BRN Discussion Ongoing

[Esq.111]

M

morning fellow brners, I was told by a Brainchip exec that a vehicle has as many as 300 chips!

Does anyone know what that could be in calculations of royalties in relation to ip licence within these x number of cars?

@Esq.111 i think u have a good calculator or anyone else please care to help me out in my fogged brain

Morning MDhere,

Purely wild speculation from someone who knows very little technicaly...

@ $20.00 per physical chip ( as has been loosely used by management in the past).
÷ by 80 nodes ( per AKIDA chip)
= $0.25 per node .
So.... if we were only selling Interlectual Property as opposed to physical chips.

$0.25 per node - 10%((cost to produce in silicon) = $0.025) = leaves $0.2475 per node ( 90% profit margin).

So....$0.2475 per node x two nodes per application????

1. Voice/word detection/recognition x 1
2. Steering wheel touch/contact sensor x1
3. in car camera x2
4. external sensors x10
5. transmission/vibration x3
6. Wheel bearing vibration x4
7. Rain detection x 1
8. Auto headlights x1
&
9. Possibly throw in NICOBO the farting comfort robot from Panasonic, on the back seat. X 1.

= 24 use cases x 2 nodes per application = 48 nodes.

48 nodes ( incorporated into ecosystem partners chips)
48 x $0.2475 per node ( IP ). = $11.88 × 1.407631 ( USD to AUD ) = $16.7226 AUD per car x 2,000,000 cars per annum ( Mercedes-Benz) to keep it simple.

= $33,445,312.56 per annum. Potentially from Mercedes-Benz alone annually.

As stated above I have absolutely no background in this sphere & as such the above workings are purely guesstimates

NOT FINANCIAL ADVICE.

Regards,
Esq.

 

[Fullmoonfever]

Cool work with the drone.

I was having a look at GitHub info re the drone programming and maybe @Diogenese can elaborate or explain something I noticed as below.

The code appears using Linux (all good) platform (Mac?) but on AMD64 architecture (Intel) not ARM64?

I understand AMD64 is more PC based power hungry and ARM64 less power hungry. Both have strengths and weaknesses.

Am I reading that right and does that mean anything or just the architecture already within the drone that we coupled with via Akida showing no issue operability with Intel.

akida_examples/ci/.drone.yml at main · Brainchip-Inc/akida_examples

Brainchip Akida Neuromorphic System-on-Chip examples and documentation. - Brainchip-Inc/akida_examples

github.com

ARM vs X86 vs AMD64​

• ARM's processors are less power-consuming, but a bit slower.
• Intel's processors are always cutting-edge fast but at the cost of needing a sizeable power source.
• ARM's processors are more on the RISC side, so your highest-level instruction is limited to multiplication-tier operations level.
• Intel's processors are more on the CISC side, so your highest-level instructions like Compute Reciprocal of Square Root ie x-1/2 = 1/sqrt(x) are seemingly arbitrarily chosen.
• ARM is for mobile phones and stuff where you are on a battery and that battery does not carry "a lot of" charge.
• Intel is for regular PCs which hook up to a power outlet most of the time, and so power consumption is not a really big deal.
• ARM is not locked to a particular vendor like Intel/AMD are the X86/x64 market.
• ARM platform is more of a recipe where a company can pick and choose the functionality they want.
• ARM processors are designed for a power requirement at first, performance second.
• x86 chips work quickly but are harder to make, expensive and not as power-conscious.
• RISC chips require fewer transistors to function. With fewer transistors to power, RISC chips see power savings and heat reduction when compared to their CISC counterparts.

