BRN Discussion Ongoing

[RobjHunt]

How wrong was I when I thought BRN was speculative at 3c?

 

[RobjHunt]

Absolutely correct @Diogenese.

Sean Hehir made crystal clear at least to me as I listened to the presentation that far from research being behind Brainchip the Perth Research Centre, Peter van der Made and his team are being elevated to protected species categorisation and will be driven to keep progressing the science underpinning the AKIDA technology advantage without any interference or distractions.

At least to me Sean was saying we understand that Peter van der Made’s genius has given us this opportunity and we understand two things:

1. He is not done yet.

2. We have an absolute responsibility to do him and his technology justice and we damn well will, so don’t stand in our way or upset Peter van der Made.

My opinion only DYOR
FF

AKIDA BALLISTA

We need to leave Peter be. He is busy!

 

[Bravo]

View attachment 7560

When I first saw this, I thought we had 3 new partners called Channel, Technical and Enablement.

 

[Sirod69]

George BouAntoun on LinkedIn: #brainchip #neuromorphic #esg #shawandpartners #leadingedge

Thank you to the BrainChip Board and Management team for providing our clients at Shaw and Partners with the opportunity to hear about the commercialisation…

www.linkedin.com

Thank you to the BrainChip Board and Management team for providing our clients at Shaw and Partners with the opportunity to hear about the commercialisation strategy for what is now an IP company leading the way with ultra low power neuromorphic edge chips for wide scale use.
It was an insightful session hearing from:
Antonio J. Viana, Sean Hehir, Peter Van Der Made, Geoff Carrick, Pia Turcinov, Ken Scarince and Tony Dawe.

 

[LM77]

 

[JK200SX]

View attachment 7636

I guess this was the secretive meeting people were talking about?

Looks good to me, and all positive!

 

[Bravo]

View attachment 7636

Shaw and Partners are one of the countries largest Stockbroking firms. I think they have close to 30,000 clients that they recommend stocks to.

 

[Fact Finder]

Those who are watching Tech Girls recording two things to note:

1. Peter van der Made’s comment that they are working on making a further 50 patents ready for filing.

2. Sean Hehir’s wording around the near term release of what was formally called AKD2000 with LSTM as IP for commercial licence.

The significance of point 1. is obvious and requires no comment.

Point 2. however does.

Every longer term holder from say early 2020 knows that the former CEO Mr. Dinardo, Peter van der Made and Anil Mankar stated and I paraphrase that they needed AKD1000 in silicon to get EAP’s and Proof of Concept customers over the commercial line because the AKIDA technology was/is so unbelievable that like doubting Thomas from the Bible these customers needed to thrust their fingers through the wounds on the living Christ’s hands so to speak. In other words they wanted absolute proof before committing. I recall that the former CEO Mr. Dinardo saying “if you are spending 30,40, 50 million” to implement a technology you want to be confident it will work.

Up until Christmas 2021 Brainchip was still going to produce AKD2000 with LSTM in silicon this year and now they are not they are just going to release the IP.

This tells me that the Mercedes Benz public validation of the technology has heralded a change in the general market acceptance of the AKIDA technology.

By the way a little bird I met today told me that after SiFive and Brainchip announced their partnership ARM appeared at the BRN front door asking to be let in straight away.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Bravo]

I Feel like a 3 legged chair

You don’t look like one!

 

[cosors]

Hmmmm...I know Intel are still working on Loihi 2, but this has got me thinking whether this is an admission that they need other AI solutions. Hello AKIDA!





View attachment 7332

AI Is Garnering a Bigger Role in Intel’s Future

Intel is moving the focus from its server CPUs to a growing roster of adjacent chips that are driving a fundamental shift in computing to AI in which

www.enterpriseai.news

I have seen that you are dealing with Intel's Loihi 2. The University of Graz has tested it or is still testing. I don't know if you already know it but I want to close the tab so here is the article:

Tue 24 May 2022
"Their research paper, "A Long Short-Term Memory for AI Applications in Spike-based Neuromorphic Hardware," published in Nature Machine Intelligence, claims that the Intel chips are up to 16 times more energy efficient in deep learning tasks than performing the same task on non-neuromorphic hardware. The hardware tested consisted of 32 Loihi chips."

 

[JK200SX]

One of the countries largest Stockbroking firms.

Preliminary discussions for NASDAQ listing?

 

[Diogenese]

Those who are watching Tech Girls recording two things to note:

1. Peter van der Made’s comment that they are working on making a further 50 patents ready for filing.

2. Sean Hehir’s wording around the near term release of what was formally called AKD2000 with LSTM as IP for commercial licence.

