BRN Discussion Ongoing

[Fact Finder]

I think it was also for the physical design work socionext did as well as the first production run. Socionext played a major part in designing our chip.

And testing after it came out of fab at TSMC and before being handed to Brainchip.

In a webinar or early podcast Mr. Dinardo spoke about the work done by Socionext and how well they continued to work together.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Violin1]

A few more slightly negative comments today. Not sure if it's the low 90s making people nervous or the odd escapee from the dark-side - but I still like to consider that we are up by 50% or so from the end of 2021 in a year where we always expected the most in the second half. So this makes me excited for July onwards. I'm not expecting massive $ in the next quarterly and I'm sure Sean said we were likely to be a net cash user until 2025 so that tempers my expectations a little as well.

But the feel from our company is all good. There's no negative sentiment in presentations or responses to queries - all very positive, streets ahead, plenty of interest. Remember, Brainchip has pretty well done what it said it would for the past three years and thrown in an odd fun surprise. I'm chilled....well I'm not actually, that's another of our crew....but you know what I mean.

You've lost nothing unless you sell. It's a great company and this is a great site to discuss it. Just don't go to the dark-side young Luke!!!

 

[Mt09]

I have been hesitant to guess/speculate on the likely sales (to date) as really it is just a guessing game. The use of the word 'mass' production of chips caused a lot of angst amongst holders as to what this actually meant. It was a shame we never got any concrete figures on this!

However, I have to hope that if that number was in the vicinity of 7,000 as previously discussed that this resulted in a similar number of products i.e. Raspberry Pi and later the PCIe Boards. Also there was a mention of production slots being booked for further chips, whatever that really meant is anyone's guess? Did we produce more or were we just waiting for a customer to come on board with an order?? Why book the production slots at all if we were not anticipating a need for them???

So I want to believe that the company would not have hesitated to produce a large number of their products in anticipation of a huge uptake considering they are the first to market in this area. It is then realistic, in my opinion, that the quarterly sales will be much higher than some are thinking.

Of course I base all of this on my assumption on how our company wanted to be perceived by it's likely competitors and the market place in general. A few hundred sales is not what I am expecting, I am hoping to see thousands. The fact that they sold out of the Raspberry Pi indicates a genuine desire for such a product in the market and is encouraging that the demand is there.

If I am right then revenue for this quarter should be up to $5 million but of course you will notice that I use the word 'guess' a lot so I will just wait and see.

The company aren’t trying to sell thousands of development boards, even sales of 50 dev kits could result in 20 IP deals, 20 orders of less than a million units of akd1000 and ten lost sales. And before you know it we’re getting royalties on millions of chips containing the IP

 

[Fox151]

JEROME NADEL JOINS BRAINCHIP AS CHIEF MARKETING OFFICER

Laguna Hills, Calif. – January 25, 2022 – BrainChip Holdings Ltd (ASX: BRN, OTCQX: BRCHF, ADR: BCHPY), a leading provider of ultra-low power high performance artificial intelligence technology and the world’s first commercial producer of neuromorphic AI chips and IP, today announced that semiconductor marketing expert Jerome Nadel has joined the company as Chief Marketing Officer. Nadel will drive corporate and product marketing to support the full commercialization of its Akida™ neuromorphic computing platforms.

Nadel, an internationally experienced executive in scaling and growth-stage companies, joins BrainChip from Rambus, a publicly traded semiconductor technology company. During his eight years at Rambus, he served as Senior Vice President, CMO, and General Manager of its fintech software division. He has also held senior positions at Sagem, Thales (formerly Gemalto), Option NV, Human Factors International, Unisys, and IBM.

He is also a board member of the Silicon Valley Executive Network, alongside BrainChip CEO Sean Hehir, as well as President of the Silicon Valley chapter of the CMO Club, a global community of senior marketing executives. He received a BS in Psychology from the University of South Florida and an MS in Industrial/Organizational Psychology from Kansas State University, where he was a doctoral candidate.

“BrainChip advancements in neuromorphic AI are truly transformative and will be the computational foundation of smart sensing in all edge devices,” said Nadel. “Our ground-breaking IP is commercially ready and relevant, and it’s exciting to see it being adopted and deployed by industry leaders. I’m excited to play a part in accelerating the company’s trajectory.”

“We’re eager to hand Jerome the reins of BrainChip’s corporate and product marketing, and proud to have him on the executive team,” said Hehir. “This is a pivotal moment for BrainChip as we steer the company from R&D breakthroughs to a focus on commercialization, strategic partnerships, and profitability.”