• ARM chips are used in heaps of other devices: routers, set-top boxes and smart TVs, smartwatches, some gaming devices, automotive infotainment systems and so on.
• The x86 has some very powerful instructions, the arm can still beat it in a fight.
• x86 can operate on direct memory as well.
• AMD64 supports 3D-Now and 3D-Now Extensions, in both 32-bit (legacy) and 64-bit (long) mode.
• AMD64 is designed to concurrently enable 32 and 64-bit processing with no loss of performance.
• Athlon 64 and Opteron are examples of some of the AMD processors that use the AMD64 architecture.
• AMD64 is the AMD-led update to the venerable x86 instruction set architecture, or ISA.
• AMD64-compatible desktop processor market is huge these days.
• AMD64-able mobile processors for laptop systems, something Intel has yet to offer.
• The AMD Athlon 64 adds a memory channel depending on the socket, twice or four times the cache memory depending on the model, and comparable and further frequencies, to beef up the performance of that model line compared to Sempron 64.

ARM vs X86 vs AMD64 | Geekboots

Know the differences between ARM, X86 and AMD64 processor - Geekboots

www.geekboots.com

 

[Fact Finder]

Something to consider that might not have jumped out and grabbed you by the scruff of the neck if you are not a @Diogenese is SCALABILITY.

If you have listened to Anil Mankar at the 2021 Ai Field Day you will recall he discussed with an audience member that if you are prepared to accept some latency increase you can use just ONE node of the AKIDA neural fabric.

The AKD1000 has 80 nodes and is designed to allow you to gang up to 64 AKD1000 should you need the power of 5,120 nodes.

What you may not be aware of is that 64 was chosen from a theoretically possible 1,024 AKD1000 chips being ganged together giving you the neuromorphic processing power of 81,920 AKIDA nodes.

When @uiux and @Zeebot were having the Nviso in cabin system demonstrated to them it was stated by Tim Llewellyn that they were not using anywhere near the full processing power of their one AKD1000 chip that you see in the vehicle and they intend to add further functions via this one chip.

So what is the point well clearly AKD1000 is ridiculously powerful and this is why the former CEO Mr. Dinardo stated that 64 AKD1000 chips was chosen because it would more than cover their target market.

However the AKIDA IP can run with 1 node right up to 81,920 nodes.

In a way what this scalability means is infinite product applications in fact imaging a product that might require 81,921 nodes is beyond my imagination.

So when Brainchip refers to a competitors product as not being scalable this is a HUGE barrier to that product taking market share from Brainchip.

My opinion only so DYOR
FF

AKIDA BALLISTA

Now here is something further to play around with and longer term holders will know that I posted a comparison of the Intel Loihi v Brainchip AKIDA styled 'Just how good is AKIDA' in 2020 which was lengthy and now lost in the mists of HC times but which I think can now be broken down into some simple figures now that most here have a pretty good grasp on the AKIDA REVOLUTION.

AKIDA runs up to 1024 AKD1000 chips before needing reengineering.

One AKD1000 has 1.2 million neuron equivalents and 10 billion synapse equivalents. So 1024 AKD1000's or 81,920 AKIDA nodes is 1billion 228million 800 thousand neuron equivalents and 1.24 trillion synapse equivalents.

Loihi 2 is fabricated on pre-production Intel 4 process and has a total of 1,048,576 artificial neurons and 120 million synapses.
(from 1 Oct 2021)

It can be seen that as you multiply out the number of Loihi chips to 1024 the gap between it and AKD1000 grows exponentially so that 1024 Loihi 2 x 1024 becomes 1 billion 73 million 741 thousand 824 neuron equivalents and 12.288 billion synapse equivalents.

So AKIDA ends up over 100 million ahead on neurons and over 1 trillion synapse ahead. The more you add the wider the gap.

Then throw in the admitted fact that Mike Davies in the following interview based article has advised that producing Loihi 2 in volume carries problems still to be resolved it becomes clear that this is indeed a one horse race.

Intel's Loihi 2 speeds effort to make neuromorphic chips like human brains

The chip also is a key product in Intel's plan to reclaim its processor manufacturing prowess.

www.cnet.com

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Diogenese]

Thanks for doing this I tried to do it for @Zedjack but could not as I know many do not like to open links. Obviously I failed but I will say again it is a must watch.

