BRN Discussion Ongoing

[Shadow59]

These researchers have found another use for an SNN Chip:

“1 Introduction
Cloud services cause an exponentially growing traffic in data centers. This requires optical transceivers to operate at lower power and lower cost. Because digital signal processing (DSP) has high power consumption, recent research envisions replacing the DSP in part with an analog frontend with lower power
consumption….

“Conclusions
In this work, we have implemented an SNN on the mixed-signal neuromorphic hardware system BrainScaleS-2 (BSS-2). While we observed a hardware penalty slightly below 1dB at a BER of 2×10−3, the BSS-2 SNN outperforms a simulated linear equalizer with the same number of taps, thanks to nonlinear processing.

The presented results confirm that neuromorphic hardware can provide the reliability required by signal processing in optical transceivers.

Promising directions
for future research include reducing the hardware penalty, reducing the architec- tural complexity and increasing the intrin- sic speed of the SNN, and optimizing the input and output interfaces of the SNN. Furthermore, processing by the SNN must be parallelized. Also, the effective power consumption of neuromorphic signal pro- cessing should be analyzed and compared to digital processing.
In a future hardware implementation, spikes could be generated directly from the electrical signal coming from the photo diode, thereby replacing the power-hungry analog-to-digital converter (ADC). In this work, we have considered hard decision demapping. Another interesting direction is the design of an SNN equalizer/demap- per with soft output.”

About Brainscale:

4 BrainScaleS-2 System

We deploy our SNN equalizer/demapper on the accelerated mixed-signal neuromor phic BSS-2 system, developed at Heidel- berg University [5] in Germany.

On its analog neural network core, it emulates up to 512 LIF neurons and 128 k synapses in analog circuits in parallel and in continu- ous time. Hardware synapses have 6-bit weights, which can be configured as in- hibitory (negative sign) or excitatory (pos- itive sign). The neurons communicate via digital spike events. Each neuron is parameterized individually.

https://arxiv.org/pdf/2206.00401

My opinion only DYOR
FF

AKIDA BALLISTA

This isn't Akida. I think it's based on Spinnaker

 

[Violin1]

Agree. Totally agree.


I have had such angst about U.S. participation in our companies prospects I am at a loss for words. What and where is Integrous??? Why is the take up in ADR shares virtually non-existent? Something is definately missing here, and I cannot put my finger on it.

I recently discussed that one of Sean's stated goals is to get the word out about Brainchip. My thinking then and now, is that the word needs to be taken directly to the investment community in the United States.

At this point I am so frustrated that I would be even willing to have Sean, or Jerome, or Rob, or Anil, etc. make an appearence on the Mad Money clown show. Something .....anything. Regards, dippY

Hey dippy - don't get distracted now when we are closing in on the prize. Sean's focus is getting the word out to users not so much investors. Sure, they'll do presentations to Shaws etc but from the AGM and last few months he is really focussed on getting the tech adopted at the ground level. Many of the ways our IP is used we'll never see or know. That's his strategy - in everything, ubiquitous, all-pervading. The ADRs have been disappointing but really we need to see the start of the revenue acceleration and the company always said it'll still be a while. I'm not expecting much this quarter, a little the next, a bit more in the December quarter and then more products using IP in 2023 which drives revenue. I still jiggle from foot to foot because I want it now, particularly the porsche, but by the time we take off at least there'll be an electric porsche and a new electric MB to choose between!

 

[GDJR69]

FF your cross examination skills are at another level. I hope you have a good book to read, he won't be back for a while.

I think that's more final submissions than cross examination but the sentiment nevertheless remains valid

 

[GDJR69]

In a significant decision handed down this week, the High Court has confirmed that parties who create a Facebook page are potentially liable for defamatory comments posted to that page by third parties as if they themselves were the publisher.9 Sept 2021

In a significant decision handed down this week, the High Court has confirmed that parties who create a Facebook page are potentially liable for defamatory comments posted to that page by third parties as if they themselves were the publisher.9 Sept 2021

This will encourage moderators to do something about trolls.

 

[Fact Finder]

This isn't Akida. I think it's based on Spinnaker

Yes that’s right but did you not read my post this is another use for an SNN chip. Brainscale is not a commercially available chip it is research only.

So logic says if you want to do this who do you go and see someone with a COTS chip or IP.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Deadpool]

HC the Bermuda Triangle of the investment world best to steer a course around that obstacle lest the siren call will rob you of your sanity.

Here too the waters will become less clear and reefs will be obscured and new participants with friendly countenance and intriguingly seductive questions which do not matter at all will ask the what if’s that subtly erode confidence.

They will claim experience that allows them to provide answers to questions they raise and the answers will be negative or designed to erode confidence.

They will throw up research and claim it points to being on the wrong course all the time ignoring the achievements of the company.

They will never however answer the single most important question which is “why will Brainchip not achieve a tiny little one percent of the world wide addressable market for Edge Ai.”

ARM alone caused 29 billion semiconductors to be shipped in the last 12 months.

One percent of this is 290 million and at a 10 cent royalty per semiconductor that is $29 million which is north of twice Brainchip’s spend making it a profitable company.

Until someone can tell us why Brainchip will not achieve one percent of the World wide market for Edge Ai said to be 70 billion US dollars by 2025 when they already are trusted by:

NASA
MERCEDES BENZ
VALEO
MEGACHIPS
RENESAS
ARM
EDGE IMPULSE and
SiFIVE

they are no more than white noise.

