BRN Discussion Ongoing

[Fact Finder]

'Royalties are paid on the product containing Brainchip IP being produced and sold in the retail market.'

Are royalties generally paid at point of production or point of sale?

If you take Prophesee as an example it is unlikely that their royalties are being paid at point of production by Sony otherwise it would have not needed to raise $50 million so close to announcing that they were about to go to mass production.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Damo4]

'Royalties are paid on the product containing Brainchip IP being produced and sold in the retail market.'

Are royalties generally paid at point of production or point of sale?

Really good question that I googled out of interest also.

https://www.design-reuse.com/news_img/Handbook_Understanding_SemiIP_BusinessProcess.pdf

Seems it could be done both ways tbh.
Might be based on the client or end user.

"In the royalty-based model, there is an ongoing payment of consideration tied to some objectively
quantifiable measure, such as the number of units sold or the number of wafers manufactured. The royalty based
model is most common in the licensing of highly differentiated SIP products and may include other
parties to the agreement such as the foundry who manufactures the device instantiating the SIP. In some
cases, royalties are paid by foundries, such as in the case of foundation SIP products; in other cases,
royalties are paid by the end user such as in the case of processor SIP products and specialty memory SIP
products."

 

[VictorG]

'Royalties are paid on the product containing Brainchip IP being produced and sold in the retail market.'

Are royalties generally paid at point of production or point of sale?

My experience with royalties is that they are paid at production.
If the product with the IP is mismanaged, over priced, not popular or for whatever reason fails to achieve sales, that should not be a problem for BRN.

 

[Xray1]

How can we say they have not already been achieved?

Income was not given as a performance indicator by which to judge his and the Boards performance.

My opinion only DYOR
FF

AKIDA BALLISTA

FF..
"Income was not given as a performance indicator by which to judge his and the Boards
performance."

IMO ........... I would submit that it was "Implied " !!!!!

 

[Baron Von Ricta]

Looks like another day of BS trading is on the cards View attachment 24383

And we'll see if this one goes through or gets pulled before open

 

[Fact Finder]

‘Warning the following comment is a response to the recent tragedy.’

The early press releases are suggesting these loonies were ‘Survivalists’ well if true they have to be marked as abject failures.

Definition:

Noun

1. a person who practises outdoor survival skills as a sport or hobby.
   "a survivalist's handbook"
   
   
2. 
   a person who tries to ensure their own survival or that of their social or national group.
   "a paranoid survivalist bracing for Armageddon"

Giving these evil lunatics such credibility by use of the name ‘Survivalists’ is irresponsible journalism.

Their suicide event proves beyond doubt that they were not interested in anything other than their own selfish egocentric narcissistic deluded view of their superiority.

Like the evil lunatic that stalked Port Arthur there is nothing to commend understanding such individuals they have nothing to offer other than a warning that such individuals or groups exist.

Regards
FF

 

[Evermont]

 

[wilzy123]

ANNNNNNN!!!!!!!!!!

Last time a "Notification regarding unquoted securities" ANN was released (8th Dec)... the market took it well.

 

[Damo4]

FF ,
Codes and associated employees known to date :
BRNAM - Anil , Ken , Sean , Antonio
BRNAJ - Peter , Pia
BRNAD - admin

just forwarding this again in relation to the latest ann
thanks @Doz


Codes and associated employees known to date :

BRNAM - Anil , Ken , Sean , Antonio

BRNAJ - Peter , Pia

BRNAD - admin

 

[Izzzzzzzzzzy]

just forwarding this again in relation to the latest ann
thanks @Doz


Codes and associated employees known to date :

BRNAM - Anil , Ken , Sean , Antonio

BRNAJ - Peter , Pia

BRNAD - admin

Mother fuckers probably got a partnership with AMD and nvidia on the same day hence the RSUs

 

[Dozzaman1977]

just forwarding this again in relation to the latest ann
thanks @Doz


Codes and associated employees known to date :

BRNAM - Anil , Ken , Sean , Antonio

BRNAJ - Peter , Pia

BRNAD - admin

So this new announcement states that none of these shares have been issued to key management personnel (kmp) but is coded BRNAM

I'd assume Anil, Ken, Sean and Antonio are key management so who did the shares go too....... Maybe xmas bonus for all the employees for the results they have been achieving???

