BRN Discussion Ongoing

[Damo4]

Quick question. The original roadmap shows Akida 2000 as being "an optimised version of AKD 1500 for LSTM and transformers". We now know that Akida 2000 includes the transformer part, but what happened to the LSTM part? Or have LSTM's been replaced with TNN's? I tried to google information on TNN's but kept getting articles about tennis, which was entertaining but not particularly helpful. But then I discovered something that @TechGirl posted from Carnegie Mellon University which describes TNN's as follows:

Processor Architecture: Temporal Neural Networks (TNN)
Temporal Neural Networks (TNNs) are a special class of spiking neural networks, for implementing a class of functions based on the space time algebra. By exploiting time as a computing resource, TNNs are capable of performing sensory processing with very low system complexity and very high energy efficiency as compared to conventional ANNs & DNNs. Furthermore, one key feature of TNNs involves using spike timing dependent plasticity (STDP) to achieve a form of machine learning that is unsupervised, continuous, and emergent.

View attachment 31412

Yeah I think we need a new roadmap, potentially something updated on the website rather than an archived screenshot.
It would be great if the specify the exact naming of current and future products in this too.
I think the E, S and P is a great idea, but it would be nice to either align it with previous naming or just start form scratch and keep it consistent.


BTW I've noticed a little chatter questioning the usefulness of AKD1000, but I believe this will increase sales, not serve as a replacement.
It's clear some customers need the extra features, but as we become ubiquitous, we will attract eyes who may not understand the use cases and see it more as a choice of which model.

I recall reading something about Coke and Pepsi vending machines next to each other increasing sales for both.
The idea is that a consumer no longer chooses whether to purchase a drink or not, instead they are left to choose which brand they would prefer.
I'd like to think the increased brand awareness as well as a true choice of nodes, efficiencies, performance etc, increase sales across the board.

 

[HopalongPetrovski]

Quick question. The original roadmap shows Akida 2000 as being "an optimised version of AKD 1500 for LSTM and transformers". We now know that Akida 2000 includes the transformer part, but what happened to the LSTM part? Or have LSTM's been replaced with TNN's? I tried to google information on TNN's but kept getting articles about tennis, which was entertaining but not particularly helpful. But then I discovered something that @TechGirl posted from Carnegie Mellon University which describes TNN's as follows:

Processor Architecture: Temporal Neural Networks (TNN)
Temporal Neural Networks (TNNs) are a special class of spiking neural networks, for implementing a class of functions based on the space time algebra. By exploiting time as a computing resource, TNNs are capable of performing sensory processing with very low system complexity and very high energy efficiency as compared to conventional ANNs & DNNs. Furthermore, one key feature of TNNs involves using spike timing dependent plasticity (STDP) to achieve a form of machine learning that is unsupervised, continuous, and emergent.

View attachment 31412

I think it likely that as with Akida 1000, changes as to what is to be released, when, were made in line with customer/user feedback.
Tweaking of attributes that will get us commercial deals done in the quickest time possible and in the light of recent developments.
I'm sure that the Company is aware that it needs to be making money ASAP and then, when that pressure is somewhat vented, can resume its orderly roll out and structured evolution. All just IMO.

 

[Dijon101]

while we are all patting ourselves on the back about this great news the sp has dropped 3.5c WTF

hahahahaha new to the game are you???


The news IS great.
Don't care what the share price does..


If you want to squabble about the day to day movement of the SP, you really should get out of this investing game.

 

[Damo4]

hahahahaha new to the game are you???


The news IS great.
Don't care what the share price does..


If you want to squabble about the day to day movement of the SP, you really should get out of this investing game.

Yeah or go to the moan and groan thread haha

 

[Vladsblood]

We need to get off the ASX ASAP. This why we have extremely low US investors. Today we had 150,000 shares traded, average is 40,000 or less. We will not get the US investor until we are listed on NASDAQ.

Agree with you Glen …..100 percent. They may be closer than 18 months out as they already have the preliminary work done to go full Nasdaq.
Our MC is still only a fraction of where it should now stand imo…Soon our unstoppable rerate will be underway heading towards tens of billions. Nothing will be able to stop Brainchips rise this year Vlad

 

[wasMADX]

WOW. From the new BRN website
"Brainchip’s holistic, distributed approach frees up the cloud and creates an explosion of capable, intelligent sensing devices on the Edge that accelerate global artificial intelligence."

 

[Vladsblood]

Our Customers will be our best advertising partners now as they preach Brainchips abilities to their customers. Vlad

 

[Rise from the ashes]

Our Customers will be our best advertising partners now as they preach Brainchips abilities to their customers. Vlad

Yes especially when they start to take away market share from the other companies in the same industry.