FF

AKIDA BALLISTA

PS: Imagine lying by the pool on your Super yacht and saying beer please and your personal drone flying off to get the beer and bringing it back and placing it in your hand.

it would blow the head off.

 

[Fact Finder]

it would blow the head off.

OK "Beer please and do not disturb the head." LOL FF

 

[Diogenese]

That drone demo made me think, as it is using single word commands that could occur in everyday situations. When controlling similar devices, we will need to be attentive to their “desire” to respond.

I found the response to the “go” command to be intriguing. The drone recognised the command but seemed to imply, by its response, a “go where?“. And I thought exactly the same.

I wonder what the world will be like when we are surrounded by devices using Akida voice/gesture/eye movement detection etc. even smells. We will have to start being very careful about some of the things we do and say in our normal day activities As devices may imadvertently jump to attention and carry out our unintentional commands.

One can only imagine the unexpected things that could result from utterances and gestures made whilst, say, watching sport on TV.

Maybe this was an early prototype for the video to demonstrate responsiveness to commands.

The production model may be initiated by "Simon says ...".

 

[GStocks123]

Wow !!

Epic!

 

[MADX]

In business, a partnership usually implies co-ownership. We (BRN) are announcing partnerships e.g. with ARM.

Would it be good to henceforth call these as "collaborator-partnerships"?

 

[AARONASX]

sorry if already shared!

Press Releases Archive - Digital Journal

A global media and thought leadership platform that elevates voices in the news cycle. Online since 1998.

www.digitaljournal.com

 

[Rise from the ashes]

BRN shared on LinkedIn

 

[Bravo]

 

[Fullmoonfever]

Cool work with the drone.

I was having a look at GitHub info re the drone programming and maybe @Diogenese can elaborate or explain something I noticed as below.

The code appears using Linux (all good) platform (Mac?) but on AMD64 architecture (Intel) not ARM64?

I understand AMD64 is more PC based power hungry and ARM64 less power hungry. Both have strengths and weaknesses.

Am I reading that right and does that mean anything or just the architecture already within the drone that we coupled with via Akida showing no issue operability with Intel.

akida_examples/ci/.drone.yml at main · Brainchip-Inc/akida_examples

Brainchip Akida Neuromorphic System-on-Chip examples and documentation. - Brainchip-Inc/akida_examples

github.com

View attachment 7771

ARM vs X86 vs AMD64​

• ARM's processors are less power-consuming, but a bit slower.
• Intel's processors are always cutting-edge fast but at the cost of needing a sizeable power source.
• ARM's processors are more on the RISC side, so your highest-level instruction is limited to multiplication-tier operations level.
• Intel's processors are more on the CISC side, so your highest-level instructions like Compute Reciprocal of Square Root ie x-1/2 = 1/sqrt(x) are seemingly arbitrarily chosen.
• ARM is for mobile phones and stuff where you are on a battery and that battery does not carry "a lot of" charge.
• Intel is for regular PCs which hook up to a power outlet most of the time, and so power consumption is not a really big deal.
• ARM is not locked to a particular vendor like Intel/AMD are the X86/x64 market.
• ARM platform is more of a recipe where a company can pick and choose the functionality they want.
• ARM processors are designed for a power requirement at first, performance second.
• x86 chips work quickly but are harder to make, expensive and not as power-conscious.
• RISC chips require fewer transistors to function. With fewer transistors to power, RISC chips see power savings and heat reduction when compared to their CISC counterparts.
• ARM chips are used in heaps of other devices: routers, set-top boxes and smart TVs, smartwatches, some gaming devices, automotive infotainment systems and so on.
• The x86 has some very powerful instructions, the arm can still beat it in a fight.
• x86 can operate on direct memory as well.
• AMD64 supports 3D-Now and 3D-Now Extensions, in both 32-bit (legacy) and 64-bit (long) mode.
• AMD64 is designed to concurrently enable 32 and 64-bit processing with no loss of performance.
• Athlon 64 and Opteron are examples of some of the AMD processors that use the AMD64 architecture.
• AMD64 is the AMD-led update to the venerable x86 instruction set architecture, or ISA.
• AMD64-compatible desktop processor market is huge these days.
• AMD64-able mobile processors for laptop systems, something Intel has yet to offer.
• The AMD Athlon 64 adds a memory channel depending on the socket, twice or four times the cache memory depending on the model, and comparable and further frequencies, to beef up the performance of that model line compared to Sempron 64.

