BRN Discussion Ongoing

[stockduck]

Sorry if mentioned before....maybe worth for reading:

Cadence Tools Paving the Way for the AI-at-the-Edge Transition

For machines to sense and replicate human intelligence, we must empower them with quick decision-making abilities for reliable operations even during adverse conditions. AI at the edge reduces latency by processing data generated by sensors/IoT devic...

community.cadence.com

Cadence Tools Paving the Way for the AI-at-the-Edge Transition​

10 May 2023 • 7 minute read

TSMC Technology Roadmap, 2023 Version

<a href="https://community.cadence.com/cadence_blogs_8/b/breakfast-bytes"></a>April brings one of the two times during the year that TSMC lays out its process roadmap, fab construction plans, and more at the TSMC Technology Symposium. The other is the Open Innovation Platform Ecosystem Forum in...

community.cadence.com

".....
Specialty technology is increasingly important: RF, CIS, PMIC, NVM, ULP, Auto (key: RF=radios, CIS=image sensors, PMIC=power management, NVM=non-volatile memory, ULP=ultra low power, auto=automotive).

Here are a few process names:

RF: N6RF, N6RF+, N4PRF (coming). Many customer tapeouts last year.

ULP: N22, N12e, N6e (coming).

....

But TSMC also invests in new variations on those processes too. In fact, TSMC's investment in specialty technology has grown (over the last 5 years) at a CAGR of 40%. For the first time ever, it is investing in building mature technology capacity and will expand by 1.5X over the next three years (this counts the TSMC/Sony/Denso fab in Kumamoto, Japan, which is all mature technology).

• ULP solution. 55ULP, 40ULP, 22ULL (0.6-1V) N12e (0.45-1V), N6e.
• RF connectivity. N6RF offers fast er performance with 49% power consumption reduction compared to 16FFC enhancement (benchmark half analog, half digital).
• N4PRF most advanced RF CMOS technology. RF on top of logic for digital-intensive RF products. N4PRF vs N6RF logic density 1.77X, power reduction 45%
• MCU is all about how to scale non-volatile memory. MRAM and RRAM in 28nm/22nm since 2019. 16/12nmsince 2022, and 6nm in planning.
• 40RRAM And 28/22RRAM in volume production
• 28RRAM achieves >10 years 125° retention (automotive grade)
• 12RRAM to be released in 1Q2024
• 22MRAM in production
• 16MRAM consumer qualified in 2022, planned for automotive 2023.
• Achieve >1M endurance cycle, 20 years 230 degrees retention.
• Power management. Smart power IC drives BCD scaling.
  • N90 BCD+ since 2021.
  • N55 BCD+ SoI (pathfinding for automotive)
  • N50BCD+ with RRAM
• CMOS image sensor. Resolution 50MP and beyond. Pixel size 11-1.4um. ISP (processor) N40 going to 12FFC.
• Roadmap for CIS has two logic layers with advanced packaging. Allows for AI in the logic.

...."


What does this mean.....?

 

[Labsy]

 

[FrederikSchack]

I'm so bloody slow sometimes, only now it hit me that neuromorphic computing in a way solves the problem af scaling down the process node, by enabling more performance in different ways:
1) Now that we have such a low power consumption and dissipated heat, it must be possible to run them more agressively.
2) As I see it, neuromorphic computing lends itself perfectly to expanding the amount of silicon used, like connecting multiple chiplets to support greater models and/or running multiple models utilizing each other. They can even be stacked not to take up any significant space.
3) It's a young technology that is already beating the old technology and has a long runway of innovation ahead of it, like the jump from Akida 1 to 2. I bet that there's a vast space of possibilities to be explored, like hardware support for n-dimensional models.

While nVidia hit the brickwall and others are struggeling with Moores law, we just got started and are seemingly already way ahead of their Jetson.

So, now I think neuromorphic computing is going to be indespensable for future performance gains and it might branch out like the three suggestions above and combinations/more branches may appear.

 

[buena suerte :-)]

I'll see if I can get any feed back from my sister as she lives in Cambridge!
Happy Friday Chippers

 

[mrgds]

View attachment 36316

If thats a "GIF " for the Unquoted Securities Announcement today, .................. very clever!

 

[Xhosa12345]

If thats a "GIF " for the Unquoted Securities Announcement today, .................. very clever!

