BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Fullmoonfever
Top 20
So, like most, I've been checking out diff partners sites when I get the chance and this guy popped up on DeGirum from a couple of days ago.
Profile isn't public but full name is, so I searched that and this guy below stood out as a co-author on a paper from late last year titled "Spiking neural networks for autonomous driving: A review".
It was also co-authored by someone affiliated with VW and did contain sections on Akida as well as some other processors so I presume, rightly or wrongly, it could be this R Parada.
If so, it would appear he would like to get his hands on or run some models / code on Akida.
DeGirum can assist by the looks with their hosting set up.
Maybe BRN HO shpuld flip him a PCIe or M.2 to play and research with.
Further details of the paper etc below.
community.degirum.com
es.linkedin.com
Tha paper published Oct last year (didn't search if previously posted here).
Profile isn't public but full name is, so I searched that and this guy below stood out as a co-author on a paper from late last year titled "Spiking neural networks for autonomous driving: A review".
It was also co-authored by someone affiliated with VW and did contain sections on Akida as well as some other processors so I presume, rightly or wrongly, it could be this R Parada.
If so, it would appear he would like to get his hands on or run some models / code on Akida.
DeGirum can assist by the looks with their hosting set up.
Maybe BRN HO shpuld flip him a PCIe or M.2 to play and research with.
Further details of the paper etc below.
Create customized SNN code
Hi, I’m interested in the neuromorphic hardware available, Akida. How can I create my own code to be launched in the platform? Thanks
community.degirum.com
Raúl Parada Medina, PhD - CTTC | LinkedIn
IoT research specialist within the connected car project at a Spanish automobile… · Experiencia: CTTC · Educación: Universitat Pompeu Fabra - Barcelona · Ubicación: Castelldefels · Más de 500 contactos en LinkedIn. Ver el perfil de Raúl Parada Medina, PhD en LinkedIn, una red profesional de más...
Tha paper published Oct last year (didn't search if previously posted here).
Dylan is/was the operations manager of Synsense prior to its merger with Inivation
https://research-repository.uwa.edu.au/en/persons/dylan-muir
Smoothsailing
Regular
Thanks Diogenese , He has an exceptional resume in the field of Neuromorphic compute, interesting Dylan now resides in Perth. I wonder if he and Peter were mates or colleagues at some point.
Last edited:
Morning Chippers ,
This chap has been making the news headlines of late.
Headsets for the millitary , on this occasion, but Anduril dabble in alot of defence related products.
Regards,
Esq.
This chap has been making the news headlines of late.
Headsets for the millitary , on this occasion, but Anduril dabble in alot of defence related products.
Regards,
Esq.
Smoothsailing
Regular
BrilliantView attachment 77536
The Nintendo Switch’s Joy-Con drift problem, explained
Everything you need to know about Joy-Con driftwww.theverge.com
What is “Joy-Con drift”?
Joy-Con drift is an issue that Switch owners have been encountering over the last few months that causes the analog sticks on the controllers to randomly move around and input commands to the console, even when they’re not being physically moved.
Nintendo Switch 2 could have an unbelievable controller feature — what we know
The console may be able to predict your button presses before you make themwww.tomsguide.com
Pre-cognitive controllers sound a little like science fiction, but the patent details a software feature that would learn from your gameplay. Over time it seems like the Switch 2 would learn from your gaming patterns and use that information to speed up button presses and other input.
It sounds a little unbelievable, but with AI on hand it might not be that unrealistic. However this sort of tech has the potential to cause problems.
The first big problem is that this feature would have to track and analyze all your gameplay in order to make these predictions. Which is something that a bunch of people may not want to see happening. There’s also the question of what happens if the console gets it wrong, and tries to initiate a button press that you didn’t intend on making.
Did I predict this Nintendo switch controller feature?
Does this patent potentially solve drift?
TheDrooben
Pretty Pretty Pretty Pretty Good
View attachment 77536
The Nintendo Switch’s Joy-Con drift problem, explained
Everything you need to know about Joy-Con drift
What is “Joy-Con drift”?
Joy-Con drift is an issue that Switch owners have been encountering over the last few months that causes the analog sticks on the controllers to randomly move around and input commands to the console, even when they’re not being physically moved.
Nintendo Switch 2 could have an unbelievable controller feature — what we know
The console may be able to predict your button presses before you make them
Pre-cognitive controllers sound a little like science fiction, but the patent details a software feature that would learn from your gameplay. Over time it seems like the Switch 2 would learn from your gaming patterns and use that information to speed up button presses and other input.