 

[Learning]

So this new announcement states that none of these shares have been issued to key management personnel (kmp) but is coded BRNAM

I'd assume Anil, Ken, Sean and Antonio are key management so who did the shares go too....... Maybe xmas bonus for all the employees for the results they have been achieving???

I too read this announcement as a positive.

BRNAM seem like AM = America staff. As Anil, Ken, Sean and Antonio is base in America.

So maybe the sale teams are getting the bonus (JMO)

Learning

 

[equanimous]

#shareman

 

[Diogenese]

@Diogenese

Maybe the answer.

In 2020 started using VeriSilicon NPU for their latest V37 at the time.

Would presume has just been iterations of same since.

Haven't looked into the VeriSilicon capabilities yet. Time for

VeriSilicon VIP9000 and ZSP are Adopted by iCatch Next Generation AI-powered Automotive Image Processing SoC

VeriSilicon VIP9000 NPU AI processor and ZSPNano DSP IP bring AI-Vision and AI-Voice to low power automotive image processing SoC Shanghai, China, May 12,

www.edge-ai-vision.com

VeriSilicon VIP9000 and ZSP are Adopted by iCatch Next Generation AI-powered Automotive Image Processing SoC

Shanghai, China, May 12, 2020 – VeriSilicon today announced that iCatch Technology, Inc. (TPEX: 6695), a global leader in low-power and intelligent image processing SoC solutions, has selected VeriSilicon VIP9000 NPU and ZSPNano DSP IP. Both will be utilized in the iCatch’s next generation AI-powered image processing SoC with embedded neural network accelerators powered by VeriSilicon’s NPU for applications such as automotive electronics, industrial, appliance, consumer electronics, AIoT, smart home, commercial and more.

VIP9000 is a highly scalable and programmable processor IP for computer vision and artificial intelligence applications and with support for all popular deep learning frameworks (TensorFlow, Pytorch, TensorFlow Lite, Caffe, Caffe2, DarkNet, ONNX, NNEF, Keras…), as well as OpenCLTM and OpenVXTM APIs. Neural network optimization techniques such as quantization, pruning, and model compression are also supported natively by VIP9000 architecture. AI applications can be easily ported to VIP9000 platforms through offline conversion by the Vivante AcuityTM SDK, or through run-time interpretation with Android NN, NN API, or ARM NN. The ZSPNano DSP provides complete programmability, industry leading power efficiency with a robust and mature SDK, enabling iCatch to extend audio and voice software applications smoothly including Acoustic Echo Cancelation (AEC), Noise Suppression and Beam Forming.

“By incorporating VeriSilicon key IP capabilities, our V37, the latest generation AI-powered image processing SoC for 4K video, can empower the edge computing capability right on the camera. Together with the embedded iCatch’s high quality image signal processing (ISP) engine and proprietary CV accelerator engine, we can drive the VeriSilicon NPU engine to the best TOPS/W performance. As a result, our AI-powered image processing SoCs not only can record high-quality video, but also perform advanced image analytics to bring intelligence to the camera devices. We successfully applied our AI-powered image processing SoCs into various applications.” said Weber Hsu, President of iCatch Technology, Inc.

“Edge AI computing has been applied in a wide range of market segments including consumer, industrial and automotive devices, especially in computer vision related applications. Image processors with built-in AI NPU have great growth momentum,” said Weijin Dai, Executive Vice President and GM of Intellectual Property Division at VeriSilicon.

“Smart eye and smart ear are essential for all the intelligent edge devices. The challenge is how to deliver the intelligence efficiently within the target product power envelope and silicon budget,” added Weijin Dai. “As a leading company in image processing, iCatch will greatly leverage our NPU and DSP in advanced solutions to perfectly address the enormous market needs. With VIP9000 and ZSPNano, iCatch’s image processing SoCs deliver advanced AI-Vision and AI-Voice capabilities at the lowest power consumptions.”

Hi Fmf,

Just brilliant research. You've ferreted out the NN, and it's CNN.

Verisilicon:

US2021382690A1 Enhanced Multiply Accumulate Device For Neural Networks


US11301214B2 Device for performing multiply/accumulate operations


CN107862378B Convolutional neural network acceleration method and system based on multiple kernels, storage medium and terminal

This explains why Prophesee was so impressed with Akida after having already committed to iCatch.

