BRN Discussion Ongoing

[Deleted member 118]

Thank God he didn't ask to try them on first beacuse that would have been weird.




View attachment 2725

 

[thelittleshort]

Please note this post starts slowly and builds to a speculative crescendo

In the article - Item 2 under the heading "Safety-First Approach will Drive Innovation" mentions two systems:

2) Rise of partial hands-free operation: L3 is the first-level of autonomous driving which operates under specific conditions. Over the next few years, OEMs are planning to release vehicles with L3 systems, operating over predefined highways or locations. Known as “geofencing,” the vehicle defines spatial boundaries and references detailed maps of the surrounding terrain. The car projects sensor data onto the maps to determine the safest route. As with L2 systems, drivers must be ready to take control of the vehicle at all times. Examples of partial hands-free operation are GM’s Super Cruise or Ford’s BlueCruise. L3 systems that feature this level of automation will begin to outpace L2 systems in the latter half of the decade.

1. Super Cruise by General Motors
2. BlueCruise by Ford Motor Company

Any chance BrainChip is involved here? Obviously we are working with Ford, so could this also link us to GM by default?

I saw back in the ASX Announcement: Update for the March 2019 Quarter that both GM and Ford were mentioned. Looks like things were looser back then as the companies listed are not confirmed as being associated with BrainChip. However some of the other companies listed are now directly involved with BrainChip - well Valeo and Safran anyway, along with Ford

* How good would it be if the companies named in the ASX announcement was the actual list - and BrainChip are checking them off one by one

Other companies on the list discussed in this forum:

• Continental by numerous including @IndepthDiver
• Aptiv by @Space Cadet among others
• Micron by @Alfie among others
• Samsung by @butcherano among others
• Boeing by @Bravo77 among others
• Sony by @Taproot among others

View attachment 2621

With Boeing also on “the list”, does anyone know what happened to their Disruptive Computing and Networks division?

I imagine the 1000 Eyes looked into them back in 2018. They seemed to burst onto the scene but kind of peter out from what can see?

Hopefully because they quickly realised BrainChip had beaten them to the punch, with superior IP!



CHICAGO, Oct. 17, 2018 /PRNewswire/ -- Boeing [NYSE: BA] today announced a new Disruptive Computing and Networks (DC&N) organization to develop computing and communications solutions for advanced commercial and government aerospace applications.

By leveraging core technologies in quantum communications and computing, neuromorphic processing and advanced sensing, the new organization will enable Boeing to develop breakthrough solutions in secure communications, artificial intelligence and complex system optimization.

 

[Fact Finder]

I posted a reply in the MegaChips thread. They are a company that does fabless production and they acquire IP from several sources which they add to their catalog. This allows their customers to choose based on features they need, so I assume which they use will be driven by customer demand.

MegaChips lists the key features and applications of both solutions in the AI/ML category on their site. They are not replacing BrainChip with Quadric, just expanding the available choices.

Hi Delekto

Diogenes looked at their offering against AKIDA and made the very valid point that they strictly speaking do not both inhabit the edge as AKIDA knows it. The following evidences his point:

The quadric processor contains the first ever instance of our fully custom architecture.​

Learn More About The Architecture

256 Vortex Cores.​

The quadric processor contains a q16 Architecture instance featuring 256 Vortex Cores.

Highly Compatible.​

The power of software defined hardware. This processor can handle any algorithm you throw at it.

Low Power.​

The processor consumes 4.5W TDP. Embeddable in a wide range of edge applications
………………………………………
Low power means different things to different people.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Fact Finder]

They were for his head. FF

 

[yogi]

I read today the world wide automotive industry is worth 2.7 trillion dollars CURRENTLY. In its transition to EV much more will be spent each year through to 2040.

I have read that the automotive semiconductor slice of this pie for EV will be 200 billion by 2040. (This is for traditional uses but does not include Brain interface etc

I personally believe it cannot credibly be argued that Brainchip will not capture some of this outrageously huge market given the present knowledge regarding Mercedes, Valeo and Ford at least.

Being impossibly conservative such share has to be at least one tenth of one percent of 200 billion US dollars.

So just from automotive this would result in a measly annual 200 million US dollars in revenue by 2040.

Add to this one percent of the current 450 billion defence spending.

