BRN Discussion Ongoing

Sirod69

bavarian girl ;-)

Nvidia and Intel team up on chips for metaverse workstations​


The need for accelerated computing power is growing exponentially due to the explosion of AI-augmented workflows, from traditional R&D and data science workloads to edge devices on factory floors or in security offices, to generative AI solutions for text conversations and text-to-image applications, the companies said.

Next-generation platform features​

With a breakthrough new compute architecture for faster individual CPU cores and new embedded multi-die interconnect bridge packaging, the Xeon W-3400 and Xeon W-2400 series of processors enable scalability for increased workload performance. Available with up to 56 cores in a single socket, the top-end Intel Xeon w9-3495X processor features a redesigned memory controller and larger L3 cache, delivering up to 28% more single-threaded and 120% more multi-threaded performance over the previous-generation Xeon W processors.

Based on the Nvidia Ada Lovelace GPU architecture, the latest Nvidia RTX 6000 brings power efficiency and performance to the new workstations. It features 142 third-generation RT Cores, 568 fourth-generation Tensor Cores and 18,176 latest-generation CUDA cores combined with 48GB of high-performance graphics memory to provide up to 2x ray-tracing, AI, graphics and compute performance over the previous generation.

And Nvidia’s ConnectX-6 Dx SmartNICs enable professionals to handle demanding, high-bandwidth 3D
rendering and computer-aided design tasks, as well as traditional office work with line-speed network
connectivity support based on two 25Gbps ports and GPUDirect technology for increasing GPU bandwidth by 10x over standard NICs. The high-speed, low-latency networking and streaming capabilities enable teams to move and ingest large datasets or to allow remote individuals to collaborate across applications for design and visualization.

The new workstations will be available in the coming months, and you can preorder from Boxx and HP starting today.

 
  • Like
Reactions: 8 users
Another new role advertised - this one in India

1676675787990.png


Check out this job at BrainChip:
https://www.linkedin.com/jobs/view/2679662150
 
Last edited:
  • Like
  • Fire
Reactions: 36 users

jtardif999

Regular
  • Like
  • Fire
Reactions: 4 users
Just posted by Edge Impulse

1676679021727.png


 
Last edited:
  • Like
  • Fire
  • Love
Reactions: 86 users

Calsco

Regular
This chart indicates we are still in a downward trend and also in a descending wedge which means as we come to the pointy end of the wedge the share price needs to make a decision if it will go up or down. There is a strong monthly support at .50c but if we drop below that then the next significant support is around .38.

I hope that an announcement or some form of licensing agreement comes out soon to encourage the market to react positively for the share price. If not I can see the share price sinking even lower.
 

Attachments

  • 5B57CA3F-3466-46B2-A61D-351008B00BEE.jpeg
    5B57CA3F-3466-46B2-A61D-351008B00BEE.jpeg
    6 MB · Views: 70
  • Like
  • Sad
Reactions: 8 users

Townyj

Ermahgerd
  • Like
  • Fire
  • Love
Reactions: 28 users

Slade

Top 20
Lead sponsors.. Oh Brainchip is definitely a joke.... Sitting right there next to TI.
San Jose, Amsterdam, Chicago, Munich, Boston and London.

Invite-only event series to explore how to harness real-world data to deploy advanced edge machine learning (ML) solutions at scale.
 
  • Like
  • Fire
  • Love
Reactions: 25 users

Steve10

Regular
The new chip tape out at Global Foundries could be for Qualcomm. Otherwise why tape out a new chip at Global Foundries?

GlobalFoundries has operations across North America, Europe, and the Asia Pacific, serving a range of clients across a variety of end markets including mobile devices, communications, IoT, and automotive. Notable customers of GlobalFoundries include Qualcomm, MediaTek, and NXP Semiconductors.

Qualcomm to spend $4.2 billion more on chips from GlobalFoundries​

 
  • Like
  • Love
  • Fire
Reactions: 27 users
I stumbled across this new technology that the US Army is using in their Enhanced Night Vision Goggles - Binoculars (ENVG-B) while following some Ukrainian chats (guilty pleasure of mine). Essentially, it's a technology released in 2019 (pre-dates Akida chips) which enables the combatants to see in low-light conditions, and the newest enhancement by L3Harris includes some cool edge-detection capabilities which enhance the target acquisition time. Instead of the typical laser-green monochrome you'd expect to associate with night-vision goggles, the ENVG-B offers views that would fit right into a futuristic video game. Clear neon white outlines of people and artillery, detailed trees and brush, bright light blue figures, and tactical information are all displayed right in front of the soldier's eyes.




View attachment 29911

It got me thinking on how much processing power is required just for the edge-detection function, and how this would be a perfect application for Akida. So I searched for Akida applications in edge detection and I found this document (Published: 11 December 2022) attached, "Implementation of the Canny Edge Detector Using a Spiking Neural Network" authored by Krishnamurthy V. Vemuru, whom some of you would recognise as a Brainchip employee.

