BRN Discussion Ongoing

[Sirod69]

Magnus Ostberg-Mercedes wrote:
"
The best ideas are developed by a team, not a single mind . That's why, for us, #LeadInCarSoftware also means working together with talented and innovative partners - like Unity. We work with Unity on our operating system MB.OS, where their solutions are integrated into #UserInterface, making real-time visualisation possible.

Cooperation and partnership are a crucial way for us to offer our customers the best possible experience. In fact, our MB.OS software is being developed in part with several excellent partners. We Mercedes-Benz AG will be responsible for complete vehicle integration. This is just one way we make excitement come to life: with new functions, services, and 3rd party apps!

I think we are a excellent partner!!!

 

[Deleted member 118]

Another market and another 50 billion dollar market. We need the Scrooge MCDuck diving into his gold coin vault gift. This is becoming a complete impossibility to arrive at any sort of reasonable valuation.

It is not fanciful to suggest a potential 500 billion US dollar addressable market.

The entire automotive industry in 2021 was less than 450 billion US dollars and that includes accessories like automotive air fresheners.

My opinion only DYOR
FF

AKIDA BALLISTA

That’s just graphic cards on top of that you’ll have VR market and by 2024 could hit 297 billion a year

VR/AR market size 2024 | Statista

Global spending on AR and VR has been growing rapidly. The market is expected to reach close to 297 billion USD by 2024.

www.statista.com

Drone market will be worth as much as 60 billion a year by 2026

UAV (Drone) Market Size, Share, Industry Report, Revenue Trends and Growth Drivers

The global UAV (Drone) Market in terms of revenue was estimated to be worth $30.2 billion in 2024 and is poised to reach $48.5 billion by 2029, growing at a CAGR of 9.9% during the forecast period.

www.marketsandmarkets.com

Computer market getting close to 1/2 trillion pounds by 2026

Please verify you are human.

www.einnews.com

And this is small fry compare the the sensor market

Global Acoustic Wave Sensor Market to 2026 - Featuring Pro-micron, Siemens and Transense Among Others - ResearchAndMarkets.com

DUBLIN, April 20, 2022--The "Acoustic Wave Sensor Market - Forecasts from 2021 to 2026" report has been added to ResearchAndMarkets.com's offering.

au.finance.yahoo.com

projected to grow at a CAGR of 16.26% to reach $2,400.540 million by 2026, from $836.174 million in 2019.


Only a small $4,000,000,000,000 a year market come 2026 for devices that could easily implement akida

 

[Fact Finder]

That’s just graphic cards on top of that you’ll have VR market and by 2024 could hit 297 billion a year

VR/AR market size 2024 | Statista

Global spending on AR and VR has been growing rapidly. The market is expected to reach close to 297 billion USD by 2024.

www.statista.com

Stop it please.

If I add the 200 billion for automotive semiconductors by 2040 plus the 153 billion for robotics plus your 53 billion for graphic cards, plus your 297 billion for VR, plus 70 billion for Edge, plus military, space and aeronautics, plus drones, plus completely changing the face of medical practice by 2030 as stated by Professor Barry Marshall I am getting to that ridiculous 1 trillion US dollar market that was published recently and that just sounds absurd and it would make my claims of investing in Brainchip to create generational wealth seem FEASIBLE.

@Rocket577 can you just go back to posting gifs please.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[miaeffect]

Yes, extremely difficult to put a valuation on this wonderfull business.
The AUD is currently tanking and I've supplied the current exchange rate with a AUD price tag of $10 per share(US $6.87) in the event of a takeover
I don't want this to happen but the little aussie bleeder may come under more pressure as the Feds next move may be up another 50 points??
Just thinking aloud.

View attachment 6393

Minimum $50 AUD per share

 

[Diogenese]

Another patent, that has been pending, seems to be in force (May, 11th)
https://register.dpma.de/DPMAregister/pat/register?AKZ=E208430314&CURSOR=1

Good evening to AUS and regards

Hi cassip,

Thanks for digging this patent application out.