 

[DollazAndSense]

@Diogenese

Maybe the answer.

In 2020 started using VeriSilicon NPU for their latest V37 at the time.

Would presume has just been iterations of same since.

Haven't looked into the VeriSilicon capabilities yet. Time for

VeriSilicon VIP9000 and ZSP are Adopted by iCatch Next Generation AI-powered Automotive Image Processing SoC

VeriSilicon VIP9000 NPU AI processor and ZSPNano DSP IP bring AI-Vision and AI-Voice to low power automotive image processing SoC Shanghai, China, May 12,

www.edge-ai-vision.com

VeriSilicon VIP9000 and ZSP are Adopted by iCatch Next Generation AI-powered Automotive Image Processing SoC

Shanghai, China, May 12, 2020 – VeriSilicon today announced that iCatch Technology, Inc. (TPEX: 6695), a global leader in low-power and intelligent image processing SoC solutions, has selected VeriSilicon VIP9000 NPU and ZSPNano DSP IP. Both will be utilized in the iCatch’s next generation AI-powered image processing SoC with embedded neural network accelerators powered by VeriSilicon’s NPU for applications such as automotive electronics, industrial, appliance, consumer electronics, AIoT, smart home, commercial and more.

VIP9000 is a highly scalable and programmable processor IP for computer vision and artificial intelligence applications and with support for all popular deep learning frameworks (TensorFlow, Pytorch, TensorFlow Lite, Caffe, Caffe2, DarkNet, ONNX, NNEF, Keras…), as well as OpenCLTM and OpenVXTM APIs. Neural network optimization techniques such as quantization, pruning, and model compression are also supported natively by VIP9000 architecture. AI applications can be easily ported to VIP9000 platforms through offline conversion by the Vivante AcuityTM SDK, or through run-time interpretation with Android NN, NN API, or ARM NN. The ZSPNano DSP provides complete programmability, industry leading power efficiency with a robust and mature SDK, enabling iCatch to extend audio and voice software applications smoothly including Acoustic Echo Cancelation (AEC), Noise Suppression and Beam Forming.

“By incorporating VeriSilicon key IP capabilities, our V37, the latest generation AI-powered image processing SoC for 4K video, can empower the edge computing capability right on the camera. Together with the embedded iCatch’s high quality image signal processing (ISP) engine and proprietary CV accelerator engine, we can drive the VeriSilicon NPU engine to the best TOPS/W performance. As a result, our AI-powered image processing SoCs not only can record high-quality video, but also perform advanced image analytics to bring intelligence to the camera devices. We successfully applied our AI-powered image processing SoCs into various applications.” said Weber Hsu, President of iCatch Technology, Inc.

“Edge AI computing has been applied in a wide range of market segments including consumer, industrial and automotive devices, especially in computer vision related applications. Image processors with built-in AI NPU have great growth momentum,” said Weijin Dai, Executive Vice President and GM of Intellectual Property Division at VeriSilicon.

“Smart eye and smart ear are essential for all the intelligent edge devices. The challenge is how to deliver the intelligence efficiently within the target product power envelope and silicon budget,” added Weijin Dai. “As a leading company in image processing, iCatch will greatly leverage our NPU and DSP in advanced solutions to perfectly address the enormous market needs. With VIP9000 and ZSPNano, iCatch’s image processing SoCs deliver advanced AI-Vision and AI-Voice capabilities at the lowest power consumptions.”

@Fullmoonfever
Mate, your research is always second to none but it seems you’ve been in overdrive the past couple of weeks! Just want to say much appreciated - always look forward to your posts.

 

[Fullmoonfever]

Hi Fmf,

Just brilliant research. You've ferreted out the NN, and it's CNN.

Verisilicon:

US2021382690A1 Enhanced Multiply Accumulate Device For Neural Networks


US11301214B2 Device for performing multiply/accumulate operations


CN107862378B Convolutional neural network acceleration method and system based on multiple kernels, storage medium and terminal

This explains why Prophesee was so impressed with Akida after having already committed to iCatch.

Cheers mate.

It would definitely seem to make more sense to use neuromorphic with neuromorphic instead of the CNN.

Maybe why the guy from Prophesee (not Luca) in the video not long ago seemed a bit non committal around their best processing option...playing safe to not totally offend iCatch whilst not confirming us yet till say an iCatch contract superceded or dissolved?

