BRN Discussion Ongoing

[Dang Son]

That's right FF, it was beautifully transcribed by me

Prophesee was so ahead of the game in tech that current (pre-BrainChip) tech was limiting their abilities to really shine & then all of a sudden a young sexy BrainChip walked through the door, Prophesee's eyes lit up & it was a match made in technology Heaven.

To me after listening to Luca, Prophesee will naturally choose BrainChip every single time it creates anything new, as why would they go forward & sell a half product to a customer when it can add BrainChip & sell a full better product to a customer, it's the "natural choice".

Buy Products | PROPHESEE

Experience Prophesee Metavision, with a range of Event-Based Vision products: event cameras, Evaluation Kits, Metavision sensors and Metavision software.

www.prophesee.ai

 

[MDhere]

That’s a very intriguing response, in fact I’d say that’s a non-denial denial. He could have said a flat no but, instead said ‘not that one’. So he didn’t say an out and out no, but just a no to what we’re alluding to being Akida 1.0. This way he’s being evasive but not lying.

My guess its Akida 1.5 with LSTM.

Before 1000 is 500 n'est pas?

 

[DingoBorat]

I don’t believe him.
The fact that there would be restrictions/NDA’s on what can and cannot be said regarding the worlds best kept secret (akida) would suggest to me that he has his poker face on!

Call me naive but I just cannot believe him with the relationship we have with Mercedes, and what they’ve already said about Brainchip publicly being the AI experts and it being up to 10x more efficient than anything else out there, and now we know that akida new and improved is due out soon, publicly anyway and that somebody has all of a sudden come up with something much more superior to Akida?

NOPE not a chance!

Wouldn't it be easier, to just not reply to a random poster, who was posting a random comment, that wasn't even posed as a question?

Seems odd and Mercedes, haven't been shy about us recently..

Definitely curious..

 

[H.E. Pennypacker]

Sorry, are you lost? Can someone point this dude back to the crapper?

How have you escaped @zeebot's moderation for this long?

Moderate me for what exactly? It’s this mentality that has stopped great posters like Uiux posting here

 

[Fact Finder]

Holy shit balls, Batwoman!

Yes and yet it does not dominate the conversation here???

Why is this so???

My opinion only DYOR
FF

AKIDA BALLISTA

 

[DingoBorat]

Mercedes is not a brand that chops and changes. It finds/invents the best technologies and work with them for the long term. That's what their F1 involvement is about for example. Now if we accept that, then the fact they have recently stated that Brainchip are the AI experts and they are working with us to bring to production the kinds of systems deployed in the EQXX it follows that the tweet could not be about anything other than Brainchip product - by whatever name. Think about it. Would Mercedes trumpet Brainchip as the best AI they can find then within months go off in another direction?????

Yes, if something better came along, they would.
There is no such thing as loyalty, when it comes to technology.
If BrainChip IP is already being incorporated though, it wouldn't be a sudden change..
(If this is even direct competition)

 

[Iseki]

the delusion around these threads sometimes is mind blowing. He confirms it’s NOT Akida but you choose not to believe him and want to convince yourself and others that it is????

If you read his posts (not sure the platform) the guy (Francois) is looking at something in the context of some up-coming award where your secretary can nominate you. ie he is not looking at it on behalf of MB. He is not looking at a patent.

 

[Quiltman]

Thanks for posting Jesse,

Holy Shit Balls that was a fantabolus podcast

Now bear with me this is all from my dodgy memory.

"we are very excited about collaboration with BrainChip, we are natural partners, before Brainchip they were facing integration problems, which was impacting sensor performance, so now combining Prophesee & Brainchip we are finishing the half story we were missing from the beginning, it is now complete, so now we can tell a full story to our customers.....WOW"

Three examples of use cases that Prophesee are currently working on & some I think he said they have been working on for the last 3 years are;

1. Industrial application - High speed fast accurate barcode detection on fast moving boxes

2. Automotive application - Event based camera state of the art object detection for sensors in cars to detect pedestrians or other objects in poor lighting & then for example applying the brakes

3. Mobile Application - Computational imaging for smart phone, our event based sensor will combine with regular frame based sensor to take best images, we capture moving object & we can help correct blurry images that would be taken with just a regular frame based sensor



This is what Prophesee thinks of Akida

And this is what BrainChip thinks about Prophesee

And this is what I think of our Partnership

We have all seen the Prophesee partners page ....
We are integral to their solution moving forward ... hence, we will be integrated with their partners in delivering commercial solutions.
These guys & MegaChips ... why is the SP still under $1.00 .... won't be for long !

