BRN Discussion Ongoing

[7für7]

That's not the way I see it. Read these two sentences again. If we do have something to do with this it'll be bigger than Ben-Hur's britches.

"This platform will progress our understanding of the brain and develop brain-scale computing applications in diverse fields including sensing, biomedical, robotics, space, and large-scale AI applications."

"Using commercial off-the-shelf configurable hardware means that the protype would be easy to replicate at data centres around the world."

Yes, I was reading it but I don’t necessarily connect everything “bain inspired” technology with brainchip. Just beause we have “brain” in our company. I think we have a lot more big Industrie player as partners we work with. This makes me more confident. But yes… if this supercomputer have brainchip in it… it will be nice. Not more and not less… for me at least… no financial advice!

 

[7für7]

You're expecting the people to read?

🫠 Not everyone is reading everything in this thread. Not because it’s not interesting. But some are just not that into technical stuff. They just want to have general information how the stock and the company is going. The company is doing well… and they expend their ecosystem. But the share price is kind of doing weak.

 

[skutza]

What can we expect from Brainchip and the Akida product in 2024?

Nothing, like 2022,2023.

That's what I now expect. But I'll be over the moon if we get something. Expect nothing, then don't get disappointed........

 

[JDelekto]

It says the supercomputer will be capable of 228 trillion operations per second and know that AKIDA-P can go to 131 trillion operations per second.

I wonder if they've been using AKIDA 3.00 a spiced up version to achieve additional TOPS?

We know Western Sydney Uni partners with Intel but Intel's Loihi isn't commercially available so can be them. Trying to find out how many TOPS Loihi can do.


View attachment 51935

I find this very intriguing and did a little bit of research. Some dots can be joined between Western Sydney University and BrainChip. There was a 4th International Workshop on Event-Based Vision in June. The usual suspects were from respective companies, with presentations by André van Schaik representing Western Sydney University and Nandan Nayampally from BrainChip. Prophesee had a presence there as well.

I did run across a 3-year-old LinkedIn post with André congratulating Tony Lewis on an award that Western Sydney University presented, so the two are not strangers.

I uncovered a research paper published in March 2018, for which André van Schaik was a co-author. The paper describes an "FPGA-based Massively Parallel Neuromorphic Cortex Simulator," with the start of the abstract as follows:

Abstract—This paper presents a massively parallel and scalable neuromorphic cortex simulator designed for simulating large and structurally connected spiking neural networks, such as complex models of various areas of the cortex. The main novelty of this work is the abstraction of a neuromorphic architecture into clusters represented by minicolumns and hypercolumns, analogously to the fundamental structural units observed in neurobiology.

The paper describes using Intel's Altera Stratix V FPGA for the simulation. This is a tool used in various applications, including as a tool for FPGA digital logic design. A PRWire article about this new neuromorphic supercomputer points out the advantages of this approach, specifically under the "Reconfigurable" and "Commercial Availability" sections.

Another thing that we know is that BrainChip's Akida is FPGA-based, as noted by one of their press releases touting an article from "The Next Platform."

Is Western Sydney University using the Akida IP in its neuromorphic supercomputer? I don't know. However, the circumstantial evidence leads me to believe it is possible.

 

[buena suerte :-)]

I'm seeing the same pattern of shorting in many different stocks around 1 Dec. It may be due to borrowers having to give the stock back to the lenders under contractual obligations. The contracts aren't open ended.

Go into shortman and compare the shape of BRN's short chart around 1 Dec with WBT, SYR, TLG, AGY, NVX

Good point! Cheers Proga

 

[Bravo]

I find this very intriguing and did a little bit of research. Some dots can be joined between Western Sydney University and BrainChip. There was a 4th International Workshop on Event-Based Vision in June. The usual suspects were from respective companies, with presentations by André van Schaik representing Western Sydney University and Nandan Nayampally from BrainChip. Prophesee had a presence there as well.

I did run across a 3-year-old LinkedIn post with André congratulating Tony Lewis on an award that Western Sydney University presented, so the two are not strangers.

I uncovered a research paper published in March 2018, for which André van Schaik was a co-author. The paper describes an "FPGA-based Massively Parallel Neuromorphic Cortex Simulator," with the start of the abstract as follows:



The paper describes using Intel's Altera Stratix V FPGA for the simulation. This is a tool used in various applications, including as a tool for FPGA digital logic design. A PRWire article about this new neuromorphic supercomputer points out the advantages of this approach, specifically under the "Reconfigurable" and "Commercial Availability" sections.

Another thing that we know is that BrainChip's Akida is FPGA-based, as noted by one of their press releases touting an article from "The Next Platform."

Is Western Sydney University using the Akida IP in its neuromorphic supercomputer? I don't know. However, the circumstantial evidence leads me to believe it is possible.

OK, so that's Tony Lewis's connection to Andre.

I've just been listening to Andre Van Schaik's interview with Sunny Bains recorded 9 Jan 2023 in which they discuss his work on Deep South.