My complaint about PvdM's patents is that they are so densely packed with landmark inventions that it is difficult to get a complete understanding of them on one or two readings, so I usually just target a particular aspect to avoid neural overload.

We have seen this one before and, at the time, I was interested in the rank coding v rate coding feature described with reference to Fig 3:

I'm still trying to reconcile this with Simon Thorpe's explanation of JAST/STDP/winner-take-all, but that's a problem for another day.

A few weeks ago there was a discussion about the changes to the Akida architecture in the final revision incorporating CNN2SNN and 4-bit weights and actuation. I speculated that the use of 4-bit weights/actuations would necessitate a change from 1*1 multipliers (AND gates) to 4*4 multipliers, and now on a less cursory review of this specification, here it is:

[0059] The choice of the number of bits is one important consideration in the design of the layers in neural network architecture. The default number of bits commonly used in neural networks is 32 bits, with multipliers being 32 bit by 32 bit. Lesser bits may be used, with a resulting smaller power consumption and chip area required, but with a loss of accuracy. Thus, a 1 bit x 1 bit multiplier would represent the lowest limit on bit choice. Initial experiments have shown that a 92% accuracy is achievable using ternary weights. Experiments by the inventors have also shown that a 4x4 bit multiplier provides a sweet spot in terms of relatively low reduction in accuracy (compared with the traditional 32 bit x 32 bit approach), while not significantly increasing the power consumption and chip area requirement over that required for a 1 bit x 1 bit solution. The sweet spot choice results in a reduction by a factor of 3 to 4 in the number of overall parameters used. In various embodiments, the number of bits used may be different on different layers, i.e., in these embodiments, a layer by layer choice of the number of bits to use (1x1, 2x2, 4x4 etc.) may be made.

Moving to 4-bits greatly increases accuracy at the expense of some speed and power, but is still much more speed and power efficient than the conventional CNN 32-bit multiplication.

... and that's still without going into the CNN2SNN aspects of the patent, and what other hidden gems there may be within

 

[Esq.111]

I was just thinking about the statement made by the former CEO Mr. Dinardo in mid 2020 "WE HAVE WELL NORTH OF 100 NDA'S" and then went on to say that they were culling them down to the best commercial opportunities and were not interested in research projects because they did not have the resources to support all these NDA's and that they would probably settle on about two dozen.

In the end this is what they did and though the exact number has remained rubbery over 20 and less than 30 seemed to be what we can extrapolate from all the different statements by key management over the last two years.

We also know that in 2015 in an AGM presentation Peter van der Made as the CEO stated that they were in discussions even back then with Fortune 500 companies. We know also that before AKD1000 became the sole focus of Brainchip that companies like Safran and Rockwell Collins and Cisco were all working with Brainchip. We then know that the former CEO Mr. Dinardo further on in 2020 qualified the statement 'best commercial opportunities' describing these companies as 'household named and Fortune 500 companies' then it becomes as a matter of mathematical probabilities much easier to identify from the above listed companies who are most likely to be well advanced because they joined the EAP program in 2020 as they need to also match being household named and/or Fortune 500.

Also we can reduce the numbers by taking out those that have actually been publicly discussed by Brainchip:

1. Mercedes Benz
2. Valeo
3. Ford
4. Renesas
5. MegaChips
6. ARM
7. NASA
8. SiFive
9. Socionext
10. TATA


So we have 10 to 20 'household named and/or Fortune 500 companies' as at 2020 still to be revealed from the original cohort and who according to both Peter van der Made the CTO and Sean Hehir the CEO of Brainchip all bar one are believed to be going to convert.

So as you look through my list above and your list which of these companies meet the definition of the former CEO Mr. Dinardo well my suggestion is:

1. Sanyo
2. Nintendo
3. Panasonic
4. Nvidia
5. DELL Technologies
6. Samsung
7. Toyota
8. BMW
9. APPLE
10. Intel
11. AWS
12. Cisco
13. Safran

So I ask this final question if of the 14 names above 10 only are the yet to be revealed EAP's that the former CEO Mr. Dinardo signed up and of that 10. only 9. adopt and use AKIDA products how huge will the market be that AKIDA technology captures when you add these 9 out of these 13 named companies to the 10. already in the public sphere.

