BRN Discussion Ongoing

Diogenese

Top 20
Too early to succeed? This happened to other companies before.
AND ... a thought I have been having for a longer while ... maybe AKIDA is too cheap? I mean there is a huge price difference between our chip and the others and maybe this also gives the impression that it can't be good enough just to try. :unsure: I am wondering why we are so cheap, is there really a need for it?
Hi Chips,

Do we know the pricing for Akida 2.

It's performance will be significantly better than Akida 1, so I hope the price will match the improvement.

That way, there will still be a market for Akida 1.
 
  • Like
  • Love
  • Fire
Reactions: 20 users

Easytiger

Regular
  • Haha
Reactions: 2 users

CHIPS

Regular
Hi Chips,

Do we know the pricing for Akida 2.

It's performance will be significantly better than Akida 1, so I hope the price will match the improvement.

That way, there will still be a market for Akida 1.

Hi Diogenese

No, not for Akida 2. I am sure it will be better than all the others before, but I also hope that the price will be much higher. The price usually gives the impression of higher quality therefore I am not pro low price. But we will see ...
 
  • Haha
  • Fire
Reactions: 3 users

Diogenese

Top 20


Helium is lightweight AI.

It requires CPU participation to implement software instructions.

documentation-service.arm.com/static/620259d9965f7d118e3f5f7f

Arm Helium technology is the M-Profile Vector Extension (MVE) for the Arm Cortex-M processor series.

Helium is an extension of the Armv8.1-M architecture and delivers a significant performance increase for Machine Learning (ML) and Digital Signal Processing (DSP) applications.

Helium lets you optimize performance by using Single Instruction Multiple Data (SIMD) to perform the same operation simultaneously on multiple data items. The Helium instructions are designed for Digital Signal Processing (DSP) and Machine Learning (ML) application.

Most Arm instructions are Single Instruction Single Data (SISD). Each instruction performs its specified operation on a single data item. Processing multiple data items therefore requires multiple instructions.

This method is slow, and it can be difficult to see how different registers are related. If the values that you are dealing with, are smaller than the maximum register size, the extra potential bandwidth is wasted with SISD instructions. For example, when adding 8-bit values together, each 8- bit value must be loaded into a separate 32-bit register. Performing many individual operations on small data items does not use machine resources efficiently.

Single Instruction Multiple Data (SIMD) instructions perform the same operation for multiple data items. These data items are packed as separate lanes in a larger register. For example, the following instruction adds four pairs of 32-bit values together. However, in this case, the values are packed into separate lanes in two pairs of 128-bit registers: the q registers. Each lane in the first source register is then added to the corresponding lane in the second source register, before being stored in the same lane in the destination register.

Being able to specify parallel operations in a single instruction like this allows the processor to do the calculations simultaneously, which increases performance and throughput. The preceding diagram shows 128-bit registers, with each register holding four 32-bit values. Operations on data elements with different data sizes are possible for Helium registers.

The Helium registers contain vectors of elements of the same data type. The same element position in the input and output registers is referred to as a lane.

Usually each Helium instruction results in n operations, where n is the number of lanes that the input vectors are divided into. Each operation is contained within the lane.

The number of lanes in a Helium vector depends on the size of the vector and the data elements in the vector. A 128-bit Helium vector can contain the following element sizes:
• Two 64-bit integers
• Four 32-bit integers or single precision float
• Eight 16-bit integers or half precision float
• Sixteen 8-bit integers Elements in a vector are ordered from the least significant bit to the most significant bit.

That is, element 0 uses the least significant bits of the register. Let’s look at an example of a Helium instruction. The instruction VADD.16 q0, q0, q5 performs a parallel addition of eight lanes of 16-bit (8 x 16 = 128) integer elements from vectors in q5 and q0, storing the result in q0.


ARM Helium is not a spiking NN.
 
  • Like
  • Love
  • Fire
Reactions: 16 users

Diogenese

Top 20
Hi Diogenese

No, not for Akida 2. I am sure it will be better than all the others before, but I also hope that the price will be much higher. The price usually gives the impression of higher quality therefore I am not pro low price. But we will see ...
Yes.