 

[TheDrooben]

Traders come and traders go. Let them play their games.

Right on cue ROFL!1
View attachment 31417

 

[AARONASX]

The Market fluctuation and shorting present does not, I repeat DOES NOT detract or hinder the day-to-day operation and the core productsss of the business, nor does it turn away customers of Brainchip. Nevertheless there will be a short squeeze most-likely in the near future...as there seem to be more and more positive chatter coming out from us and Brainchip's customers, take yesterday's announcement for example, I for one did not see this coming, and if they did it once they'll surprise me again, bigger and better than the last.

In part shorters at the moment rely on fear, no anns and negativity, but, Brainchip's Team are already proving time and time again they're ahead of the competition by a country mile!!!, where you as a company looking into AI would have to ask yourself wait maybe for something better (LOL) or get left behind. IMO

 

[skutza]

Sugar fix for traders yesterday. But wow, what great publicity we are seeing for this next stage of the company. Some big names coming out of the woodwork praising the company. Tick tock tick tock. What is neuromorphic computing anyway? Well the world is about to find out.

 

[skutza]

View attachment 31419

Only if you see it as a problem? I don't, just knowing the future is what I'm best at

Money, money, money.

Once it flows for the company, it'll flow for all it's shareholders, simple basic maths.

 

[Damo4]

Sorry if it's already been mentioned in the last 24hrs, but the partners tab of the website has been updated and is looking phat as:
Brainchip Partners

Technology Partners​

Technology partners improve vertical value with demonstrated pre-integration interoperability. Our neuromorphic IP is processor and OS independent. Let’s demonstrate the power of Essential AI together.
We work with our partners to support integration and market-relevant use cases.
BrainChip is a member of the Arm AI Partner Program, an ecosystem of hardware and software specialists enabling the next generation of intelligent AI solutions.

BrainChip is a partner with Intel Foundry Services – IP Alliance. Partners in this alliance collaborate with IFS to enable designers to access high-quality IPs, supporting their design needs and project schedule, while optimizing for performance, power and area. Building upon Intel’s advanced technology offering.

BrainChip partners with Prophesee driving the optimization for computer vision AI performance and efficiency to deliver next generation intelligent platforms for OEM’s looking to integrate event-based vision systems with high levels of AI performance coupled with an ultra-low power framework.

Integrating BrainChip’s Akida technology and SiFive’s multi-core capable RISC-V processors will provide a highly efficient solution for integrated edge AI compute.

Enablement Partners​

Enablement partners optimize the performance of their AI/ML development, modeling, or inference software with Akida integration. Pre-integration provides a vertically complete solution to simplify evaluation and implementation.
We work with our partners to create integrated AI solutions that are rich in capability with unparalleled performance.

The Minsky AI Engine from AI Labs and the ultra efficient sensory inference capabilities of BrainChip’s Akida™ provides a compelling and cost-effective solution in system health monitoring in industrial efficiency and productivity. Detecting anomalies, identifying challenges, analyzing impacts are easily addressed with AI Labs and BrainChip working seamlessly together.

Combining BrainChip’s Akida technology and the Edge Impulse platform, tools, and services allows current and future customers to achieve their ML objectives with fast and efficient development cycles to get to market quicker and achieve a competitive advantage.

BrainChip’s technology combined with emotion3D’s state-of-the-art computer vision and machine learning software for image-based analysis of in-cabin environments enables our mutual customers to achieve an ultra-low-power working environment with on-chip learning while processing everything locally on the device within the vehicle to ensure data privacy.

BrainChip and NVISO are targeting battery-powered applications in robotics and mobility devices addressing the need for high levels of AI performance in ultra low power environments. Implementing NVISO’s AI solutions with BrainChip’s Akida drive next generation solutions.

Integration Partners​

Integration partners utilize BrainChip’s technology through designing the Akida IP into System on Chip (SoC) products that will be into ready-to-use systems or implementing Akida silicon into ready-to-use modules.

Partnering with BrainChip, MegaChips is incorporating the Akida technology into it’s ASIC solutions service. Enabling the development and support required to design and manufacture integrated circuits and systems on chips with intelligence that will drive AI into the next generation of edge based devices.

Teksun focuses on end to end IoT product development and enabling intelligent solutions, such as predictive and preventative maintenance devices, analytics and diagnostics for portable healthcare, and vision based devices for security and surveillance. The partnership between BrainChip and Teksun proliferates intelligence through the Teksun product development channels.