The significance of point 1. is obvious and requires no comment.

Point 2. however does.

Every longer term holder from say early 2020 knows that the former CEO Mr. Dinardo, Peter van der Made and Anil Mankar stated and I paraphrase that they needed AKD1000 in silicon to get EAP’s and Proof of Concept customers over the commercial line because the AKIDA technology was/is so unbelievable that like doubting Thomas from the Bible these customers needed to thrust their fingers through the wounds on the living Christ’s hands so to speak. In other words they wanted absolute proof before committing. I recall that the former CEO Mr. Dinardo saying “if you are spending 30,40, 50 million” to implement a technology you want to be confident it will work.

Up until Christmas 2021 Brainchip was still going to produce AKD2000 with LSTM in silicon this year and now they are not they are just going to release the IP.

This tells me that the Mercedes Benz public validation of the technology has heralded a change in the general market acceptance of the AKIDA technology.

By the way a little bird I met today told me that after SiFive and Brainchip announced their partnership ARM appeared at the BRN front door asking to be let in straight away.

My opinion only DYOR
FF

AKIDA BALLISTA

Yep - ARM has RISC-IV and SiFive has RISC-V. There was no way ARM could have not joined the Akida bandwagon ... and when the word gets out about the loaves and fis...? sorry the 250 fps to 1000 fps, they'll be glad they did.

 

[Fox151]

Hi all
I won’t grace the nonsense with a rebuff but I will support those who have reported a positive response to the AGM.

It was as reported an entirely positive investment affirming experience.

I did not have an opportunity to speak to Sean Hehir or any one else at the meeting about the company as I was keeping socially distant and masked as I have a responsibility to minimise the risk of transferring Covid to my wife.

I did shake hands with Ken Scarince before leaving and thanked him for the work he has done with LDA Capital.

I then attended with some others a demonstration by Nviso of its in cabin system that we have all seen the video of and it is more impressive in real life. This was entirely independent of Brainchip and it was organised by Jesse Chapman.

In due course thanks to the work of Jesse and Acca we here may be remembered when the Nviso IPO is launched later this year, early next year.

At the time of the video we saw AKIDA was processing 250 frames per second with the Nvidia CPU which without AKIDA was accordingly to Tim Llewellyn only processing up to 10 frames per second.

Since they made the video two weeks ago Anil Mankar has suggested some tweaks to AKIDA and this has taken the 250 frames per second to 1,000 frames per second which was the speed I/we witnessed.

It was clear from the answers Tim Llewellyn gave that without AKIDA what he was presenting would not be possible. He was very derogatory of the technology currently in phones and it was clear the only way phones would be able to run Nviso Apps properly was with AKIDA.

As for the share price movement this is clearly manipulation and on this point there has been significant increase in institutional interest.

From another shareholder I have been made aware that two meetings took place with large institutions after the AGM that were requested late by these institutions through Shaw Partners and plans had to be changed suddenly so they could be accommodated by Sean Hehir & Co.

I have no reason to believe that the share price manipulation today and these meetings were/are connected.

It is not unreasonable though to say if ARM and Nviso have been the catalyst for friendly conservative institutions to take a look at Brainchip then bad actors would also have been put on notice.

I have made the point before that as a retail investor my approach is to ignore the manipulation and trust in my research.

I see no reason to change my approach and having seen the AKIDA Nviso demonstration I am completely convinced this is the correct decision.

My opinion only DYOR
FF

AKIDA BALLISTA

So, FF, what you are saying is that, by upgrading Akida from 250 fps to 1,000 fps capable, the Akida 1000 in my mini PCIe card that is currently sitting with the framer is already out of date? Does that make it more or less valuable?

 

[Diogenese]

So, FF, what you are saying is that, by upgrading Akida from 250 fps to 1,000 fps capable, the Akida 1000 in my mini PCIe card that is currently sitting with the framer is already out of date? Does that make it more or less valuable?

Hi Fox51,

No, there was no change to the silicon - the same Akida 1000 chip was used as the one you have at the picture hangers.

Akida has 20 nodes each consisting of 4 NPUs (network processing Units) all connected by a switchable connexion matrix which can be configured by the ARM Cortex MPU in Akida.

This allows the arrangement of the nodes to be changed to suit the particular application. For instance, there may be:
5 layers of 4 nodes,
4 layers of 4 nodes (I don't know if you can have an odd number of nodes in a layer so I'm leaving 4 nodes idle),
3 layers of 6 nodes (2 idle nodes), or
2 layers of up to10 nodes.