 

[HopalongPetrovski]

Moderate me for what exactly? It’s this mentality that has stopped great posters like Uiux posting here

Uiux is free to post here, as are you.
As this is Zeeb0t's creation, He is the only one with the power to moderate anyone.
He has however asked us all to help him by pointing out possible infringements of his rules, which some do more zealously than other's,
as is they're bent.
Some individuals occasionally get over excited and begin to act as if they may dictate what other's post, or what is suitable, or what is politically correct according to some personal version of truth or morality held.
We are stronger as 1000 eye's when disparate and wide ranging views and opinions are able to be displayed as this can lead to interesting cross pollination and a strengthened overall viewpoint of our company from a multidisciplinary technical/corporate/ethical/financial/governance/research viewpoint. Just another anonymous opinion.

AKIDA BALLISTA
AKIDA'S EVERYWHERE
GLTAH

 

[TheFunkMachine]

Anil Mankar made these comments at the 2021 Ai Field Day regarding GPU’s:

“Because (we) do computations neuromorphic it is a clear advantage over deep learning GPUs…a quick comparison between (our) neuromorphic chip…and GPUs…(we) can work in microwatts to milli watts while deep learning GPUs are using 100’s of watts…(our) neuromorphic chip… can run much faster, we are running at 300Mhz but the circuits can go to 1Ghz…”

Anil Mankar then goes on to speak about the fact that AKIDA is like the human brain in that it can learn.

So Mercedes’ Benz IP pictures of brains you would have to be a WANCA to not be able to work out these clues all add up to the only logical choice it’s AKIDA.

My opinion only DYOR
FF

AKIDA BALLISTA

I read someone else alluding this to potentially be Synsense and not Akida. Have we explored that further? Does the synsense tech boast 100x power efficiency over GPUs ?

If this is Not Akida, then it is cause for concern, especially as we know Mercedes has boasted 5-10x power savings on their “hey Mercedes” applications.

In my opinion this has to be Akida as I have t heard of anyone else boasting these kind of results, but I don’t want to ignore potential competitors if there is any.

 

[Fact Finder]

Yes, if something better came along, they would.
There is no such thing as loyalty, when it comes to technology.
If BrainChip IP is already being incorporated though, it wouldn't be a sudden change..

Hi @DingoBorat

There is a statement by Peter van der Made that AKIDA is 1,000 times more efficient than GPUs. Not sure where it is now but its out there somewhere to be found.

Prophesee has been looking for the event based processor to match the performance of their event based sensor and have now stated via this podcast and their CEO that AKIDA has provided the missing piece of their puzzle. Clearly not SynSense who runs around claiming all sorts of things about itself uninhibited by the ASX or SEC rules.

This fellow who not one of us ever mentioned before yesterday posts one random claim about an unidentified IP as being 100 more time efficient than a GPU and when asked if it is AKIDA says not that one.

Well the first thing to say is according to Peter van der Made AKIDA is 1,000 times more efficient so 10 times better than what is being described.
Second thing is that someone else I think @Cardpro posted other posts of this chap where he is saying Moores Law is not dead and conventional technology has a long way to go before it is dead.

Then @Bravo found and posted about a new clock technology that improves the performance of a GPU 100 times and references RISC-V technology as being made more efficient.

Having reflected on @Bravo 's posted information I am of the mind that this is what this poster was referencing however as SiFive adds AKIDA to its RISC-V all this new clock technology does is make more efficient the whole system which includes AKIDA it does not replace AKIDA.

So whether it is @Bravo 's discovery or whether it is AKIDA2.0 it in no way changes the importance of the Prophesee statements concerning why they are going forward with AKIDA in Mobile Phone vision systems, Industrial vision systems and Automotive vision systems.