University AI Accelerator Program​

BrainChip is bringing its neuromorphic technology into higher education institutions via the BrainChip University AI Accelerator Program, which shares technical knowledge, promotes leading-edge discoveries and positions students to be next-generation technology innovators.
BrainChip’s University AI Accelerator Program provides hardware, training, and guidance to students at higher education institutions with existing AI engineering programs. BrainChip’s products can be leveraged by students to support projects in any number of novel use cases or to demonstrate AI enablement. Students participating in the program will have access to real-world, event-based technologies offering unparalleled performance and efficiency to advance their learning through graduation and beyond.
Current university participants include:

By partnering with BrainChip’s AI Accelerator Program, universities are able to ensure that students have the tools and resources needed to encourage development of cutting-edge technologies that will continue to usher in an era of essential AI solutions.
Have your university become AI smarter, join the University AI Accelerator Program today.

 

[Evermont]

A lot to digest today. This one caught the eye.

 

[Rise from the ashes]

Only if you see it as a problem? I don't, just knowing the future is what I'm best at

Money, money, money.

Once it flows for the company, it'll flow for all it's shareholders, simple basic maths.

Once the kraken gets released we are going to have a huge re-rate. Then most of us long-term holders won't complain about a few cents here or there. I look forward to the days when it drops 50 cents and we just giggle about it

 

[TheDrooben]

Only if you see it as a problem? I don't, just knowing the future is what I'm best at

Money, money, money.

Once it flows for the company, it'll flow for all it's shareholders, simple basic maths.

I agree. Just don't know what you are trying to imply about my post regarding the SP gap. If you are trying to imply that I am posting it for my own personal gain then you are dead wrong. I have been buying when I can since 2020 and have never sold a single share.
The point in my post was that when TA traders and shorters see that gap they will use it as a reason to push it to that price. Without further news it is a good possibility.

Larry

 

[Fact Finder]

Anyone who has not read the Forbes article make time. It confirms that AKIDA ESP offers 1,2,4&8 bit activations and much much more.

The last two paragraphs inspired me to break my golden rule and buy more BRN:

“Brainchip’s bio-inspired Akida platform is certainly an unusual way to tackle AI/ML applications. While most other NPU vendors are figuring out how many MACs they can fit – and power – on the head of a pin, Brainchip is taking an alternative approach that’s been proven by Mother Nature to work over many tens of millions of years.

In Tirias Research’s opinion, it’s not the path taken to the result that’s important, it’s the result that counts. If Brainchip’s Akida event-based platform succeeds, it won’t be the first time that a radical new silicon technology has swept the field. Consider DRAMs (dynamic random access memories), microprocessors, microcontrollers, and FPGAs (field programmable gate arrays), for example. When those devices first appeared, there were many who expressed doubts. No longer. It’s possible that Brainchip has developed yet another breakthrough that could rank with those previous innovations. Time will tell.”

My mistake and opinion only so DYOR
FF

AKIDA BALLISTA

PS: Just love the “head of a pin comment” wish I had said it.

Anyone for some performance figures from the same Forbes article:


"Brainchip offers three Akida IP variants: E, S, and P:

E. The Max Efficiency variant (Akida-E) provides as many as four nodes (or 16 NPUs), runs as fast as 200MHz, delivers the equivalent of 200 GOPS (giga-operations per second) of performance, and needs only milliwatts of power for operation. The company says that this smaller variant is designed for running simpler AI/ML networks and is intended for use in continuously operating equipment where power consumption is at a premium.

S. The Sensor Balanced variant (Akida-S) can be configured with as many as eight nodes (32 NPUs), runs as fast as 500MHz, and delivers the equivalent of 1 TOPS (trillion operations per second) of performance, which is capable of running object detection and classification workloads.

P. The Performance variant (Akida-P) accommodates as many as 128 nodes (512 NPUs), operates as fast as 1.5GHz, and delivers the equivalent of 50 TOPS. The most capable version of the Performance variant includes optional hardware support for vision transformer networks that take the form of additional nodes in the internal Akida mesh network.

The high-end variant can run the full gamut of AI/ML models in Brainchip’s model zoo to perform tasks including classification, detection, segmentation, and prediction. Together, these Akida variants allow a design team to use one AI/ML architecture that scales from low-power configurations that consume mere microwatts of power to high-performance configurations that deliver dozens of TOPS, and according to BrainChip, can perform HD video object detection at 30 frames per second while consuming less than 75 milliwatts, which could result in very compelling portable vision solutions."