ARM vs X86 vs AMD64 | Geekboots

Know the differences between ARM, X86 and AMD64 processor - Geekboots

www.geekboots.com

Ok. Appears it is for the target operating system so presume the drone is using AMD64?

Platform​

Use the platform section to configure the target operating system and architecture and routes the pipeline to the appropriate runner. If unspecified, the system defaults to Linux amd64.

Example Linux amd64 pipeline:

 

[MADX]

Marc Kennis from Stocksdownunder with Brainchip CEO Sean Hehir

https://stocksdownunder.com/video/i...Facebook&utm_medium=post&utm_campaign=organic

WOW. I was at the AGM and there was some negativity at how Sean delivered his speech by reading it out. Have look at his fluidity of delivery on the link. He's got his job crystal clear in his brain. I've commented elsewhere how PVDM looked pleased when Sean got up to speak at the meeting and regards him as his golden-hehir'd discovery. I'm a fan too. (Am I getting the hint about his hair through or should I leave it alone?)

 

[Esq.111]

OK "Beer please and do not disturb the head." LOL FF

Afternoon Fact Finder ,

Had a funny experience, probably 10 years ago, now.

I was sitting in the bar at our local marina having a quiet beer, when I noticed down on the pontoon one of my father's mates release all the lines to his boat without jumping aboard first.
Watching his boat ($3 mill plus) drift out some 25 plus metres from the dock with knowone aboard, I ran down ready to jump in and save it .

The owner, smiling, then proceed to pull a remote controller from his pocket.

Standing on the dock, he proceeded to rotate his boat ( 60 odd footer with some 2000 horsepower, twin screw) 180 degrees , and backed it up to dock , where he casually stepped onto the back deck and departed.

All with a tiny remote.

Amazing.

Regards,
Esq.

 

[HopalongPetrovski]

That drone demo made me think, as it is using single word commands that could occur in everyday situations. When controlling similar devices, we will need to be attentive to their “desire” to respond.

I found the response to the “go” command to be intriguing. The drone recognised the command but seemed to imply, by its response, a “go where?“. And I thought exactly the same.

I wonder what the world will be like when we are surrounded by devices using Akida voice/gesture/eye movement detection etc. even smells. We will have to start being very careful about some of the things we do and say in our normal day activities As devices may imadvertently jump to attention and carry out our unintentional commands.

One can only imagine the unexpected things that could result from utterances and gestures made whilst, say, watching sport on TV.

 

[Deleted member 118]

I wonder if BrN or Dyson are aware of one another

Dyson working on home robots

The company is setting up a major robotics research centre in the UK.

www.bbc.com

 

[TheFunkMachine]

Gee, you were quick off the mark, I just posted it!

Is this you in the interview JK? If so, thanks for doing what you do and asking the hard questions!

 

[Diogenese]

Cool work with the drone.

I was having a look at GitHub info re the drone programming and maybe @Diogenese can elaborate or explain something I noticed as below.

The code appears using Linux (all good) platform (Mac?) but on AMD64 architecture (Intel) not ARM64?

I understand AMD64 is more PC based power hungry and ARM64 less power hungry. Both have strengths and weaknesses.