View attachment 29908

View attachment 29909
View attachment 29910


As you can see from the images, the SNN Canny edge detector picks up more detail than the conventional Canny breed.

In the interest of power, space and weight savings, not to mention detail, I'd be surprised if L3Harris isn't already experimenting with Akida.

Get that sheet to Ukraine!
 
  • Like
  • Fire
Reactions: 6 users

Esq.111

Fascinatingly Intuitive.
Afternoon Chippers,

We got a mention in the weekend financial paper, . ...... damit.

Esq.
 

Attachments

  • 20230218_122352.jpg
    20230218_122352.jpg
    2.9 MB · Views: 269
  • Sad
  • Haha
  • Like
Reactions: 10 users

BaconLover

Founding Member
  • Like
  • Fire
  • Love
Reactions: 22 users
Afternoon Chippers,

We got a mention in the weekend financial paper, . ...... damit.

Esq.
Just had a visit from a mate who I convinced to buy brainchip a while back 😔
He's cool but I still feel like shit.
 
  • Like
  • Sad
  • Thinking
Reactions: 22 users
This chart indicates we are still in a downward trend and also in a descending wedge which means as we come to the pointy end of the wedge the share price needs to make a decision if it will go up or down. There is a strong monthly support at .50c but if we drop below that then the next significant support is around .38.

I hope that an announcement or some form of licensing agreement comes out soon to encourage the market to react positively for the share price. If not I can see the share price sinking even lower.
The 5 day chart shows an uptrend, since the attack and drop where over 5 million "reported" shorts, were taken out (1.8 million were taken out the day before).

_20230218_123356.JPG


_20230218_123711.JPG


The share price rose despite another 1.6 million shorts being taken out on the 16th.
Although it can't be known if these have already been sold and scooped up by buyers.

What are the effects of shorting, on chart dynamics?
 
Last edited:
  • Like
Reactions: 11 users

AARONASX

Holding onto what I've got
  • Like
  • Fire
  • Love
Reactions: 72 users
I see our mates at TCS are running a tutorial at this upcoming conference in June.

2023 IEEE International Conference on Acoustics, Speech and Signal Processing

They list some examples of dedicated sophisticated neuromorphic but we know which one of those is commercially available and also been linked to TCS.

Hopefully they discuss our product in more detail in the tutorial.



A Tutorial on Bio-Inspired Neuromorphic Computational Paradigms
Presenters

Manan Suri (IIT Delhi), Sounak Dey (Tata Consultancy Services Ltd.), Arun M. George (TCS Research & Innovation)


INTRODUCTION

Embedding intelligence at the edge has become a critical requirement for many industry domains, especially disaster management, healthcare, manufacturing, retail, surveillance, remote sensing etc. Classical Machine learning or Deep learning (ML/DL) based systems, being heavy in terms of required computation and power consumption, are not suitable for Edge devices such as robot, drones, automated cars, satellites, routers, wearables etc. which are mostly battery driven and have very limited compute resource. Inspired from the extreme power efficiency of mammalian brains, an alternative computing paradigm of Spiking Neural Networks (SNN) also known as Neuromorphic Computing (NC), has evolved with a promise to bring in significant power efficiency compared to existing edge-AI solutions. NC follows non-von Neumann architecture where data and memory are collocated like brain neurons and SNNs handle only sparse event-based data (spikes) in asynchronous fashion. Inherently SNNs are very efficient to understand features in temporally varying signals and is found to efficiently classify/process auditory data, gestures/actions from video streams, spot keywords from audio streams, classify & predict time series from different sensors used in IoT, regenerate temporal patterns etc. The community is pursuing multiple sophisticated dedicated Neuromorphic hardware platforms such as: Intel Loihi, IBM TrueNorth, Brainchip Akida, SpiNNaker, DYNAPs to name a few.

Moreover, ultra-advanced and futuristic nanoelectronic devices and materials are being explored to build energy efficient neuromorphic computers. So this domain, as well as this tutorial, lines in the intersection of Computational Neuroscience, Machine Learning and In-memory Neuromorphic Computation techniques.

RATIONALE AND STRUCTURE OF THE TUTORIAL

The ICASSP community is at the forefront of research in the domain of signal processing. Thus it is extremely relevant to conduct a tutorial on advances in the domain of Neuromorphic Computing at the forum. We are proposing two valuable cross vertical elements in this tutorial:

(i) Firstly, the proposed tutorial has been developed keeping both academic and industrial/application interests in mind. The speakers represent leading academic and industry research teams on the subject with several years of theoretical and applied experience on the topic. Over last few years, while solving customer requirements related to edge computing, TCS Research has successfully taken neuromorphic research to real market applications. At the same time, group at IIT-Delhi has contributed significantly towards development of cutting-edge neuromorphic hardware and memory-inspired computing.