Talk about a versatile solution package.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Diogenese]

Anyone for some performance figures from the same Forbes article:


"Brainchip offers three Akida IP variants: E, S, and P:

E. The Max Efficiency variant (Akida-E) provides as many as four nodes (or 16 NPUs), runs as fast as 200MHz, delivers the equivalent of 200 GOPS (giga-operations per second) of performance, and needs only milliwatts of power for operation. The company says that this smaller variant is designed for running simpler AI/ML networks and is intended for use in continuously operating equipment where power consumption is at a premium.

S. The Sensor Balanced variant (Akida-S) can be configured with as many as eight nodes (32 NPUs), runs as fast as 500MHz, and delivers the equivalent of 1 TOPS (trillion operations per second) of performance, which is capable of running object detection and classification workloads.

P. The Performance variant (Akida-P) accommodates as many as 128 nodes (512 NPUs), operates as fast as 1.5GHz, and delivers the equivalent of 50 TOPS. The most capable version of the Performance variant includes optional hardware support for vision transformer networks that take the form of additional nodes in the internal Akida mesh network.

The high-end variant can run the full gamut of AI/ML models in Brainchip’s model zoo to perform tasks including classification, detection, segmentation, and prediction. Together, these Akida variants allow a design team to use one AI/ML architecture that scales from low-power configurations that consume mere microwatts of power to high-performance configurations that deliver dozens of TOPS, and according to BrainChip, can perform HD video object detection at 30 frames per second while consuming less than 75 milliwatts, which could result in very compelling portable vision solutions."

Talk about a versatile solution package.

My opinion only DYOR
FF

AKIDA BALLISTA

Just imagine being one of the students at the chosen universities - getting hands-on experience with the future of AI ... and it won't do their employment prospects much harm either.

 

[Bravo]

SiFive has design wins with more than 100 customers, including 8 of the top 10 semiconductor companies, in applications including automotive, AR/VR, client computing, datacenter, and the intelligent edge.

So, I guess this means many of their customers will become ours in due course.

Oh, and speaking of Si-FIve customers, here's what Ziad Asghar, VP of Product Management for Snapdragon Technology and Roadmap at Qualcomm said in a statement in November 2022.

Baring in mind that Prophesee loves us and would no doubt be pushing for our inclusion in to Snapdragon platforms. And SiFive loves us too as can be attested in the latest statement from Phil Dworsky (Global Head of Strategic Alliances at SiFive).

So, come on Qualcomm, just go with the flow, because resistance is literally futile.

Compact RISC-V Cores Bring the Power to Wearables, Consumer Devices

With its latest 64-bit RISC-V CPU cores, the P670 and P470, SiFive is putting the focus on compute density.

www.electronicdesign.com

 

[perceptron]

Totally agree. All bases covered. Mind is racing thinking about the possibilities that will be created. Would be great to see videos of the new platform in action ie: the always on function, through to the most suitable applications for AKIDA-P and it's 512 NPU's.

Anyone for some performance figures from the same Forbes article:


"Brainchip offers three Akida IP variants: E, S, and P:

E. The Max Efficiency variant (Akida-E) provides as many as four nodes (or 16 NPUs), runs as fast as 200MHz, delivers the equivalent of 200 GOPS (giga-operations per second) of performance, and needs only milliwatts of power for operation. The company says that this smaller variant is designed for running simpler AI/ML networks and is intended for use in continuously operating equipment where power consumption is at a premium.

S. The Sensor Balanced variant (Akida-S) can be configured with as many as eight nodes (32 NPUs), runs as fast as 500MHz, and delivers the equivalent of 1 TOPS (trillion operations per second) of performance, which is capable of running object detection and classification workloads.

P. The Performance variant (Akida-P) accommodates as many as 128 nodes (512 NPUs), operates as fast as 1.5GHz, and delivers the equivalent of 50 TOPS. The most capable version of the Performance variant includes optional hardware support for vision transformer networks that take the form of additional nodes in the internal Akida mesh network.

The high-end variant can run the full gamut of AI/ML models in Brainchip’s model zoo to perform tasks including classification, detection, segmentation, and prediction. Together, these Akida variants allow a design team to use one AI/ML architecture that scales from low-power configurations that consume mere microwatts of power to high-performance configurations that deliver dozens of TOPS, and according to BrainChip, can perform HD video object detection at 30 frames per second while consuming less than 75 milliwatts, which could result in very compelling portable vision solutions."

Talk about a versatile solution package.

My opinion only DYOR
FF

AKIDA BALLISTA