Yup sure was lol

 

[davidfitz]

Sorry if this has already been mentioned but I rarely visit the forum these days.

2023 Shortlist - Global Business Tech Awards

Congratulations to our 2023 Global Business Tech Awards finalists. Finalist Assets

globalbusinesstechawards.com

MANUFACTURING TECH DISRUPTOR OF THE YEAR​

BrainChip Ltd. – Neuromorphic convolutional chip saves up to 97% of power while offering high performance Artificial Intelligence processing.

Phaos Technology

Porotech

SMARTECH Manewfacturing™ Technologies

MARCH 24​

Shortlist Announced

MAY 11​

Winners Announced

The Awards are judged by leading tech industry experts from around the world and judging is ethical, transparent and fair will all entrants receiving feedback on their submissions.

We didn't win in any of the 3 categories we were nominated for but I am happy to see the share price is holding up well despite any positive ASX announcements of late.

 

[MDhere]

Megachips fiscal result comes out today

 

[Damo4]

View attachment 36316

 

[Stable Genius]

RT is a fan of Hugging Face and Edge impulse is excited and commented on this this as well. Looks like it’s another software technology advancement to help with the implementation of our transformers to be released soon:

Exciting news! We can't wait to try it out to help speed up dataset creation for edge AI — many uses for augmentation, auto-labelling, and data synthesis! In fact, check out how we've used OpenAI to… | Edge Impulse (a Qualcomm company)

Exciting news! We can't wait to try it out to help speed up dataset creation for edge AI — many uses for augmentation, auto-labelling, and data synthesis! In fact, check out how we've used OpenAI to generate datasets at Edge Impulse: https://lnkd.in/eUcReCRB

www.linkedin.com

We just released Transformers' boldest feature: Transformers Agents.

This removes the barrier of entry to machine learning: it offers more than 100,000 HF models to be controlled with natural language.

This is a truly fully multimodal agent: text, images, video, audio, docs, and way more to come.

Read the documentation here: https://lnkd.in/eAmxV_zi

What is it?

Create an agent using LLMs (OpenAssistant, StarCoder, OpenAI ...) and start talking to transformers and diffusers through curated tools.

How does it work?

It's straightforward prompt-building:

• Tell the agent what it aims to do
• Give it tools
• Show examples
• Give it a task

The agent uses chain-of-thought reasoning to identify its task and outputs Python code using the tools. It comes with a myriad of built-in tools such as document, text, image QA, Speech-to-text and text-to-speech, text classification, summarization, translation, image edition tools, text-to-video...

But it is EXTENSIBLE by design.

Tools are elementary: a name, a description, a function. Designing a tool and pushing it to the Hub can be done in a few lines of code.

The toolkit of the agent serves as a base: extend it with your tools, or with other community-contributed tools:

LinkedIn

This link will take you to a page that’s not on LinkedIn

lnkd.in

Please play with it, add your tools, and let's create super-powerful agents together.

Here's a notebook to get started: https://lnkd.in/eYsh9eqG

 

[Diogenese]

I'm so bloody slow sometimes, only now it hit me that neuromorphic computing in a way solves the problem af scaling down the process node, by enabling more performance in different ways:
1) Now that we have such a low power consumption and dissipated heat, it must be possible to run them more agressively.
2) As I see it, neuromorphic computing lends itself perfectly to expanding the amount of silicon used, like connecting multiple chiplets to support greater models and/or running multiple models utilizing each other. They can even be stacked not to take up any significant space.
3) It's a young technology that is already beating the old technology and has a long runway of innovation ahead of it, like the jump from Akida 1 to 2. I bet that there's a vast space of possibilities to be explored, like hardware support for n-dimensional models.

While nVidia hit the brickwall and others are struggeling with Moores law, we just got started and are seemingly already way ahead of their Jetson.

So, now I think neuromorphic computing is going to be indespensable for future performance gains and it might branch out like the three suggestions above and combinations/more branches may appear.

Hi Frederick,

1. Akida is asynchronous, responding to input events, so I don't think it can be overclocked. However, producing it as, say, 7nm will increase speed and reduce power.