(ii) Secondly, the proposed tutorial will not only cover foundational basics (i.e. algorithms, bio-inspiration, mathematics) of the subject, but will also delve in to real hardware-level implementation and actual application use-cases (such as gesture recognition in robotics, time series classification and prediction in IoT, continuous health monitoring, remote sensing via satellite etc.) as pursued in industry so far.

The tutorial is structured to cover all relevant aspects of SNN and NC as detailed below.

Biological Background, Software & Simulation of SNNs:
We will start the tutorial by giving a background of modus operandi of biological neurons, their equivalent computational models, spike generation process, synaptic weight updates and learning rules. Next, we will speak in detail about (i) existing software tools and SNN simulators, (ii) how to create a basic SNN, feed data into it and do a simple classification task, (iii) how to tune it towards better performance.

Speaker: Sounak Dey, duration: 50 minutes.

Neuromorphic Hardware Basics:
Second part of the tutorial will cover the basics of dedicated hardware approaches for implementing neuromorphic algorithms in real world. Key techniques will be bench-marked for their pros and cons. How specialized neuromorphic hardware can offer performance benefits will be discussed. The session will end with discussion of futuristic nanomaterials proposed for neuromorphic computation.

Speaker: Manan Suri, duration: 50 minutes.

Application & Implementation:
Third and final part of the tutorial will provide a detailed landscape of the applications that have been developed and tested so far using SNN and NC, including our own experiences. Aspects such as spike encoding techniques, conversion of ANNs to SNNs, FPGAs will be discussed.

Speaker: Arun George, duration: 50 minutes.

A flexible and interactive model of discussion and Q-A with the audience will be followed throughout the tutorial. Duration:10-30 minutes.
 
  • Like
  • Fire
  • Love
Reactions: 23 users

TopCat

Regular
This is new on the nViso website.

66A2D589-9B39-43BA-B2A9-83BF49CFB656.jpeg
 
  • Like
  • Thinking
Reactions: 11 users

TECH

Regular
Partnering with major companies who can not only "talk the talk" but "walk the walk".

We can't travel this road alone anymore, hence all the partnerships announced over the last 12 months or so, they all open up
doors that may never have been possible, especially in the short term, and while the current share price is frustrating for many,
hovering around 52 week lows, we have taken a number a big steps forward in the overall organization of the company, managed
to expand the staff base worldwide, control day-to-day outgoings and to top that off, we have a more advanced version of Akida
under production, which is only really 3 months out approximately, as well as engineering teams and designers working hard to
keep Brainchip at our favorite position on the grid.

This video below, merely shows off just one of our partners, whose reach is massive. I see our connection within India really expanding
over the next 5 years, Anil knows the massive potential from within his original homeland, just think for a moment of all his contacts and the respect he enjoys within the chip industry.

To bag Brainchip is to bag yourself, focus on the positive, "we are making good progress"!

 
  • Like
  • Fire
  • Love
Reactions: 63 users

Slade

Top 20
We are 100 % going to be part of Mercedes Benz, the worlds number 1 car marker .

The rest will , follow the Leader .

Some exciting quotes below

@Tothemoon24 you keep giving us gems.

“Mercedes has experimented with a new type of processor that performs tasks in “neuromorphic spikes”. Put simply, this means that the computer stores up tasks and executes them in one go once a threshold is reached, saving energy and boosting driving range in the process.

Mercedes EQA saloon 2

Unlike premium rivals such as Volvo and Polestar, who have paired with Google for their infotainment software, the German firm is going it alone with the development of MB.OS, and Chief Software Officer Magnus Östberg told Automotive Daily: “We are the architects of our own house, as it were. It’s important to have ownership of your own OS for safety and security.”
 
  • Like
  • Love
  • Fire
Reactions: 58 users

Kachoo

Regular
Just posted by Edge Impulse

View attachment 29920

This is happening every week or month. There are new articles out daily with blurbs about AKIDA or BRN. How can a well informed person really think not much is going on with our little company is beyond me.

Look one can be dirty that we have not locked in more revenue question certain aspects that makenrhem in happy but that's for you to write TD and iron out your issues then decied if your not happy and will move on or happy and stay.

I can guess this whole dramatic effect was used to get cheap shares scare off some investors and holders that are either not well informed or had enough on their plate.

It's easy kick the company when total shares being sold will over whelm the buy side. Trigger stops and move it up a few days weeks later why.

LDA selling
Some shorts selling but really buying cheaper
Bombard the TSE and HC with negativity to a point that even the mosts informed people have had enough and leave (bully attitude.)
Then a push down gets bigger with people's stops.

This is my opinion what I believe has happened.

Again tommorow is a new day and Nueromorphic is just gaining traction.

GLA
Happy Weekend!
 
  • Like
  • Love
  • Fire
Reactions: 38 users
Top Bottom