 

[Fullmoonfever]

@Fullmoonfever
Mate, your research is always second to none but it seems you’ve been in overdrive the past couple of weeks! Just want to say much appreciated - always look forward to your posts.

Welcome DAS.

Appreciated.

 

[Diogenese]

A bit more information:

Summary
For over a decade, CPU and GPU design companies have been using Synopsys VC Formal Datapath Validation (DPV) app with its HECTOR™ technology to verify their data processing elements because traditional verification methods cannot exhaustively verify the correctness of mathematical computations in these designs. Like CPUs and GPUs, AI processors are also datapath heavy with mathematical functions like addition, subtraction, matrix multiplication, and square root in its compute engines, making these designs a good fit for formal datapath validation.
This webinar will introduce the Synopsys ARC® NPX Neural Processing Unit (NPU) IP family of embedded AI processors and use of Synopsys VC Formal DPV to verify its datapath functions. The Synopsys ARC NPX6 processor supports the latest and most complex neural networks, such as CNN, RNN, and transformers, targeted for AI SoCs that are widely used in automotive, data center, high end gaming, next generation augmented reality, and surveillance. At the heart of the NPX6 neural network processor are convolution and tensor accelerators that are optimized to perform light speed computations. Correctness of these functional units are key to correct facial, audio, image processing and recognition which could have safety implications for automotive applications. We will explore Synopsys VC Formal DPV, the gold standard for datapath validation and signoff for the last 20 years, and how it is used to ensure correctness of the core algorithms of the ARC NPX6 processor’s compute engines.

Speakers

Neelabja Dutta
Sr. Manager, Applications Engineering
Synopsys

Shuaiyu Jiang
Sr. ASIC Digital Design Engineer
Synopsys

Long term holders of Brainchip shares will know that Vorago in the Phase 1 NASA application to harden AKD1000 described it as a CNNRNN processor.

My interest only so DYOR
FF

AKIDA BALLISTA

Synopsys have gone through the CNN wormhole in the space-time fabric and are trapped in MAC land.

US10846591B2 Configurable and programmable multi-core architecture with a specialized instruction set for embedded application based on neural networks

A programmable architecture specialized for convolutional neural networks (CNNs) processing such that different applications of CNNs may be supported by the presently disclosed method and apparatus by reprogramming the processing elements therein.


Their party trick is swapping between 16 bit and 8 bit MACs

US2021377122A1 MIXED-PRECISION NEURAL NETWORKS

[006] … The operations include receiving a target bandwidth increase for a neural network including a plurality of binary large objects (BLOBs) and weights of a first data type represented by a first number of bits. The target bandwidth increase relates to changing at least some of the plurality of BLOBs and weights to a second data type represented by a second number of bits different from the first number of bits.

[0026] Further, any suitable supervised ML techniques can be used to train the floating-point ML model. For example, the initial ML model 114 can be a neural network (e.g., a convolutional neural network (CNN), an adversarial neural network, or any other suitable neural network) and can be trained using any suitable technique. FIG. 1 is merely one example.

 

[Diogenese]

MF

Difficult to argue with this paragraph

"But given this alliance, other recent agreements, management’s bullish rhetoric, and its lofty market capitalisation, the market will no doubt be expecting Brainchip to start delivering some very big sales in 2023."

And it provides BRN.AX some exposure to generate investor interest.
OK admittedly trying to put some lipstick on the pig.

. ... and a very handsome pig it is too ... but why did you put the lipstick on that end?

 

[jtardif999]

Thanks D

Like you, I had a quick patent skim and couldn't find anything to do with AI per se either.

I did notice the CNN reference on the Vi37 but then just NPU on others as I said hence the question on CNN2SNN if viable.

When you say missing link I presume you mean the AI processing component (not us in that IMX) to create that package?

I'll try dig around for earliest Prophesee/ iCatch connection mentions maybe tomoz.

Begs question maybe then of how long working with iCatch and how long working with us before either public news releases and was there any parallel Testing / Dev?

Yeah as our management have perviously described that it’s all about intersecting at the right point in their design phase to create the quickest turnaround. I would assume we will finish up as IP embedded directly in a later/next? vision sensor and we know from the Renesas experience how long it might take before we see the related product. But a licence fee should be up coming if we have intersected at the right time.