2. Because Akida does not need cooling because it is so low power, it would be an ideal buried layer in a stacked chip arrangement. One potential application would be the Sony/Prophesee image sensor which uses Sony's 3D technology. I imagine the Prophesee DVS is also low power as it only fires on events. Also some IR night-vision systems need to be cooled, so Akida would be a good fit as it would reduce cooling requirements. Akida is designed for multiple chip applications.

One of my friends whom I convinced to buy BRN was asking about Ai following the recent ChatGPT publicity, so I sent the following:

You asked whether BrainChip's Akida was involved in AI, and I subsequently realized that ChatGPT has been in the news lately, not always favourably.

It is not related to ChatGPT. It could be used on the input side, but, at this stage, not on the output side generating the responses. That is all done with software using a web browser and language interpreting software.

Akida is a spiking neural network which imitates the brain's neural/synaptic processes. It does not use the same digital mathematical methods as conventional computers which consume enormous amounts of processing power in performing object recognition. Think of objects in a field of view as having a line around the edge, ie, where the pixel illumination changes. Akida compares the edge line with stored edge lines in a library of edge lines, whereas an object recognition program in a conventional computer does a pixel-by-pixel comparison of full frame images. Akida also does it in silicon, not in software. It is loaded with its compact object library data and does a hardware comparison, not a software comparison.

Akida is used to classify input signals, voice, video, etc. It does not generate replies.

It can be used for autonomous driving sensors such as LiDaR, radar, video, and event-cameras (shutterless cameras which detect only changes in the pixel illumination) aka DVS (Dynamic Vision Systems) to recognize the surroundings. It is also be used in-cabin driver monitoring, speech recognition, and any application which involves sensor signal interpretation such as vibration sensing to anticipate the need for maintenance. I suppose it could detect when a Boeing spits a turbine blade out.

It is used in Mercedes EQXX concept car which achieved 1000 km in the in-vehicle voice recognition system where it performed 5 to 10 times more efficiently that the other systems they had tested in a system where every Watt counts.

It can be used for autonomous drone navigation and image detection.

NASA are trialing it for navigation, Mars rovers, as well as wireless communication via a company called Intellisense.

Similarly, ISL (Information Systems Limited) is using it in USAF trials.

Akida is listed by ARM as being compatible with all their processors, as well as being compatible with ARM's upstart challenger SiFive which use RISC-V architecture compared with ARM's RISC-IV.

It is also available to Intel Foundry Services (IFS) customers.

It is also part of a few US university computer courses including Carnegie Mellon.

BrainChip's main business is licensing the design of Akida, a similar business model to ARM. This does limit the customer base to those who can afford the licence fee and the cost of incorporating the design in their product and manufacturing the chips. It also introduces a large time lag for designing, making and testing the chips.

A company called Renesas has licenced the Akida IC design and will be bringing out microprocessors capable of handling 30 frames per second (fps) later this year. Akida is capable of much faster fps (> 1000) but Renesas only licenced two Akida nodes out of a possible 80.

Similarly, MegaChips is also in the process of producing chips containing the Akida design.

A second generation of Akida will also be available later this year. This will have the capability to determine the speed and direction of objects in hardware rather than relying on software to process the images identified by Akida. Software is much slower and more power hungry.

 

[Dougie54]

Well every cloud ...

Unemployment in Russia is down 30%, overcrowding in jails has been reduced ...

Unemployment in Russia and overcrowding in jails being reduced would surely be a direct result of Putin sending every able body to the Ukraine to fight.!!!

 

[Damo4]

Hi Frederick,

1. Akida is asynchronous, responding to input events, so I don't think it can be overclocked. However, producing it is, say, 7nm will increase speed and reduce power.

2. Because Akida does not need cooling because it is so low power, it would be an ideal buried layer in a stacked chip arrangement. One potential application would be the Sony/Prophesee image sensor which uses Sony's 3D technology. I imagine the Prophesee DVS is also low power as it only fires on events. Also some IR night-vision systems need to be cooled, so Akida would be a good fit as it would reduce cooling requirements. Akida is designed for multiple chip applications.

One of my friends whom I convinced to buy BRN was asking about Ai following the recent ChatGPT publicity, so I sent the following:

You asked whether BrainChip's Akida was involved in AI, and I subsequently realized that ChatGPT has been in the news lately, not always favourably.