The discussion picks up pace at about 15 mins in and at about 26.48mins Andre talks about the FPGA's that are being used for the supercomputer that are commercially available. He doesn't mention which FPGA but he says at 29 mins approx. that they are "high-end FPGA's" so they're not something everyone will be able to afford to buy. Given the University's funding they were able to buy 100 FPGA's for the project.

It sounds like he's referring to Intel's Altera Stratix V FPGA's.

What would be really interesting is to find out if you could replicate Deep North using AKIDA instead of Intel's Altera Stratix and to compare the costs.

https://www.eetimes.com/podcasts/andre-van-schaik-discusses-new-neuromorphic-simulator/

 

[MDhere]

What can we expect from Brainchip and the Akida product in 2024?

Expect the unexpected

 

[Esq.111]

Afternoon Chippers ,

Having a cleanout of paperwork in the office & came upon this random note from 14 Feb 2021.

Dare say one of our super slouthes would be all over this .

Brainchip Akida Early Access Evaluation Kit

On 12/8/20 Shared Services Center issued Special Notice 80NSSC21745612Q for Brainchip Akida Early Access Evaluation Kit due 12/15/20

www.highergov.com

Regards,
Esq.

 

[Diogenese]

I find this very intriguing and did a little bit of research. Some dots can be joined between Western Sydney University and BrainChip. There was a 4th International Workshop on Event-Based Vision in June. The usual suspects were from respective companies, with presentations by André van Schaik representing Western Sydney University and Nandan Nayampally from BrainChip. Prophesee had a presence there as well.

I did run across a 3-year-old LinkedIn post with André congratulating Tony Lewis on an award that Western Sydney University presented, so the two are not strangers.

I uncovered a research paper published in March 2018, for which André van Schaik was a co-author. The paper describes an "FPGA-based Massively Parallel Neuromorphic Cortex Simulator," with the start of the abstract as follows:



The paper describes using Intel's Altera Stratix V FPGA for the simulation. This is a tool used in various applications, including as a tool for FPGA digital logic design. A PRWire article about this new neuromorphic supercomputer points out the advantages of this approach, specifically under the "Reconfigurable" and "Commercial Availability" sections.

Another thing that we know is that BrainChip's Akida is FPGA-based, as noted by one of their press releases touting an article from "The Next Platform."

Is Western Sydney University using the Akida IP in its neuromorphic supercomputer? I don't know. However, the circumstantial evidence leads me to believe it is possible.

Hi JD,

It is true that the configuration of Akida is field programmable as it has a programmable communication matrix interconnecting the nodes, but in my mind Akida is not FPGA-based. The initial proof-of-concept circuit was built in FPGA (Xylink, or was that BrainChip Accelerator?), but the commercial Akida 1 chip would be better described as an ASIC.

FPGA is field programmable gate array, a prefabricated chip with lots of different logic gates which the user can selectively interconnect to make a number of different circuits with for different purposes*. As a result there are many redundant logic gates and the layout is far from optimal. In the case of Akida, it would have been for fewer nodes than Akida 1, and it's performance would be inferior to an ASIC version of Akida. FPGAs are commonly used as test chips.

ASIC (application specific integrated circuit) is a purpose-built chip with only the necessary gates and in which the layout would be optimized by the circuit designer.

From the article you cited:

When we last spoke with BrainChip in 2018, the company was on the verge of rolling out its FPGA-based spiking neural network (SNN) accelerator, known as Akida (Greek for spike). At that point, the plan was to get its hardened SoC into the market by 2019.

The reference to "FPGA-based" was to the proof-of-concept chip. The "hardened SoC" is the ASIC. The use of the adjective "hardened" can be thought of as implying the FPGA design is malleable.

The presence of Zurich Uni as a partner would be put on the scales on the side of analog MemRistor SNNs, but this is not conclusive.

* On reflection, I had to put in the bit about different purposes to better distinguish Akida from my definition of a FPGA. Akida is a single purpose SNN with field programmable nodes and NPUs.

 

[gilti]

Question
The after close auction timing varies from a few seconds after 4.10 to 40-50 seconds.
What sets the cut off time. It appears that it fluctuates until the price has been dropped then stops.
Surely there must be a fixed limit.

 

[Diogenese]

Question
The after close auction timing varies from a few seconds after 4.10 to 40-50 seconds.
What sets the cut off time. It appears that it fluctuates until the price has been dropped then stops.
Surely there must be a fixed limit.

Maybe the cut-off is for orders in the queue by 4:10 - anything after that misses the bus?

After all, us bots do have homes to go to.

 

[FJ-215]

If anyone is wondering where the shorts have gone it might be worth having a look in the Lithium space.

PLS currently have 22% of the float shorted!!!

 

[Iseki]

Hi JD,

It is true that the configuration of Akida is field programmable as it has a programmable communication matrix interconnecting the nodes, but in my mind Akida is not FPGA-based. The initial proof-of-concept circuit was built in FPGA (Xylink, or was that BrainChip Accelerator?), but the commercial Akida 1 chip would be better described as an ASIC.