Can I say this I think you can reduce the odds further by removing Nintendo because EAP or not it has to be part of the use cases targeted by MegaChips in any event so you probably are selecting 9 out of the 12 names in this list.

These are exciting times if I can steal the quote of Rob Telson.

Please note it is clear there are many more companies and industries engaged now than in 2020 but I am trying to narrow the possible suspects as to that original group.

My opinion only DYOR
FF

AKIDA BALLISTA

Evening Chippers,

This may blow some minds, it has mine, and possibly make Fact Finder's list of quite possible engagements.

Please bear with me...

Regarding the Deloitte valuation of BRN back in 2019, our share price was around $ 0.05 and Deloitte put a potential 16,000% possible price rise on BRN which would put us at $8.00 per share, give or take.

Now Deloitte are not stupid.
150 year old company.
286,000 odd global employees / partners in the firm.

Deloitte subsidiaries in the US serve more than 85% Of US Fortune 500 Companys & member firms of Deloitte Touche Tohmatsu Limited collectively serve more than 70% Of the Fortune Global ( FG )500 Companys.

This being the case thay would have a very good idea of what these company's are up to.

Deloitte is a private company ( Not listed on any exchange ).

This is but some of their clients ( this entire article I have written is from information sourced from Googleing Deloitte and as such I personally cannot vouch for the veracity of the info heirin).

Client List

Procter & Gamble
GlaxoSmithKline
Microsoft
GM
Boeing
UPS
Apple
IBM
HP
State Grid Corperation of China ( largest utility company in the world )
BAE Systems
NAVEY
NASA
Government of Australia
Monsanto
Berkshire Hathaway
The Blackstone Group
Morgan Stanly
&
Chipotle ( Mexican Grill)
African Cycling Team for the Tour De France.

To name but a few , past and pressent.

So through this truely massive collection of data , collated, Deloitte would have a front seat when giving share price predictions based on known facts only to them.

Make of it what you will , but I think Deloitte alone is a possible contender.

Regards,
Esq.

 

[Deleted member 118]

Stop it please.

If I add the 200 billion for automotive semiconductors by 2040 plus the 153 billion for robotics plus your 53 billion for graphic cards, plus your 297 billion for VR, plus 70 billion for Edge, plus military, space and aeronautics, plus drones, plus completely changing the face of medical practice by 2030 as stated by Professor Barry Marshall I am getting to that ridiculous 1 trillion US dollar market that was published recently and that just sounds absurd and it would make my claims of investing in Brainchip to create generational wealth seem FEASIBLE.

@Rocket577 can you just go back to posting gifs please.

My opinion only DYOR
FF

AKIDA BALLISTA

I made it $4,000,000,000,000

 

[Baisyet]

Here we go sorry if it has been posted before

Day 2

 

[Learning]

Back in the UK I use to be big into online gaming playing FPS games likes of unreal tournament, far cry, halo, quake, doom, call of duty, counter strike and red faction. This 250fps just makes me want to build a new computer when I can get hold of a graphic card with Akida running it.