Am I reading that right and does that mean anything or just the architecture already within the drone that we coupled with via Akida showing no issue operability with Intel.

akida_examples/ci/.drone.yml at main · Brainchip-Inc/akida_examples

Brainchip Akida Neuromorphic System-on-Chip examples and documentation. - Brainchip-Inc/akida_examples

github.com

View attachment 7771

ARM vs X86 vs AMD64​

• ARM's processors are less power-consuming, but a bit slower.
• Intel's processors are always cutting-edge fast but at the cost of needing a sizeable power source.
• ARM's processors are more on the RISC side, so your highest-level instruction is limited to multiplication-tier operations level.
• Intel's processors are more on the CISC side, so your highest-level instructions like Compute Reciprocal of Square Root ie x-1/2 = 1/sqrt(x) are seemingly arbitrarily chosen.
• ARM is for mobile phones and stuff where you are on a battery and that battery does not carry "a lot of" charge.
• Intel is for regular PCs which hook up to a power outlet most of the time, and so power consumption is not a really big deal.
• ARM is not locked to a particular vendor like Intel/AMD are the X86/x64 market.
• ARM platform is more of a recipe where a company can pick and choose the functionality they want.
• ARM processors are designed for a power requirement at first, performance second.
• x86 chips work quickly but are harder to make, expensive and not as power-conscious.
• RISC chips require fewer transistors to function. With fewer transistors to power, RISC chips see power savings and heat reduction when compared to their CISC counterparts.
• ARM chips are used in heaps of other devices: routers, set-top boxes and smart TVs, smartwatches, some gaming devices, automotive infotainment systems and so on.
• The x86 has some very powerful instructions, the arm can still beat it in a fight.
• x86 can operate on direct memory as well.
• AMD64 supports 3D-Now and 3D-Now Extensions, in both 32-bit (legacy) and 64-bit (long) mode.
• AMD64 is designed to concurrently enable 32 and 64-bit processing with no loss of performance.
• Athlon 64 and Opteron are examples of some of the AMD processors that use the AMD64 architecture.
• AMD64 is the AMD-led update to the venerable x86 instruction set architecture, or ISA.
• AMD64-compatible desktop processor market is huge these days.
• AMD64-able mobile processors for laptop systems, something Intel has yet to offer.
• The AMD Athlon 64 adds a memory channel depending on the socket, twice or four times the cache memory depending on the model, and comparable and further frequencies, to beef up the performance of that model line compared to Sempron 64.

ARM vs X86 vs AMD64 | Geekboots

Know the differences between ARM, X86 and AMD64 processor - Geekboots

www.geekboots.com

Hi Fmf,

Akida is processor agnostic, so it can work with RISCs (reduced instruction set computer) or CISCs (Complex instruction set computer).

https://brainchip.com/akida-neural-processor-soc/

RISC use a large number of short instructions (simple actions), while CISC instructions are longer(complex actions), so CISC use fewer instructions to perform an action which RISC would take more instructions to perform.

To enable Akida to talk to any processor, the communication protocol between Akida and the other processor is in a standard format, ie, they talk the same language. The format of the signals/data exchanged is set in accordance with the protocol, eg, PCIe or USB.

 

[jtardif999]

still not up yet !

Interesting that ComSec state BRN is their number 1 traded stock and then take the drum roll approach to releasing the video - engaging their audience to maximise viewing. Shows just how far BRN have come and how seriously they are now being taken. AIMO.

 

[Fullmoonfever]

Hi Fmf,

Akida is processor agnostic, so it can work with RISCs (reduced instruction set computer) or CISCs (Complex instruction set computer).

https://brainchip.com/akida-neural-processor-soc/

View attachment 7774

RISC use a large number of short instructions (simple actions), while CISC instructions are longer(complex actions), so CISC use fewer instructions to perform an action which RISC would take more instructions to perform.

To enable Akida to talk to any processor, the communication protocol between Akida and the other processor is in a standard format, ie, they talk the same language. The format of the signals/data exchanged is set in accordance with the protocol, eg, PCIe or USB.

Thanks D

So totally intergratable with Intel CISC...cool.

Presume Intel CISC more prevalent in server settings and as such there is the possibility of Akida to servers as well as Edge devices?

Like eX3 mob testing in the TaiShan and KunPeng?