Indeed all of the concerns that some have raised about AKIDA being dropped by Mercedes Benz for better technology are real except they are real for SynSense over at BMW where Prophesee has its foot in the door and on todays presentation the only solution they would be moving forward with is one where they are utilising AKIDA in vision systems for automotive.

If anything the Prophesee podcast should have everyone asking will this increase the use of AKIDA in Mercedes Benz as they will no doubt be introducing Mercedes Benz to what the combined AKIDA Prophesee vision sensor is capable of achieving for them.

Perhaps Brainchip is also over at Valeo introducing the Prophesee AKIDA Vision System for Valeo to consider as yet another automotive sensor technology to combine with its present offerings.

I have an enormous sense of frustration with the continuing discussion of one random engineer who makes claims about testing he has not even completed. For all we know this other technology if it is not AKIDA could be like Elon Musks Chip and needs 500 watts of power to achieve this 100 times improved performance.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[alwaysgreen]

In other news, how good is the share price action today!

 

[ndefries]

In other news, how good is the share price action today!

Almost getting upto where it was before 10M shorts taken out - that have not as yet been bought back

 

[MDhere]

"Hey fellow brners"

Let me make this clear - Hey Mercedes - is us. IMO the below uses Hey Mercedes and i agree its a hot word

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.

This has come including the bold words, directly from https://group-media.mercedes-benz.com/

of course IMO

 

[buena suerte :-)]

Hi @DingoBorat

There is a statement by Peter van der Made that AKIDA is 1,000 times more efficient than GPUs. Not sure where it is now but its out there somewhere to be found.

Prophesee has been looking for the event based processor to match the performance of their event based sensor and have now stated via this podcast and their CEO that AKIDA has provided the missing piece of their puzzle. Clearly not SynSense who runs around claiming all sorts of things about itself uninhibited by the ASX or SEC rules.

This fellow who not one of us ever mentioned before yesterday posts one random claim about an unidentified IP as being 100 more time efficient than a GPU and when asked if it is AKIDA says not that one.

Well the first thing to say is according to Peter van der Made AKIDA is 1,000 times more efficient so 10 times better than what is being described.
Second thing is that someone else I think @Cardpro posted other posts of this chap where he is saying Moores Law is not dead and conventional technology has a long way to go before it is dead.

Then @Bravo found and posted about a new clock technology that improves the performance of a GPU 100 times and references RISC-V technology as being made more efficient.

Having reflected on @Bravo 's posted information I am of the mind that this is what this poster was referencing however as SiFive adds AKIDA to its RISC-V all this new clock technology does is make more efficient the whole system which includes AKIDA it does not replace AKIDA.

So whether it is @Bravo 's discovery or whether it is AKIDA2.0 it in no way changes the importance of the Prophesee statements concerning why they are going forward with AKIDA in Mobile Phone vision systems, Industrial vision systems and Automotive vision systems.

Indeed all of the concerns that some have raised about AKIDA being dropped by Mercedes Benz for better technology are real except they are real for SynSense over at BMW where Prophesee has its foot in the door and on todays presentation the only solution they would be moving forward with is one where they are utilising AKIDA in vision systems for automotive.

If anything the Prophesee podcast should have everyone asking will this increase the use of AKIDA in Mercedes Benz as they will no doubt be introducing Mercedes Benz to what the combined AKIDA Prophesee vision sensor is capable of achieving for them.

Perhaps Brainchip is also over at Valeo introducing the Prophesee AKIDA Vision System for Valeo to consider as yet another automotive sensor technology to combine with its present offerings.

I have an enormous sense of frustration with the continuing discussion of one random engineer who makes claims about testing he has not even completed. For all we know this other technology if it is not AKIDA could be like Elon Musks Chip and needs 500 watts of power to achieve this 100 times improved performance.

My opinion only DYOR
FF

AKIDA BALLISTA

BrainChip founder and CTO Peter van der Made presented at Stocks Down Under’s Semiconductor Conference on 30 November 2021.