It is not related to ChatGPT. It could be used on the input side, but, at this stage, not on the output side generating the responses. That is all done with software using a web browser and language interpreting software.

Akida is a spiking neural network which imitates the brain's neural/synaptic processes. It does not use the same digital mathematical methods as conventional computers which consume enormous amounts of processing power in performing object recognition. Think of objects in a field of view as having a line around the edge, ie, where the pixel illumination changes. Akida compares the edge line with stored edge lines in a library of edge lines, whereas an object recognition program in a conventional computer does a pixel-by-pixel comparison of full frame images. Akida also does it in silicon, not in software. It is loaded with its compact object library data and does a hardware comparison, not a software comparison.

Akida is used to classify input signals, voice, video, etc. It does not generate replies.

It can be used for autonomous driving sensors such as LiDaR, radar, video, and event-cameras (shutterless cameras which detect only changes in the pixel illumination) aka DVS (Dynamic Vision Systems) to recognize the surroundings. It is also be used in-cabin driver monitoring, speech recognition, and any application which involves sensor signal interpretation such as vibration sensing to anticipate the need for maintenance. I suppose it could detect when a Boeing spits a turbine blade out.

It is used in Mercedes EQXX concept car which achieved 1000 km in the in-vehicle voice recognition system where it performed 5 to 10 times more efficiently that the other systems they had tested in a system where every Watt counts.

It can be used for autonomous drone navigation and image detection.

NASA are trialing it for navigation, Mars rovers, as well as wireless communication via a company called Intellisense.

Similarly, ISL (Information Systems Limited) is using it in USAF trials.

Akida is listed by ARM as being compatible with all their processors, as well as being compatible with ARM's upstart challenger SiFive which use RISC-V architecture compared with ARM's RISC-IV.

It is also available to Intel Foundry Services (IFS) customers.

It is also part of a few US university computer courses including Carnegie Mellon.

BrainChip's main business is licensing the design of Akida, a similar business model to ARM. This does limit the customer base to those who can afford the licence fee and the cost of incorporating the design in their product and manufacturing the chips. It also introduces a large time lag for designing, making and testing the chips.

A company called Renesas has licenced the Akida IC design and will be bringing out microprocessors capable of handling 30 frames per second (fps) later this year. Akida is capable of much faster fps (> 1000) but Renesas only licenced two Akida nodes out of a possible 80.

Similarly, MegaChips is also in the process of producing chips containing the Akida design.

A second generation of Akida will also be available later this year. This will have the capability to determine the speed and direction of objects in the hardware rather than relying on software to process the images identified by Akida. Software is much slower and more power hungry.

Great post Dio!

It's also important to zoom out and think about who wants Akida and why.
They aren't chasing extra flops (equivalent) per watt, they are trying to reduce watts per operation.
The super computers and Nordic self-cooling warehouses can handle the big stuff, but only low power, edge NM technology will solve the use-cases you mention above.

Great post again, it's easy to lose track of all the connections and reasons why Akida excels.

 

[Stable Genius]

Blackberry has been hinted as being an EAP before:

Maybe they’ll be the first one to adopt the Prophesee/Akida combo?

 

[Draed]

Bottomed out at 40c with low buyong volumes... I think the shorters will try to close soon or are starting too... I'm hoping they start feeling some panic.. maybe something coming towards the agm?

 

[SERA2g]

Thank you for the lesson in comprehension. I'll interpret what you said how I spelled it out last time. You implied that our involvement with Valeo and their $1 billion sale would be a company maker for us. It will be a very nice little earner but I disagree that it will be a company maker. Let's move on.

I think you may need to take some of your own advice regarding what I wrote in relation to my holding and "sale" not purchase of shares at $2.13.

Sorry, missed this response.

My bad on not reading your post correctly. Well played on the $2.13 sale!

For the rest of it, let's agree to disagree

 

[MadMayHam | 合氣道]

Blackberry has been hinted as being an EAP before:

View attachment 36323


Maybe they’ll be the first one to adopt the Prophesee/Akida combo?

I thought that the obvious application for a low SWaP neuromorphic system would be to analyse patterns in network packet data so I actually did the introductions between Rob and a senior executive in BB (whom I coldcalled) in 2021 over email, though unfortunately of course I'm not privy to the outcome of those discussions. Obviously I'm also hoping that something comes to fruition.