Akida 1 was priced according to performance, the more nodes a customer licensed, the higher the royalty. So now we have extended the upper end performance with Akida 2, the royalties should also increase with performance.

Akida 2 is denominated in 3 flavours Akida 2E, Akida 2S, and Akida 2P all with TeNN.

Each flavour is sub-divided by the number of nodes:
E has 1 to 4 nodes;
S has 2 to 8 nodes;
P has 8 to 128 nodes with optional ViT.

This is in addition to Akida 1 (available with up to 20 nodes) and Akida 1.5 probably with the same node options as Akida 1, but without the processor.

https://brainchip.com/akida2-0/

The 2nd generation IP platform will support a very wide range of market verticals and will be delivered in three classes of product.​


Akida-E: Extremely energy-efficient, for always on operation very close to, or at sensors.
Akida-S: Integration into MCU or other general purpose platforms that are used in broad variety of sensor-related applications
Akida-P: Mid-range to higher end configurations with optional vision transformers for ground breaking and yet efficient performance
.

So there will be a range of royalty rates depending on the selected model and "horsepower".
 
Last edited:
  • Like
  • Love
  • Fire
Reactions: 44 users

CHIPS

Regular

Nvidia vs. AMD vs. Intel: Comparing AI Chip Sales​


1693124105010.png

1693124177298.png

1693124380938.png


1693124302469.png


1693123995816.png
 
  • Like
  • Fire
Reactions: 7 users

Getupthere

Regular

Nvidia vs. AMD vs. Intel: Comparing AI Chip Sales​


View attachment 43107
View attachment 43108
View attachment 43110

View attachment 43109

View attachment 43106
Looks like the AI chip market is growing rapidly and Nvidia is getting more market share.

Other big boys needs some help.
 
  • Like
Reactions: 6 users

Bravo

If ARM was an arm, BRN would be its biceps💪!
Hi Bravo. As much as I appreciate the enthusiasm and sharing of potential significant information, if the article does not mention Brainchip, Akida, SNN, Neuromorphic, edge computing, or any other key words that would normally qualify for Akida suspicious activity, I would avoid referring to a 99.9% certainty of our engagement. All it does is create false hope and expectation which leads to potential disappointment and emotional rash investor mindset if it fall trough.

I encourage your involvement and find the article interesting but as far as I can see, all we have to go by in the article is Robs Like which is amusing at best.

Please correct me if I have missed some crucial contextual evidence with our engagement to this particular space mission

All the best, Funk

What about this Funky?


Screen Shot 2023-08-27 at 7.03.45 pm.png


S02.png


So1.png
 
Last edited:
  • Like
  • Love
  • Fire
Reactions: 47 users

IloveLamp

Top 20
  • Like
  • Fire
  • Love
Reactions: 4 users
New job just posted....... :) (As always, I love the free lunch)



1693130665396.png


About the job​



This job is sourced from a job board.

The ML/AI Research Scientist’s primary role is to is to support the research team in developing commercial applications for BrainChip’s Akida Neuromorphic System-on-Chip (NSoC). Some of the target applications include work in computer vision (object classification/detection and face recognition), audio processing (keyword spotting), and sensor fusion. Additionally, this team member will support the research team’s algorithm development for the next version of the Akida NSoC. BrainChip’s Akida fully-digital neuromorphic AI processor has the first of its kind, licensable product that delivers great efficiency in Edge AI devices along with unique capabilities like learning on the edge. Akida’s success is dependent on finding the best product market fit in large markets, supported by a product roadmap that has the right combination of features and capabilities that delivers the best differentiated value, with many internal functions and external entities. This is an ON-SITE ROLE ESSENTIAL JOB DUTIES AND RESPONSIBILITIES: Quickly prototyping, training, and testing machine learning (ML) solutions using online code repositories, research publications, or customer specifications Keeping abreast of developments in ML such as new development tools, libraries, and frameworks as well as new ML models/architectures, training techniques, and application pipelines Participating in ML algorithm/hardware co-design tasks Performing, documenting, and presenting detailed analyses related to ML algorithm development, software/hardware benchmarking, and application development Obtaining a thorough understanding of the Akida 1.0 hardware device and associated software stack (MetaTF) Interfacing with customers to discuss ML application goals, constraints, and opportunities JOB REQUIREMENTS: Education/Experience: Master’s Degree in Computer Science or equivalent. PhD preferred. Course work in machine learning and computer vision Strong programming skills in Python Three years’ experience developing ML applications in either TensorFlow/Keras and/or PyTorch Excellent communication skills Experience in one or more of the following application fields: Image Processing/Computer Vision, ADAS, Anomaly Detection, Audio/Speech Processing, or Sensor Fusion. Preferred Qualifications: Ten plus years’ experience developing ML applications in TensorFlow/Keras and PyTorch Multi-project experience in object classification, object detection, face recognition, and/or keyword spotting Knowledge of deep learning quantization techniques Experience with Docker and Git Experience with Scrum/Agile software development (e.g. Jira) Benefits Offered:

  • Competitive Pay depending on experience $130K-190K base
  • Restricted Stock Units
  • Bonus Pay
  • 401K with matching
  • Free Lunch Daily
  • Company-paid Medical HMO, Dental PPO, and Vision Insurance
  • Company-paid Life Insurance and AD&D
  • Paid Time Off
  • Holiday Pay
  • Employee Assistance Program, Caregiver Support, Adoption Assistance Program
  • Flexible Spending Account
  • Health Savings Account
  • Commuter Benefit Program
  • Employee Discounts BrainChip is an equal opportunity employer. All aspects of employment including the decision to hire, promote, discipline, or discharge, will be based on merit, competence, performance, and business needs. We do not discriminate on the basis of race, color, religion, marital status, age, national origin, ancestry, physical or mental disability, medical condition, pregnancy, genetic information, gender, sexual orientation, gender identity or expression, veteran status, or any other status protected under federal, state, or local law.
 
  • Like
  • Fire
  • Love
Reactions: 18 users

Ian

Founding Member
Too early to succeed? This happened to other companies before.
AND ... a thought I have been having for a longer while ... maybe AKIDA is too cheap? I mean there is a huge price difference between our chip and the others and maybe this also gives the impression that it can't be good enough just to try. :unsure: I am wondering why we are so cheap, is there really a need for it?
What chip? .....we sell IP duffus 😮‍💨
 
  • Like
  • Fire
Reactions: 7 users

Getupthere

Regular
 
  • Fire
  • Like
Reactions: 3 users

IloveLamp

Top 20
  • Like
  • Fire
  • Thinking
Reactions: 11 users

TheFunkMachine

seeds have the potential to become trees.
Yes Bravo! This information is exciting, factual and public knowledge! We can all rightfully be excited about all of this!:) but I’m not sure if it proves with 99.9 certainty that ISRO is using Akida. Yes, it is a good connection that NASA has been involved in this mission. Yes, it is exciting the Brainchip can be integrated with any space vehicle in the future! So technically if falls within the realm of Possibility. I was simply stating that the article you provided did not in my opinion provide a conclusive narrative with the information available.

I am certainly open to the possibility of Brainchip being on the moon. We know they are going into space soon for the robotic arm project. But unfortunately it is not public knowledge and therefor I simply wanted to encourage you to channel your excitement towards known facts and continue speculation with a more conservative % estimate:)

Again, thanks for the information and hopefully you are right! 👍🏻
 
  • Like
  • Love
  • Fire
Reactions: 19 users

Cgc516

Regular
Has anyone got bath towel ready for tomorrow?
 
  • Sad
  • Haha
Reactions: 2 users

wilzy123

Founding Member
New job just posted....... :) (As always, I love the free lunch)



View attachment 43132

About the job​



This job is sourced from a job board.