Current top graphic card doesn’t even have 1/2 the fps

The Best Graphics Cards Shortlist

GPU	Performance Rank	DXR Rank	Value Rank – online (MSRP)
Nvidia GeForce RTX 3090 Ti	1 – 132.4 fps	1 – 84.4 fps	13 – $2,000 ($1,999)
Nvidia GeForce RTX 3080	4 – 116.3 fps	2 – 66.3 fps	12 – $949 ($699)
AMD Radeon RX 6900 XT	2 – 130.6 fps	3 – 49.8 fps	11 – $1,020 ($999)
AMD Radeon RX 6800 XT	3 – 124.5 fps	4 – 46.1 fps	10 – $920 ($649)
AMD Radeon RX 6800	5 – 111.7 fps	6 – 39.3 fps	9 – $800 ($579)
Nvidia GeForce RTX 3060 Ti	7 – 91.5 fps	5 – 43.3 fps	7 – $580 ($399)
AMD Radeon RX 6700 XT	6 – 96.0 fps	8 – 30.5 fps	6 – $515 ($489)
Nvidia GeForce RTX 3060	9 – 70.2 fps	7 – 32.3 fps	5 – $390 ($329)
AMD Radeon RX 6600 XT	8 – 78.2 fps	9 – 23.6 fps	4 – $410 ($379)
AMD Radeon RX 6600	10 – 66.7 fps	11 – 19.7 fps	3 – $325 ($329)
Nvidia GeForce RTX 3050	11 – 51.4 fps	10 – 22.8 fps	2 – $300 ($249)
AMD Radeon RX 6500 XT	12 – 30.4 fps	12 – 5.6 fps	1 – $210

And with the current graphic card market worth above $50 billion a years. It’s going to be interesting who implements it 1st.

Totally agree Rocket577.

Its exactly what's i was thinking about a month ago.

I believe Akida would fits in the graphic cards market nicely.

Low power and image processing capabilities is better then what's offered in the current market.

Sooner or later Nvidia or AMD will have to embedded Akida to stay ahead. (My opinion)

It's great to be a shareholder

 

[Deleted member 118]

@Diogenese anything with amd regarding patents

AMD Radeon RX 7000 GPUs to Feature Machine Learning Chip for Improved AI Upscaling in Games [Report] | Hardware Times

AMD’s Radeon RX 7000 graphics cards are set to launch in the third (or fourth) quarter of 2022. Leveraging TSMC’s 5nm process, and a chiplet-based design, these GPUs are slated to be the first consumer offerings to combine multiple dies on the same substrate. Not only that, but they’ll also be...

www.hardwaretimes.com

 

[Deleted member 118]

@Facts we forgot the smart phone and tablet market combined @ $800 billion give or take a few billion

The global smartphone market is anticipated to reach $493.13 billion by 2026 | Klever Finance

In 2022, the number of smartphone users in the world today is 6.648 Billion, which…

klever.finance

The Global Rugged Tablet Market is expected to grow by $ 307.39 mn during 2022-2026 progressing at a CAGR of 7.29% during the forecast period

Global Rugged Tablet Market 2022-2026 The analyst has been monitoring the rugged tablet market and it is poised to grow by $ 307. 39 mn during 2022-2026 progressing at a CAGR of 7.New York, April 13, 2022 (GLOBE NEWSWIRE) -- Reportlinker.com announces the release of the report "Global Rugged...

au.sports.yahoo.com

 

[Fact Finder]

Why did Peter van der Made have to mention that AKIDA could be implemented on 5 nm? FF


AKIDA BALLISTA

 

[Deleted member 118]

Why did Peter van der Made have to mention that AKIDA could be implemented on 5 nm? FF


AKIDA BALLISTA

 

[Learning]

Why did Peter van der Made have to mention that AKIDA could be implemented on 5 nm? FF


AKIDA BALLISTA

Its great to be a shareholder.

 

[Diogenese]

@Diogenese anything with amd regarding patents

AMD Radeon RX 7000 GPUs to Feature Machine Learning Chip for Improved AI Upscaling in Games [Report] | Hardware Times

AMD’s Radeon RX 7000 graphics cards are set to launch in the third (or fourth) quarter of 2022. Leveraging TSMC’s 5nm process, and a chiplet-based design, these GPUs are slated to be the first consumer offerings to combine multiple dies on the same substrate. Not only that, but they’ll also be...

www.hardwaretimes.com

It seems that AMD are aware of CNNs.

https://worldwide.espacenet.com/pat...machine learning" AND nftxt = "graphics card"