So if you’re looking at our products, the Akida 1000 is the major product, which is the chip. The neural multiprocessor mimics…and I’ll look it up in the dictionary, it mimics neurobiological architecture present in the neural nervous system. So it doesn’t work like a processor that is applied in your laptop, it works like a mini-brain. It is very good at recognizing things, recognizing smells, recognizing images, recognizing tastes. We have done all these things as examples and made them available to the market. We also have an artificial retina image to spike converter. What the eye is doing, it takes images, converts those images into spikes that then the brain is processing. We do something very similar with our artificial retina image to spike converter.

The chip is up to 10 times more energy-efficient than any other edge AI processor. Except if you’re looking at very tiny things that can’t be really used for anything. But anything that isn’t the same class as the BrainChip Akida processor is using at least 10 times more energy here. And it’s up to a thousand times more energy-efficient than GPUs.

 

[Bravo]

Re the Francois Piednoel post, I was curious to find out more about Movellus because of the comment from Aakash Jani (Marketing & Brand Growth at Movellus). I'm pretty sure it was the wink that motivated me to take a closer look.

Anyhoo, Movellus describes itself in these terms "Movellus solves SoC-level clock distribution challenges in high performance and ultra low power chips. The company’s flagship platform, Maestro™ is a clock distribution network platform that delivers chip-level architectural innovations to improve SoC performance while reducing power consumption. All-digital, fully synthesizable clocking solutions achieve quantum leaps in energy efficiency. This accelerates the move to carbon-neutral data centers and enables ultra-low voltage operation for edge AI, IoT devices, and radiation hardened satellites".

What's interesting is that Movellus have worked with the following big names: Syntiantcorp, MosChip, SiFive, Intel Foundary Services, Esperanto, BAE Systems, etc.

I also noticed an Esperanto announcement, that they had adopted the Movellus intelligent clocking solution. I also noticed it said 100x better energy efficiency, which was very similar to what Francois Piednoel was talking about in the IP he was trialing, which leads me to wonder if it was the Esperanto chip he was referring to.



View attachment 18046

Esperanto Technologies Adopts Movellus Maestro AI, Intelligent Clock Networks for Its ET-SoC-1 Chip

Esperanto Technologies Adopts Movellus Maestro AI, Intelligent Clock Networks for Its ET-SoC-1 Chip News

www.design-reuse-embedded.com

Let's say for arguments sake that Francoise was talking about the new Esperanto chip. Do you think they could be a competitor or is there a possibility that we could be complementary to each other? I say this because of Esperanto's links with SiFive.

In another article by Sally Ward-Foxton it stated:

“BrainChip can run [AI] algorithms on their own, but when they move into a larger system, they will need a host processor,” Chris Jones, vice president, product at SiFive, told EE Times. “You could pick a host processor that does nothing but scheduling, or you could pick a host processor that actually contributes to the AI processing, and that’s where the SiFive Intelligence product comes in.”

In an SoC design for edge AI, the AI workload would typically be split between host processor, vector processor, and AI accelerator — some parts of edge workloads are better suited to general purpose compute rather than a dedicated AI accelerator, Jones said.

This 1,000-core RISC-V processor is generating buzz in the AI space​

By Anthony Spadafora
published April 26, 2022
New AI chip hold promise for machine learning recommendation

(Image credit: Esperanto Technologies)


A new 1,000-core RISC-V processor from Esperanto Technologies (opens in new tab) is currently being evaluated by Samsung SDS and other ‘lead customers”.

According to a press release (opens in new tab) from the computer software company, its new ET-SoC-1 AI Inference Accelerator is undergoing initial evaluations by a number of firms ahead of its release.

Esperanto Technologies itself was founded back in 2014 by semiconductor industry veteran Dave Ditzel who has previously worked at both Sun Microsystems and Intel. The company is now led by President and CEO Art Swift who in addition to being the former CEO of Wave Computing also spent some time working at the RISC-V Foundation.

As part of its evaluation program, Samsung SDS (opens in new tab) and other potential customers will get a chance to obtain performance data after running a variety of off-the-shelf AI (opens in new tab) models using the ET-SoC-1 AI Inference Accelerator and so far the results have been quite impressive.