Edit: Rob liking a Blackberry executive's post in light of the above...

 

[Diogenese]

I thought that the obvious application for a low SWaP neuromorphic system would be to analyse patterns in network packet data so I actually did the introductions between Rob and a senior executive in BB in 2021 over email though unfortunately of course I'm not privy to the outcome of those discussions. Obviously I'm also hoping that something comes to fruition.

Well done!

That is several levels above and beyond dot joining.

Let's hope something comes of it.

 

[DollazAndSense]

Hi Frederick,

1. Akida is asynchronous, responding to input events, so I don't think it can be overclocked. However, producing it as, say, 7nm will increase speed and reduce power.

2. Because Akida does not need cooling because it is so low power, it would be an ideal buried layer in a stacked chip arrangement. One potential application would be the Sony/Prophesee image sensor which uses Sony's 3D technology. I imagine the Prophesee DVS is also low power as it only fires on events. Also some IR night-vision systems need to be cooled, so Akida would be a good fit as it would reduce cooling requirements. Akida is designed for multiple chip applications.

One of my friends whom I convinced to buy BRN was asking about Ai following the recent ChatGPT publicity, so I sent the following:

You asked whether BrainChip's Akida was involved in AI, and I subsequently realized that ChatGPT has been in the news lately, not always favourably.

It is not related to ChatGPT. It could be used on the input side, but, at this stage, not on the output side generating the responses. That is all done with software using a web browser and language interpreting software.

Akida is a spiking neural network which imitates the brain's neural/synaptic processes. It does not use the same digital mathematical methods as conventional computers which consume enormous amounts of processing power in performing object recognition. Think of objects in a field of view as having a line around the edge, ie, where the pixel illumination changes. Akida compares the edge line with stored edge lines in a library of edge lines, whereas an object recognition program in a conventional computer does a pixel-by-pixel comparison of full frame images. Akida also does it in silicon, not in software. It is loaded with its compact object library data and does a hardware comparison, not a software comparison.

Akida is used to classify input signals, voice, video, etc. It does not generate replies.

It can be used for autonomous driving sensors such as LiDaR, radar, video, and event-cameras (shutterless cameras which detect only changes in the pixel illumination) aka DVS (Dynamic Vision Systems) to recognize the surroundings. It is also be used in-cabin driver monitoring, speech recognition, and any application which involves sensor signal interpretation such as vibration sensing to anticipate the need for maintenance. I suppose it could detect when a Boeing spits a turbine blade out.

It is used in Mercedes EQXX concept car which achieved 1000 km in the in-vehicle voice recognition system where it performed 5 to 10 times more efficiently that the other systems they had tested in a system where every Watt counts.

It can be used for autonomous drone navigation and image detection.

NASA are trialing it for navigation, Mars rovers, as well as wireless communication via a company called Intellisense.

Similarly, ISL (Information Systems Limited) is using it in USAF trials.

Akida is listed by ARM as being compatible with all their processors, as well as being compatible with ARM's upstart challenger SiFive which use RISC-V architecture compared with ARM's RISC-IV.

It is also available to Intel Foundry Services (IFS) customers.

It is also part of a few US university computer courses including Carnegie Mellon.

BrainChip's main business is licensing the design of Akida, a similar business model to ARM. This does limit the customer base to those who can afford the licence fee and the cost of incorporating the design in their product and manufacturing the chips. It also introduces a large time lag for designing, making and testing the chips.

A company called Renesas has licenced the Akida IC design and will be bringing out microprocessors capable of handling 30 frames per second (fps) later this year. Akida is capable of much faster fps (> 1000) but Renesas only licenced two Akida nodes out of a possible 80.

Similarly, MegaChips is also in the process of producing chips containing the Akida design.

A second generation of Akida will also be available later this year. This will have the capability to determine the speed and direction of objects in hardware rather than relying on software to process the images identified by Akida. Software is much slower and more power hungry.

Brainchip should employ you to proof read all future marketing releases - not a single mention of Alida or Akido!

 

[MadMayHam | 合氣道]

Well done!

That is several levels above and beyond dot joining.

Let's hope something comes of it.

View attachment 36316

Hopefully I'll get rewarded in shares for my efforts when it comes to pass.