FPGA is field programmable gate array, a prefabricated chip with lots of different logic gates which the user can selectively interconnect to make a number of different circuits with for different purposes*. As a result there are many redundant logic gates and the layout is far from optimal. In the case of Akida, it would have been for fewer nodes than Akida 1, and it's performance would be inferior to an ASIC version of Akida. FPGAs are commonly used as test chips.

ASIC (application specific integrated circuit) is a purpose-built chip with only the necessary gates and in which the layout would be optimized by the circuit designer.

From the article you cited:

When we last spoke with BrainChip in 2018, the company was on the verge of rolling out its FPGA-based spiking neural network (SNN) accelerator, known as Akida (Greek for spike). At that point, the plan was to get its hardened SoC into the market by 2019.

The reference to "FPGA-based" was to the proof-of-concept chip. The "hardened SoC" is the ASIC. The use of the adjective "hardened" can be thought of as implying the FPGA design is malleable.

The presence of Zurich Uni as a partner would be put on the scales on the side of analog MemRistor SNNs, but this is not conclusive.

* On reflection, I had to put in the bit about different purposes to better distinguish Akida from my definition of a FPGA. Akida is a single purpose SNN with field programmable nodes and NPUs.

Thanks as always for your brilliance.

Regards,

Depressed of Batemans Bay.

PS Do you have a take on why the satellite fix-it co chose akida 1000? I can see that it could have been taught a lot of different moves with the robotic arm - rules like " If you see this then do this..." Have you any intel on what the robotic arm can in fact do?

 

[Iseki]

Afternoon Chippers ,

Having a cleanout of paperwork in the office & came upon this random note from 14 Feb 2021.

Dare say one of our super slouthes would be all over this .

Brainchip Akida Early Access Evaluation Kit

On 12/8/20 Shared Services Center issued Special Notice 80NSSC21745612Q for Brainchip Akida Early Access Evaluation Kit due 12/15/20

www.highergov.com

Regards,
Esq.

I believe Washington Foundries did the man-made-stone "granite" tops to the reception counter at our new premises at Laguna Hills

 

[Tothemoon24]

VVDN - Looking very polished

Camera Design and Manufacturing Services - VVDN

With an experience of more than 15+ years, VVDN helps global OEMs with E2E design, development, and manufacturing of smart cameras across multiple domains

www.vvdntech.com

 

[Tothemoon24]

The robot operates in autonomous moving or manual remote control mode, with a battery life of at least 4 hours, and it offers real-time processing with side-by-side inspection and reporting.

RIIS005 Railway Inspection Robot - Innovative solution for the railway industry | Bhavishek Malik

RIIS005 Railway Inspection Robot - Innovative solution for the railway industry RIIS005 Railway Inspection Robot is equipped with a self-moving motion platform and various high-precision intelligent sensor modules to achieve rapid and accurate detection of track defects. The robot is designed...

www.linkedin.com

 

[HopalongPetrovski]

Thanks as always for your brilliance.

Regards,

Depressed of Batemans Bay.

PS Do you have a take on why the satellite fix-it co chose akida 1000? I can see that it could have been taught a lot of different moves with the robotic arm - rules like " If you see this then do this..." Have you any intel on what the robotic arm can in fact do?

 

[Pom down under]

Anyone else getting that special feeling in their potatoes that January is gonna be...........

View attachment 51914

 

[wilzy123]

Question
The after close auction timing varies from a few seconds after 4.10 to 40-50 seconds.
What sets the cut off time. It appears that it fluctuates until the price has been dropped then stops.
Surely there must be a fixed limit.

On a regular trading day, CSPA ends somewhere between 4:10-4:12PM AEST.

Specifically, it's 4:10pm + a time programmatically chosen at 'random' (in seconds - i think the range is between -60 and 60) + 60 seconds.

So, lets pretend today's random number was 17 seconds. CSPA would settle at 4:11:17 (i.e. 4:10 + 17 seconds + 60 seconds).

During this time, any outstanding orders to buy or sell securities are matched at a single price to determine the final trades for the day. This single price is based on the weighted average of the last trade prices for the securities during this two-minute auction period. The aim is to "ensure an orderly and fair closing of the market" - lol.

 

[gilti]

On a regular trading day, CSPA ends somewhere between 4:10-4:12PM AEST.

Specifically, it's 4:10pm + a time programmatically chosen at 'random' (in seconds - i think the range is between -60 and 60) + 60 seconds.

So, lets pretend today's random number was 17 seconds. CSPA would settle at 4:11:17 (i.e. 4:10 + 17 seconds + 60 seconds).

During this time, any outstanding orders to buy or sell securities are matched at a single price to determine the final trades for the day. This single price is based on the weighted average of the last trade prices for the securities during this two-minute auction period. The aim is to "ensure an orderly and fair closing of the market" - lol.

Thanks for the information. i had not been able to find it anywhere. So it is just a fantastic coincidence that 3 days last week the final trade for the day was a 1 share transaction which just also strangely tipped the balance down by half a cent?