ET-SoC-1 AI Inference Accelerator​

Esperanto’s ET-SoC-1 features 1,088 energy-efficient, 64-bit bit processor cores that utilize the RISC-V (opens in new tab) instruction set architecture which is quickly becoming a viable alternative to those of both x86 and ARM. The company’s new chip also includes four high-performance RISC-V cores along with 160m bytes of on-chip SRAM as well as interfaces for flash memory and external DRAM.

What sets the ET-SoC-1 apart from similar chips is its speed together with its low-power requirements. While the chip can run any type of machine learning (opens in new tab) workload, Esperanto says that it excels at machine learning recommendation which is used by Meta, Amazon and other hyperscalers.

VP of AI at Samsung SDS, Dr. Patrick Bangert provided further insight on the experience the company’s data science team had when evaluating the ET-SoC-1, saying:

“Our data science team was very impressed with the initial evaluation of Esperanto’s AI acceleration solution. It was fast, performant and overall easy to use. In addition, the SoC demonstrated near-linear performance scaling across different configurations of AI compute clusters. This is a capability that is quite unique, and one we have yet to see consistently delivered by established companies offering alternative solutions to Esperanto.”

While Esperanto has given Samsung SDS and other potential customers a chance to test out its new AI chip (opens in new tab), we’ll have to wait and see if the evaluation was impressive enough for orders to start coming in.

This 1,000-core RISC-V processor is generating buzz in the AI space

New AI chip hold promise for machine learning recommendation

www.techradar.com

 

[TechGirl]

We have all seen the Prophesee partners page ....
We are integral to their solution moving forward ... hence, we will be integrated with their partners in delivering commercial solutions.
These guys & MegaChips ... why is the SP still under $1.00 .... won't be for long !

View attachment 18054

I never get sick of looking at that Partners page

 

[Fact Finder]

BrainChip founder and CTO Peter van der Made presented at Stocks Down Under’s Semiconductor Conference on 30 November 2021.

So if you’re looking at our products, the Akida 1000 is the major product, which is the chip. The neural multiprocessor mimics…and I’ll look it up in the dictionary, it mimics neurobiological architecture present in the neural nervous system. So it doesn’t work like a processor that is applied in your laptop, it works like a mini-brain. It is very good at recognizing things, recognizing smells, recognizing images, recognizing tastes. We have done all these things as examples and made them available to the market. We also have an artificial retina image to spike converter. What the eye is doing, it takes images, converts those images into spikes that then the brain is processing. We do something very similar with our artificial retina image to spike converter.

The chip is up to 10 times more energy-efficient than any other edge AI processor. Except if you’re looking at very tiny things that can’t be really used for anything. But anything that isn’t the same class as the BrainChip Akida processor is using at least 10 times more energy here. And it’s up to a thousand times more energy-efficient than GPUs.

Thank you @ BienSuerte

This mini panic by some is but one more example of why the CEO Sean Hehir went to lengths at the AGM to educate shareholders to the sad fact that out there many, many companies are making false claims about their efficiency which are in no way shape or form close to the power and performance advantages offered by AKIDA.

Given the statement today by Luca Verre CEO of Prophesee TWO more pretenders have been exposed.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[MDhere]

But i dont think we are limited to the Hey Mercedes. I read that they have sensors that only act on a spikes which saves car energy, so basically I believe that the smart sensors in the vehicles will be equipped with Akida, Thats my thoughts and im sticking to it.

 

[TechGirl]

We have all seen the Prophesee partners page ....
We are integral to their solution moving forward ... hence, we will be integrated with their partners in delivering commercial solutions.
These guys & MegaChips ... why is the SP still under $1.00 .... won't be for long !

View attachment 18054

Plus lets not forget Prophesee's new partners

Prophesee Showcases Event Based Vision Partners at VISION 2022

The 2021 VISION Award winner is back with a powerful suite of hardware and software sensing solutions and new announcements with its fast-growing partner’s ecosystem including CIS, Datalogic, Framos, Lucid and MVTec.

www.prophesee.ai

Prophesee Showcases the Solutions and Partners Driving Event Based Vision at VISION 2022​

​

The 2021 VISION Award winner is back with a powerful suite of hardware and software sensing solutions and new announcements with its fast-growing partner’s ecosystem including Century Arks, CIS, Datalogic, Framos, Lucid and MVTec.​

PARIS – October 3, 2022 – The full extent of the power and accessibility of Prophesee’s advanced neuromorphic vision systems will be on display at the 2022 VISION show, with an array of technology showcases, partner demonstrations and live interactions that showcase the company’s breakthrough event-based vision approach to machine vision.