US2021383527A1 AUTOMATED ARTIFACT DETECTION

[0023] The orchestrator 206 trains the network 200 by providing noise to the generator 202 , which outputs generated images 203 . The orchestrator 206 causes the discriminator 204 to output a classification 205 that indicates whether the generated image 203 includes or does not include a glitch. The orchestrator 206 trains the discriminator 204 to accurately classify the input image set 201 (i.e., as either including a glitch if the input image set 201 includes only glitched images or as not including a glitch if the input image set 201 does not include any glitched images), and to accurately classify the generated images 203 . The orchestrator 206 trains the generator 202 to “fool” the discriminator 204 , by maximizing the error rate of the discriminator 204 . Thus the orchestrator 206 continually improves the ability of the generator 202 to generate “realistic” images, and improves the ability of the discriminator 204 to accurately classify an image as including either a glitch or not including a glitch. In some implementations, the discriminator 204 is a convolutional neural network. In some implementations, the generator 202 is a deconvolutional neural network
...
[0035] At step 308 , the orchestrator 206 provides a training set image 201 to the discriminator 204 to classify the training set image 201 . As with images from the generator 202 , the discriminator 204 processes an image through the neural network of the discriminator 204 to generate a classification classifying an image as either “real” or “fake.”

...
[0043] At step 354 , the trained discriminator 204 feeds the input image through the neural network layers of the discriminator in sequence. More specifically, as described elsewhere herein, each neural network layer accepts an input from either a previous layer or an input to the network (e.g., the input image itself), processes that input through the layer, and provides an output either to a subsequent layer or to the output of the network itself (e.g., as a classification). In examples, the discriminator 204 is implemented as a convolutional neural network, which includes one or multiple convolutional layers that perform convolution operations.


But they are very short on detail.

I don't think they would recognize a spike if they sat on one. .

 

[Deleted member 118]

One more for ya @Fact Finder

$350 billlion a year

Cybersecurity Market Expected to Be Worth $345.4 billion by 2026 - Exclusive Report by MarketsandMarkets™

The major factors fueling the Cybersecurity market include rising frequency and sophistication of target-based cyber attacks, increasing demand for the...

www.globenewswire.com

 

[Deleted member 118]

How about this as well video surveillance

A small $50 billon

Video Surveillance Market Size, Forecast and Industry Trends | 2030

The global Video Surveillance Market in terms of revenue was estimated to be worth $53.7 billion in 2023 and is poised to reach $83.3 billion by 2028, growing at a CAGR of 9.2% during the forecast period.

www.marketsandmarkets.com

 

[Deleted member 118]

Are we at $10 trillion year yet @Fact Finder

$35 billion

Global Smart Home Appliances Market 2024-2028 - Research and Markets

This report features 18 companies, including Koninklijke Philips N.V., Whirlpool Corp., Apple Inc., Sharp Corp., Panasonic Holdings Corp., BSH Hausgerate GmbH

www.researchandmarkets.com

 

[Fact Finder]

One more for ya @Fact Finder

$350 billlion a year

Cybersecurity Market Expected to Be Worth $345.4 billion by 2026 - Exclusive Report by MarketsandMarkets™

The major factors fueling the Cybersecurity market include rising frequency and sophistication of target-based cyber attacks, increasing demand for the...

www.globenewswire.com

@Rocket577 I give up OK.

I just have to accept that Brainchip is a once in a lifetime opportunity never to be repeated in my particular lifetime.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Deleted member 118]

@Rocket577 I give up OK.

I just have to accept that Brainchip is a once in a lifetime opportunity never to be repeated in my particular lifetime.

My opinion only DYOR
FF

AKIDA BALLISTA

One more before I fall asleep $80 billion lol

Global Wearable Medical Devices Markets, 2021-2026 - Increasing Penetration of AI and Big Data in Healthcare to Create Smart and Adaptive Wearable Devices

/PRNewswire/ -- The "Wearable Medical Devices: Technologies and Global Markets 2021-2026" report has been added to ResearchAndMarkets.com's offering. The...

www.prnewswire.com