It sounds a little unbelievable, but with AI on hand it might not be that unrealistic. However this sort of tech has the potential to cause problems.
The first big problem is that this feature would have to track and analyze all your gameplay in order to make these predictions. Which is something that a bunch of people may not want to see happening. There’s also the question of what happens if the console gets it wrong, and tries to initiate a button press that you didn’t intend on making.
Did I predict this Nintendo switch controller feature?
Does this patent potentially solve drift?
Sony also looking at predictive gameplay........
Happy as Larry
7für7
Top 20
New partner already posted?
multicorewareinc.com
Accelerated Software & Development Company | MulticoreWare
MulticoreWare is an accelerated software development company with products in broadcast, OTT, autonomous vehicles, and video analytics.
multicorewareinc.com
Last edited:
suss
Regular
Partner = enabler/channel for commercial products
How many of these partners have resulted commercial products? Until this happens, we're all just in a holding pattern.. The nervousness is understandable....
Bravo
If ARM was an arm, BRN would be its biceps💪!
With any luck this standardized benchmarking framework will help BrainChip to validate Akida's performance and efficiency compared with conventional methods.
www.nature.com
The neurobench framework for benchmarking neuromorphic computing algorithms and systems - Nature Communications
Brain-inspired neuromorphic algorithms and systems have shown essential advance in efficiency and capabilities of AI applications. In this Perspective, the authors introduce NeuroBench, a benchmark framework for neuromorphic approaches, collaboratively designed by researchers across industry and...
We have entered that "period again"...the no solid news, but the solid speculative news !
Does anybody here think that Peters et al technology is being purposely held back, so other unnamed companies can "catch up" ?
The fact no company has signed a solid license agreement really bothers me, yes I know all about Renasas & MegaChips, but that in was 2020, wasn't it ?
Currently, the trading, (manipulation) pattern is so, so obvious, please don't give your shares away, that's not financial advice, that's " common sense"..hang tough, it's a real tug of war between "good and greedy".
No disrespect to my friend Peter, but I personally like how Tony speaks his truth on Linkedin and conducts himself on/in podcasts and interviews, he's honest and positive, and in many ways he is really blessed to have been given this opportunity from both Peter and Anil...when we "finally" succeed, please remember to acknowledge my friends, without them, Brainchip wouldn't exist.
Fact.....Akida we allu....Tech.
Yes, absolutely.
eg It's hard to understand the delay in signing up the sub-contractor for the AFRL deal.
I asked TD if BRN would announce the signing when it happens on the ASX, and he replied that it most likely would.
But with the delay lengthening, the same thought came to my mind, that RTX, or whoever the subcontractor might be, are not as committed as the AFRL.
This pattern seems to be same as with MB, Renesas, and other companies who are not supporting us financially, the very same companies who, ironically, suffered badly with supply chain issues in the past, and who, with a bit of work, could own their own chip that could be manufactured in multiple countries.
Clearly potential customers wanting to trial us would be also looking at what alternatives like NVIDIA have to offer with non-neuromorphic NPU's
I'd say that we are probably being held back by the constant fear amongst our potential clients that mega companies like NVIDIA and INTEL will destroy us or anyone wanting to use us in the future. You know the (unwritten) spiel - If you add an Akida chip to an NVIDIA system, then you will lose all the NVIDIA support you are currently receiving.
Nvidia Federal Litigation Filings - Company Legal Profiles
Justia - Company Profiles
companyprofiles.justia.com
Sean is a bit to blame, as his vision of us surviving in a few thousand drive-thru's or a few thousand satelites never looked feasible, especially since Akida was to be everywhere. Similarly the whole concept of an edge-box seems like waving the white flag: It's not up to BRN to find a use case for akida, so here, try yourself...
Whoever on the board stated that NVIDIA wasn't going to be a competitor but rather a partner, clearly doesn't know NVIDIA.
It will always seem impossible to break into the old boys club and I would love to see BRN do a joint venture in a product that goes up against a popular non-neuromorphic ai product. We retail investors are a hardy lot, so let the games begin.
Smoothsailing
Regular
Games have been going on for years getting sick of them that’s for sure, however I still believe we will hear an announcement prior to AGM.Yes, absolutely.
eg It's hard to understand the delay in signing up the sub-contractor for the AFRL deal.
I asked TD if BRN would announce the signing when it happens on the ASX, and he replied that it most likely would.
But with the delay lengthening, the same thought came to my mind, that RTX, or whoever the subcontractor might be, are not as committed as the AFRL.
This pattern seems to be same as with MB, Renesas, and other companies who are not supporting us financially, the very same companies who, ironically, suffered badly with supply chain issues in the past, and who, with a bit of work, could own their own chip that could be manufactured in multiple countries.