Prophesee will have in-depth product demonstrations at its Booth 8B29 at the premier industry event happening October 4-6 in Stuttgart, Germany.

Prophesee will also deliver an overview of its event-based platform on Wednesday, 6 October at 12:20 (Hall 8 Booth C70), entitled “The World Between Frames; Event Camera Come of Age.”

Secure your private meeting today for in-depth insights from our experts: http://prophesee.ai/meet

Prophesee’s Metavision® platform has gained traction with leading developers of machine vision systems in industrial automation, robotics, automotive and IoT. Its event-based vision approach significantly reduces the amount of data needed to capture information. Among the benefits of the sensor and AI technology are ultra-low latency, robustness to challenging lighting conditions, energy efficiency, and low data rate.

The company’s breakthrough sensors, accompanying development tools, open-source algorithms and models are at the foundation of several announcements and demonstrations at the show, including:

• CENTURY ARKS: Century Arks announces it will release mid-October the SilkyEvCam HD, first commercial Event-based Vision camera featuring the IMX636MP sensor realized in collaboration between Sony and Prophesee.
• CIS: Prophesee announces collaboration with CIS, a leading machine vision provider, to build the first Event-Based 3D sensing evaluation platform, leveraging advanced VCSEL technology: https://www.prophesee.ai/2022/10/03/cis-prophesee-structured-light-evaluation-kit/
• Datalogic: A global technology leader in the automatic data capture and factory automation markets begins landmark partnership with Prophesee to bring the performance and efficiency of neuromorphic vision to its industrial products: https://www.prophesee.ai/2022/10/03/datalogic-prophesee-event-camera-partnership/
• Framos: A leading global supplier of imaging products, custom vision solutions and OEM services is releasing their brand new Event-Based Vision development kit based on NVIDIA Jetson Xavier, featuring IMX636 sensor realized in collaboration between Sony and Prophesee.
• Lucid Vision Labs: Following early prototype announcement at VISION 2021, Machine Vision leader Lucid announces commercial availability of its new Triton EVS, featuring the Prophesee Metavision sensor inside.
• MVTec: Prophesee and MVTec Partner to Support Integration of Prophesee Event-Based Metavision® Cameras with MVTec HALCON Machine Vision Software. https://www.prophesee.ai/2022/10/03/mvtec-prophesee-halcon-integration-partnership/

“We are very pleased to see the fast-growing ecosystem around event-based vision and the variety of applications this powerful approach to machine vision can be used for. We have made great strides in the Machine Vision world even from last year when we were named best in show by VISION, and this year we shine the spotlight on not just our innovations but the broad range of use cases being enabled by our partners around the world,” said Luca Verre, CEO and co-founder of Prophesee.

ABOUT PROPHESEE​

Prophesee is the inventor of the world’s most advanced neuromorphic vision systems.
The company developed a breakthrough Event-Based Vision approach to computer vision. This new vision category allows for significant reductions of power, latency and data processing requirements to reveal what was invisible to traditional frame-based sensors until now. Prophesee’s patented Metavision® sensors and algorithms mimic how the human eye and brain work to dramatically improve efficiency in areas such as autonomous vehicles, industrial automation, IoT, security and surveillance, and AR/VR.
Prophesee is based in Paris, with local offices in Grenoble, Shanghai, Tokyo and Silicon Valley. The company is driven by a team of more than 100 visionary engineers, holds more than 50 international patents and is backed by leading international equity and corporate investors including 360 Capital Partners, European Investment Bank, iBionext, Intel Capital, Prosperity7 Ventures, Robert Bosch Venture Capital, Sinovation, Supernova Invest, Will Semiconductor, Xiaomi.
Learn more: www.prophesee.ai