Add to this NASA revenue.

Add to this MegaChips revenue.

Add to this Renesas MCU revenue.

Add to this the medical technology market revenue via Biotome and NaNose.

As I said it all makes FF very happy.

My opinion only DYOR
FF

AKIDA BALLISTA

Hi FF
Below is what is coming up soon from Audi this car hasn't been showcased anywhere yet.

 

[thelittleshort]

With Boeing also on “the list”, does anyone know what happened to their Disruptive Computing and Networks division?

I imagine the 1000 Eyes looked into them back in 2018. They seemed to burst onto the scene but kind of peter out from what can see?

Hopefully because they quickly realised BrainChip had beaten them to the punch, with superior IP!



CHICAGO, Oct. 17, 2018 /PRNewswire/ -- Boeing [NYSE: BA] today announced a new Disruptive Computing and Networks (DC&N) organization to develop computing and communications solutions for advanced commercial and government aerospace applications.

By leveraging core technologies in quantum communications and computing, neuromorphic processing and advanced sensing, the new organization will enable Boeing to develop breakthrough solutions in secure communications, artificial intelligence and complex system optimization.

Good presentation by Charles Toups from Disruptive Computing & Networks organization about neuromorphic computing and quantum technologies in relation to Boeing and aerospace

Skip to 3:50

 

[Bravo]

I know it's a bit belated but Happy St Patrick's day everyone! Especially to all our Irish Brainers out there!

And before anyone asks, yes I did manage to hire my leprechaun boots out, but at a heavily discounted price, "mates rates" actaully seeing as it was my sister who needed to borrow them.

 

[Dallas]

Golem.de: IT-News für Profis

www.golem.de

 

[Dallas]

New article about BMW and autonomous driving☝

 

[AusEire]

I know it's a bit belated but Happy St Patrick's day everyone! Especially to all our Irish Brainers out there!

And before anyone asks, yes I did manage to hire my leprechaun boots out, but at a heavily discounted price, "mates rates" actaully seeing as it was my sister who needed to borrow them.


View attachment 2728
View attachment 2729

Lá Fhéile Pádraig sona duit

 

[Deleted member 118]

They were for his head. FF

Lol

 

[Diogenese]

I noticed megachips have a stake in quadric which they mentioned they will aim to use in their lidar. Given they have a stake in quadric, are they more inclined to use it over akida?

Hi JB,

Quadric make what they call a "Supercomputer" which integrates the functions of CPU/GPT/AI accelerator using a modified CNN for image classification.

Its basic unit seems to be their Vortex ALU.

It won't be used for doorbells.

The White Paper 2019​

May, 2019

https://www.quadric.io/post/the-white-paper

we founded Quadric to build a product that brings server-class performance to the edge.

By 2015, researchers were proving classification results on the Imagenet challenge that exceeded human error rates. One such advanced CNN network architecture, RESNET50 strikes a balance between total computational network complexity and error rate.

####################################################################################

https://brainchipinc.com/wp-content...brief_6-How-BrainChip-is-Changing-AI_v1.2.pdf
The Akida Event-Based Neural Processor
The Akida event-based neural processor is a fundamentally different approach that breaks the linear relationship between high power consumption and performance seen in traditional accelerators. The Akida processor is 10x to 30x more energy-efficient than its nearest competitor for inferencing on industry-standard benchmarks such as MobileNet and Google Keyword Spotting DNNs, and is easy to use. Trained on MobileNet’s Imagenet 1000 data set, the Akida neural processor can classify all 1.2 million images, and 1,000 classes, at 30 frames per second within a power budget of just 156 milliwatts in 28nm, compared to several watts for a Google Edge TPU. Audio keyword recognition using the Google keyword database runs at an extremely low power of 150 microwatts.

30 frames per second in 156 mW = 192 frames per second per Watt. (This is the 2019 Akida, not the improved 2021 commercial version - to be fair, the Quadric figures are from a 2019 white paper). So Quadric is in the same ballpark, but Akida has made home base, while Quadric just making 3rd.

The main difference is that Quadric is a Frankenstein amalgam of CPU, GPU and AI accelerator. Their AI accelerator uses ResNet50, an improved CNN arrangement which enables skip connections when intermediate layers are not required.

