting applications at the sensor edge to high-performance yet power-efficient solutions at the network edge. It is sensor-agnostic and has been demonstrated on a variety of sensors. As a self-managed neural processor that executes most networks completely in hardware without CPU intervention, it addresses key congestion and system bandwidth challenges in embedded SoCs while delivering highly efficient performance. With support for INT8 down to INT1 and skip connections, it handles most complex networks today, along with spiking neural nets. This led NASA to select BrainChip’s first silicon platform in 2021 to demonstrate in-space autonomy and cognition in one of the most extreme power- and thermally-constrained applications. Similarly, Mercedes Benz demonstrated BrainChip in their EQXX concept vehicle that can go over 1000 km on a single charge. In the latest generation, Brainchip has taken another big step of adding Temporal Event Based Neural Nets (TENNs) and complementary separable 3D convolutions that speed up some complex time-series data applications by 500x while radically reducing model size and footprint, but without compromising accuracy. This enables a new class of compact, cost-effective devices to support high-res video object detection, security/surveillance, audio, health, and industrial applications. While neuromorphic computing is still discussed as a future paradigm, BrainChip is already bringing this paradigm to market.