Clearly potential customers wanting to trial us would be also looking at what alternatives like NVIDIA have to offer with non-neuromorphic NPU's
I'd say that we are probably being held back by the constant fear amongst our potential clients that mega companies like NVIDIA and INTEL will destroy us or anyone wanting to use us in the future. You know the (unwritten) spiel - If you add an Akida chip to an NVIDIA system, then you will lose all the NVIDIA support you are currently receiving.
Nvidia Federal Litigation Filings - Company Legal Profiles
Justia - Company Profiles
Sean is a bit to blame, as his vision of us surviving in a few thousand drive-thru's or a few thousand satelites never looked feasible, especially since Akida was to be everywhere. Similarly the whole concept of an edge-box seems like waving the white flag: It's not up to BRN to find a use case for akida, so here, try yourself...
Whoever on the board stated that NVIDIA wasn't going to be a competitor but rather a partner, clearly doesn't know NVIDIA.
It will always seem impossible to break into the old boys club and I would love to see BRN do a joint venture in a product that goes up against a popular non-neuromorphic ai product. We retail investors are a hardy lot, so let the games begin.
uiux
Regular
Why I am stoked about the Iron Dome:
Post in thread 'America's Iron Dome' https://thestockexchange.com.au/threads/americas-iron-dome.260527/post-450604
Post in thread 'America's Iron Dome' https://thestockexchange.com.au/threads/americas-iron-dome.260527/post-450604
Hope so but this time last year we were all hoping for the same thing. The SP spiked from 16c to 48c in the weeks prior to the AGM on this same hope. Then the AGM was released with nothing new was announced and the SP slid back down to 16c after two months.
I guess the BoM can't really time their announcements to fit in with AGM's - if a deal comes in, it has to be announced straight away - so the ann. would have to be a matter of change of BRN strategy going forward.
Smoothsailing
Regular
I disagree with your analogy and believe they are steering the ship not the shareholders and can time their announcements accordingly to a degree that works best for the company.
Last edited:
toasty
Regular
They cannot time their announcements to suit the company. The regulations are quite clear about what can be announced and when. If it is a material matter, the board has no choice, except in some special circumstances. My current view is that there are a lot of parties interested but noone is yet willing to stump up the cash. The prospects may well be in the "noone ever got fired for buying big blue" stage...........
Last edited:
rgupta
Regular
If new sub contractor is still not assigned then how brainchip told us on asx that subcontractor costs can be around $800000 for the project.Yes, absolutely.
eg It's hard to understand the delay in signing up the sub-contractor for the AFRL deal.
I asked TD if BRN would announce the signing when it happens on the ASX, and he replied that it most likely would.
But with the delay lengthening, the same thought came to my mind, that RTX, or whoever the subcontractor might be, are not as committed as the AFRL.
This pattern seems to be same as with MB, Renesas, and other companies who are not supporting us financially, the very same companies who, ironically, suffered badly with supply chain issues in the past, and who, with a bit of work, could own their own chip that could be manufactured in multiple countries.
Clearly potential customers wanting to trial us would be also looking at what alternatives like NVIDIA have to offer with non-neuromorphic NPU's
I'd say that we are probably being held back by the constant fear amongst our potential clients that mega companies like NVIDIA and INTEL will destroy us or anyone wanting to use us in the future. You know the (unwritten) spiel - If you add an Akida chip to an NVIDIA system, then you will lose all the NVIDIA support you are currently receiving.
Nvidia Federal Litigation Filings - Company Legal Profiles
Justia - Company Profiles
Sean is a bit to blame, as his vision of us surviving in a few thousand drive-thru's or a few thousand satelites never looked feasible, especially since Akida was to be everywhere. Similarly the whole concept of an edge-box seems like waving the white flag: It's not up to BRN to find a use case for akida, so here, try yourself...
Whoever on the board stated that NVIDIA wasn't going to be a competitor but rather a partner, clearly doesn't know NVIDIA.
It will always seem impossible to break into the old boys club and I would love to see BRN do a joint venture in a product that goes up against a popular non-neuromorphic ai product. We retail investors are a hardy lot, so let the games begin.
Also if the team brainchip is so slow in such a big contract then what to expect from them?
That is where team brainchip jumps on a news without thinking ahead, how the market will perceive the same.
e.g last time about akida edge boxes, team brainchip start making noises somewhere mid 23 that it will be launched end 23 and even after end 24 we have no revenue from edge boxes.
To me it is a very communication, if it is right.
Dyor
Similar threads