Layers constitute a processing delay as the signals must pass through each layer in series from the sensor to the output, so the fewer layers the faster the processing time. Also the more nodes in a layer, the more NPUs are in parallel, which speeds up the processing (or increases accuracy?).

In addition, the clock speed can be increased to increase the processing speed.

But none of this requires any changes to the silicon architecture. the flexible/configurable connexion network enables the reconfiguration of the nodes.

It would have been impossible to put a new Akida silicon SoC through the design and manufacturing processes in 2 weeks.


PS: The term layers is used in the electrical sense, not in the physical sense.

 

[Mugen74]

Hi Fox51,

No, there was no change to the silicon - the same Akida 1000 chip was used as the one you have at the picture hangers.

Akida has 20 nodes each consisting of 4 NPUs (network processing Units) all connected by a switchable connexion matrix which can be configured by the ARM Cortex MPU in Akida.

This allows the arrangement of the nodes to be changed to suit the particular application. For instance, there may be:
5 layers of 4 nodes,
4 layers of 4 nodes (I don't know if you can have an odd number of nodes in a layer so I'm leaving 4 nodes idle),
3 layers of 6 nodes (2 idle nodes), or
2 layers of up to10 nodes.

Layers constitute a processing delay as the signals must pass through each layer in series from the sensor to the output, so the fewer layers the faster the processing time. Also the more nodes in a layer, the more NPUs are in parallel, which speeds up the processing (or increases accuracy?).

In addition, the clock speed can be increased to increase the processing speed.

But none of this requires any changes to the silicon architecture. the flexible/configurable connexion network enables the reconfiguration of the nodes.

It would have been impossible to put a new Akida silicon SoC through the design and manufacturing processes in 2 weeks.


PS: The term layers is used in the electrical sense, not in the physical sense.


View attachment 7637

Reading your posts sometimes(who am I kidding mostly)makes me feel stupid.

 

[Diogenese]

I think it's time:

 

[Mugen74]

I think it's time:

Great song,great band. I Prefer

 

[Beebo]

Did any BRCHF holders receive voting proxy from their respective brokerages?
Just curious…
I received nothing from SCHWAB, nothing from TDameritrade.

 

[dippY22]

I think you have it! at least it seems logical because the formulation or terminology is not unique....
I was first at the development platform Bonseyes who talk about deep edge where Nviso, ARM, ZF and other are founding members. Nviso also describes devices beyond the edge or in the deep edge for example in the medical sector, sensors or handhelds. I wanted to search further with them.

I came across the link to their app for phones. If you press on it, this appears (https://www.bonseyes.com/):
View attachment 7625
and in the app store is not yet to be found. Maybe it is still in development. Pity, would have liked to watch it to see what it's all about:
View attachment 7631 I don't know if this is the same or a tool of Bonsapp.

Can of course have nothing to do with Akida and beyond the edge. But maybe it goes in the direction of smart devices like watches or bracelets that can be controlled via an app from the marketplace app by Bonseyes. The definition seems to be important not only for Nviso, beyond the edge, deep edge or edge of the edge.

View attachment 7630
https://www.bonseyes.com/

Here also Renesas or Nvidia... Maybe this is all connected via a large network, like a mycelium. The chairman mentioned what is important to him
"I, personally, will always challenge the company to strike the right balance between technology, timing and the corresponding deployment strategy. Let me stress - these are not simply words... this is real...
...
they forgot about ecosystem development and integration concerns... "

Maybe we are already looking at it. Or maybe it's all nonsense on my part and it's about edge data centers.

Bravo and Cosars,......BINGO. I also think this is a huge market potential for Brainchip IP. None other than Tim Cook believes that direction is the right way for his company to go too in the future with wonderful potential benefits to society (i.e. health). Apple cars??? Meh.

But more Apple wearables in the future !!!

And the Apple zealots will consume whatever the Cook serves up. So stir in the Akida secret sauce and serve it up to the masses Apple. Let's go.

 

[zeeb0t]

I’ll echo other peoples experience with my own… it was fantastic meeting a number of folks from here, some of which also shouted me a drink - that was very nice of you, thank you.

Meeting the BRN team face to face was also nice, I particularly enjoyed meeting Ken because we’ve had a dialogue open for years now.

As the admin of this forum the ASX doesn’t really allow me specifically to say too much but what I will say is to echo Fact Finders own words, “an investment reaffirming experience” sounds like a perfect one line summary of the day / evening.