The ML/AI Research Scientist’s primary role is to is to support the research team in developing commercial applications for BrainChip’s Akida Neuromorphic System-on-Chip (NSoC). Some of the target applications include work in computer vision (object classification/detection and face recognition), audio processing (keyword spotting), and sensor fusion. Additionally, this team member will support the research team’s algorithm development for the next version of the Akida NSoC. BrainChip’s Akida fully-digital neuromorphic AI processor has the first of its kind, licensable product that delivers great efficiency in Edge AI devices along with unique capabilities like learning on the edge. Akida’s success is dependent on finding the best product market fit in large markets, supported by a product roadmap that has the right combination of features and capabilities that delivers the best differentiated value, with many internal functions and external entities. This is an ON-SITE ROLE ESSENTIAL JOB DUTIES AND RESPONSIBILITIES: Quickly prototyping, training, and testing machine learning (ML) solutions using online code repositories, research publications, or customer specifications Keeping abreast of developments in ML such as new development tools, libraries, and frameworks as well as new ML models/architectures, training techniques, and application pipelines Participating in ML algorithm/hardware co-design tasks Performing, documenting, and presenting detailed analyses related to ML algorithm development, software/hardware benchmarking, and application development Obtaining a thorough understanding of the Akida 1.0 hardware device and associated software stack (MetaTF) Interfacing with customers to discuss ML application goals, constraints, and opportunities JOB REQUIREMENTS: Education/Experience: Master’s Degree in Computer Science or equivalent. PhD preferred. Course work in machine learning and computer vision Strong programming skills in Python Three years’ experience developing ML applications in either TensorFlow/Keras and/or PyTorch Excellent communication skills Experience in one or more of the following application fields: Image Processing/Computer Vision, ADAS, Anomaly Detection, Audio/Speech Processing, or Sensor Fusion. Preferred Qualifications: Ten plus years’ experience developing ML applications in TensorFlow/Keras and PyTorch Multi-project experience in object classification, object detection, face recognition, and/or keyword spotting Knowledge of deep learning quantization techniques Experience with Docker and Git Experience with Scrum/Agile software development (e.g. Jira) Benefits Offered:

  • Competitive Pay depending on experience $130K-190K base
  • Restricted Stock Units
  • Bonus Pay
  • 401K with matching
  • Free Lunch Daily
  • Company-paid Medical HMO, Dental PPO, and Vision Insurance
  • Company-paid Life Insurance and AD&D
  • Paid Time Off
  • Holiday Pay
  • Employee Assistance Program, Caregiver Support, Adoption Assistance Program
  • Flexible Spending Account
  • Health Savings Account
  • Commuter Benefit Program
  • Employee Discounts BrainChip is an equal opportunity employer. All aspects of employment including the decision to hire, promote, discipline, or discharge, will be based on merit, competence, performance, and business needs. We do not discriminate on the basis of race, color, religion, marital status, age, national origin, ancestry, physical or mental disability, medical condition, pregnancy, genetic information, gender, sexual orientation, gender identity or expression, veteran status, or any other status protected under federal, state, or local law.

YES!

To all the persistent whingers whinging about the lack of revenue and forever feeling butthurt about one thing to the next... take note of this job advert.

BRN are looking to take on a new ongoing role, in RESEARCH.... not sales or marketing or some other area.... RESEARCH... and for $130-$190K base salary (US).

Let that sink in (if that's at all possible).

PS: LMAO
 
  • Like
  • Haha
  • Wow
Reactions: 9 users

wilzy123

Founding Member
  • Haha
  • Like
Reactions: 11 users

DK6161

Regular
YES!

To all the persistent whingers whinging about the lack of revenue and forever feeling butthurt about one thing to the next... take note of this job advert.

BRN are looking to take on a new ongoing role, in RESEARCH.... not sales or marketing or some other area.... RESEARCH... and for $130-$190K base salary (US).

Let that sink in (if that's at all possible).

PS: LMAO
Feels like we are forever researching and collaborating. 😔
 
  • Like
  • Sad
  • Fire
Reactions: 8 users

wilzy123

Founding Member
  • Like
  • Fire
Reactions: 7 users

DK6161

Regular
  • Like
  • Love
Reactions: 8 users
Top Bottom