Also remember Akida's performance at key word spotting:

IPS = inferences (identified words) per second

 

[Diogenese]

Today I was dropping some things off to the local Salvation Army Op-Shop in my Toyota Echo and there was a guy who looked to be in his early twenties who asked the shop assistant if there were any undies in the store. The shop assistant said that it was unlikely, however if underpants were to be anywhere they'd be located in the far left corner and he added helpfully that there was an Aldi's next door and that they sold extremely cheap underwear.

Well, the young dude looked a bit crest-fallen but he continued his search and he actually succeeded in finding what may have been the only underpants in the entire op-shop and he bought them over to the counter whereupon the shop assistant exclaimed. "These are brand new! It must be your lucky day!"

"Oh, really?" said the dude "In that case, I'll leave them".

And that's a true story.

Oh dear!

 

[hamilton66]

Hi JB,

Quadric make what they call a "Supercomputer" which integrates the functions of CPU/GPT/AI accelerator using a modified CNN for image classification.

Its basic unit seems to be their Vortex ALU.

It won't be used for doorbells.

The White Paper 2019​

May, 2019

https://www.quadric.io/post/the-white-paper

we founded Quadric to build a product that brings server-class performance to the edge.

View attachment 2723


By 2015, researchers were proving classification results on the Imagenet challenge that exceeded human error rates. One such advanced CNN network architecture, RESNET50 strikes a balance between total computational network complexity and error rate.

View attachment 2724

####################################################################################

https://brainchipinc.com/wp-content...brief_6-How-BrainChip-is-Changing-AI_v1.2.pdf
The Akida Event-Based Neural Processor
The Akida event-based neural processor is a fundamentally different approach that breaks the linear relationship between high power consumption and performance seen in traditional accelerators. The Akida processor is 10x to 30x more energy-efficient than its nearest competitor for inferencing on industry-standard benchmarks such as MobileNet and Google Keyword Spotting DNNs, and is easy to use. Trained on MobileNet’s Imagenet 1000 data set, the Akida neural processor can classify all 1.2 million images, and 1,000 classes, at 30 frames per second within a power budget of just 156 milliwatts in 28nm, compared to several watts for a Google Edge TPU. Audio keyword recognition using the Google keyword database runs at an extremely low power of 150 microwatts.

30 frames per second in 156 mW = 192 frames per second per Watt. (This is the 2019 Akida, not the improved 2021 commercial version - to be fair, the Quadric figures are from a 2019 white paper). So Quadric is in the same ballpark, but Akida has made home base, while Quadric just making 3rd.

The main difference is that Quadric is a Frankenstein amalgam of CPU, GPU and AI accelerator. Their AI accelerator uses ResNet50, an improved CNN arrangement which enables skip connections when intermediate layers are not required.

View attachment 2731


Also remember Akida's performance at key word spotting:
View attachment 2726

IPS = inferences (identified words) per second

MSN

www.msn.com

D, food for thought. GLTA

 

[Bravo]

Oh dear!

No, that's the funny thing Dodgy-knees they weren't dear at all, actually they were really cheap! But even-so he didn't want to touch them with a 10 foot barge pole. I can only guess as to what the moral of this story is?

 

[Potato]

Just a lovely reminder of our management team

 

[GpHiggsBoson]

Hi all, I have obviously not been here for sometime.

I want you to know I will always own some Brainchip stock. I believe in the technology, I have been following for years. Yes it will have a role in future technology for sure, imo.

I am concerned with world affairs and ask you to recheck your risk to ratio analysis of your investment, no advice, imo, dyor.

We have had a fantastic journey. Many beautiful souls here and I wish you all well. Your are all intelligent and amazing people.

I am heading off for sometime but I wanted to wish you all the best before I go.

Hope to catch up in the future at some stage. May God bless you all and your families. Stay safe.

AKIDA BALLISTA!!

 

[Bravo]

Everyone has gone to bed. So annoying and boring. I am babysitting some echidnas so I can't go to sleep and I so wish it was Friday already.

 

[Diogenese]

Everyone has gone to bed. So annoying and boring. I am babysitting some echidnas so I can't go to sleep and I so wish it was Friday already.

Did you name one of them Spike?

 

[Bravo]

So now you're a clairvoyant as well as being the biggest brainiac I know! Phffff!