By the way one percent of 70 billion US dollars is 700 million US dollars.

Annual outgoings at Brainchip are about $12 million so even if we bump that up to $50 million that is still $650 million in profit.

My opinion only DYOR
FF

AKIDA BALLISTA

Fact Finder down ramping again

 

[Bravo]

This is a great read if you want to understand more about how Driver Monitoring Systems are set to be more stringently evaluated moving forward.

Driver monitoring gets spotlight amid safety questions

Two leaders in motor vehicle safety testing said Thursday that they will rate the driver monitoring systems that are supposed to help make technologies like Tesla's Autopilot safe.

edition.cnn.com

 

[Earlyrelease]

I should add that while Bosch is a truly great company last year they sat at number 6 in a list of global providers of semiconductors to the automotive industry and Renesas was at number 3.

Once again Renesas who also had an Ai Edge research program jumped on board the Brainchip Bus and the one percent question raises its ugly head.

My opinion only DYOR
FF

AKIDA BALLISTA

Hey FF
I think you have done enough to convince those similar conservative LTH that we should be starting to quote at least 2% market share

 

[Shadow59]

Hey FF
I think you have done enough to convince those similar conservative LTH that we should be starting to quote at least 2% market share

LOL... the difference between 1 and 2% doesn't sound much until you realise that it is double!

 

[Slymeat]

BrainChip out with, interestingly, a quite easy to read "white Paper". It's only about a 5 minute read and nicely explains how Akida saves power, processes data, and (to a degree) how Akida learns.

Well worth the read.

What Is the Akida Event Domain Neural Processor?​

As an example of the simple language employed here's an excerpt documenting the difference between how traditional AI and Akida process a piece of blank paper. It is a nice way of demonstrating the true worth of sparsity - i.e. the way the brain also works.

"consider an extreme case. You could give a “standard” Convolutional Neural Network a blank page to process, and it will take every pixel and process it through millions of multiply-accumulate instructions to find out that the page is blank.The Akida event-based processing method works like how the human brain would process a blank page: since there are no lines or colors on the page, it receives no events, so it does not need to process anything. It is this reduction of data that must be processed, known as “sparsity,” that leads to significant power savings."

 

[hotty4040]

LOL... the difference between 1 and 2% doesn't sound much until you realise that it is double!

Shadow, you must have been top of the class at Maths I reckon,


Akida ballista

h...

 

[Reuben]

Looks like chip crunch is slowly easing,

 

[Shadow59]

Shadow, you must have been top of the class at Maths I reckon,


Akida ballista

h...

Class of 1 too! Lol.... made it to the top most years, there were a couple of times it was a near thing though!

 

[Fullmoonfever]

I wonder if the communications company Sean Hehir the CEO referred to that they are engaged with currently takes the time to read NASA SBIR
applications.

If so this might have caught their attention and been the driver of a telephone call to Brainchip to have a chat:

"The NECR technology can be integrated into commercial communication systems to enhance cognitive sensing and communication capability."

Then again this would just kick sand in the face of those over at HC who denied that there was any commercial value to be had with engaging with NASA.

My opinion only DYOR
FF

AKIDA BALLISTA

Was having a dig around on the NASA NECR SBIR to see if any changes anywhere as proposed Ph 1 end date was Nov 21.

Last comments on TSE I find were FF so thought reply to that.

Anyway, some snips off US Gov procurement sites I use.

If I read right (happy to be corrected), the Ph 1 did complete as proposed, the contract does not appear to have been closed (highlighted) and also further down is flagged multi year (highlighted).

So, to me, I like to think this may indicate moving to Ph 2 at some stage? Thoughts?

 

[TECH]

Hey @chapman89 if you happen to be catching up with Ken in Sydney tonight, please pass on that we
WA shareholders really appreciated getting to spend some time with him last Friday night and that we hope he has
a safe flight home to the US...he fly's back either tonight or Saturday...hope the plane gets airborne with all that WA
wine he picked up down Margaret River way

He is a man of many talents, and we are grateful that he represents "Team Brainchip"

 

[Wags]

hope your better soon Macca , well said and we all know FF's 1% comment was stating the obvious (and it is at least 1%)

thanks MD. You know that Manflu, I should be able to milk it for another week or so, haha

Yep, looks like another week or so, COVID confirmed this morning.
I doubt there is any connection, but just a heads up I was at the AGM and Pub and NVISO demo.

 

[HopalongPetrovski]

Yep, looks like another week or so, COVID confirmed this morning.
I doubt there is any connection, but just a heads up I was at the AGM and Pub and NVISO demo.

Look after yourself Macca. They say that even if symptoms are not too bad to get plenty of rest and chicken soup so that you get a full and complete recovery. Otherwise, you might just get a visit from Angela.

 

[Baisyet]

 

[Fastback6666]

I am looking forward to the ASX200 rebalance announcements this afternoon - I believe they come out after market.

It will make for some interesting trading on Monday for BRN if it has made it in the ASX200.

Is this the last chance to get quite cheap shares below $1.10??? Hmmm

 

[Murphy]

LOL... the difference between 1 and 2% doesn't sound much until you realise that it is double!

Interesting thing I only just learned about percentages - they work in reverse exactly the same! If, for example, you were trying to work out 8% of 25, then just reverse it to 25% of 8 and it becomes easy! Life hack I only just learned about last week!!


If you don't have dreams, you can't have dreams come true!