BrainChip deploys AI at the edge in a way that existing technologies cannot. The company’s tech is both high-performance and ultra-low power, enabling a range of capabilities including on-chip, in-device one-shot learning. BrainChip’s IP can be used in a wide range of applications from industrial IoT, cybersecurity, and autonomous vehicles to smart sensors that can detect and act on visual features, odors, taste, touch, and sound

BRAINCHIP ACHIEVES FULL COMMERCIALIZATION OF ITS AKD1000 AIOT CHIP WITH AVAILABILITY OF MINI PCIE BOARDS IN HIGH VOLUME

Company welcomes enterprises to the Edge with AIoT PCIe board and board design layout​

Laguna Hills , Calif. – January 17, 2022 – BrainChip Holdings Ltd (ASX: BRN, OTCQX: BRCHF, ADR: BCHPY), a leading provider of ultra-low power high performance artificial intelligence technology and the world’s first commercial producer of neuromorphic AI chips and IP, today announced that it has begun taking orders for the first commercially available Mini PCIe board leveraging its Akida™ advanced neural networking processor, rounding out its suite of AKD1000 offerings.

The AKD1000-powered Mini PCIe boards can be plugged into a developer’s existing system to unlock capabilities for a wide array of edge AI applications, including Smart City, Smart Health, Smart Home and Smart Transportation. BrainChip will also offer the full PCIe design layout files and the bill of materials (BOM) to system integrators and developers to enable them to build their own boards and implement AKD1000 chips in volume as a stand-alone embedded accelerator or as a co-processor.

The new boards help usher in a new era of AI at the edge due to their performance, security, low power requirements, and the ability to perform AI training and learning on the device itself, without dependency on the cloud. The production-ready chips provide high-speed neuromorphic processing of sensor data at a low cost, high speed and very low power consumption. The PCIe boards are immediately available for pre-order on the BrainChip website. Pricing starts at $499.

“I am excited that people will finally be able to enjoy a world where AI meets the Internet of Things,” said Sean Hehir, BrainChip CEO. “We have been working on developing our Akida technology for more than a decade and with the full commercial availability of our AKD1000, we are ready to fully execute on our vision. Other technologies are simply not capable of the autonomous, incremental learning at ultra-low power consumption that BrainChip’s solutions can provide. Getting these chips into as many hands as possible is how the next generation of AI becomes reality.”

The launch of BrainChip’s new PCIe board closely follows the company’s development kit offerings introduced in October. The two development kits – an x86 Shuttle PC development kit, as well as an ARM-based Raspberry Pi development kit – both include the AKD1000 chip on a Mini PCIe board and are available to partners, large enterprises and OEMs. BrainChip’s AKD1000 chips and PCIe board can be purchased at shop.brainchipinc.com or via the Buy Now button at www.brainchip.com/.

INFORMATION SYSTEMS LABS JOINS BRAINCHIP EARLY ACCESS PROGRAM

Laguna Hills, Calif. – January 9, 2022 – BrainChip Holdings Ltd (ASX: BRN, OTCQX: BRCHF, ADR: BCHPY), a leading provider of ultra-low power, high performance artificial intelligence technology and the world’s first commercial producer of neuromorphic AI chips and IP, today announced that Information Systems Laboratories, Inc. (ISL) is developing an AI-based radar research solution for the Air Force Research Laboratory (AFRL) based on its Akida™ neural networking processor.

ISL is an employee-owned technology development corporation that performs expert research and complex analysis, software and systems engineering, advanced hardware design and development, and high-quality specialty fabrication for a variety of customers worldwide.

ISL specializes in the areas of advanced signal processing, space exploration, undersea technologies, surveillance and tracking, cyber security, advanced radar systems, and energy independence. As a member of BrainChip’s Early Access Program, ISL will be able to evaluate boards with the Akida device, software and hardware support and dedicated engineering resources.

“As part of BrainChip’s EAP, we’ve had the opportunity to evaluate firsthand the capabilities that Akida provides to the AI ecosystem,” said Jamie Bergin, Senior VP, Manager of Research, Development and Engineering Solutions Division at ISL.

BrainChip brings AI to the edge in a way that existing technologies are not capable. The Akida processor is ultra-low power with high performance, supporting the growth of edge AI technology by using a neuromorphic architecture, a type of artificial intelligence that is inspired by the biology of the human brain. Devices currently available to BrainChip’s EAP customers provide partners with capabilities to realize significant gains in power consumption, design flexibility and true learning at the Edge.

“ISL has decided to use Akida and Edge-based learning as a tool to incorporate into their portfolio of research engineering and engineering solutions in large part due to our innovative capabilities and production-ready status that provides go-to-market advantages,” said Sean Hehir, BrainChip CEO. “We are pleased to be included as the AI- and Edge-based learning component of ISL’s research sponsored by AFRL. We feel that the combination of technologies will help expedite its deployment into the field.”

Akida is currently available now to be licensed as IP, as well as available for orders for production release in silicon. Its focus is on low power and high-performance, enabling sensory processing, for applications in Beneficial AI, as well as applications including Smart Healthcare, Smart Cities, Smart Transportation and Smart Home. Those interested in learning how BrainChip has solved the problems inherent in moving AI out of the data center to the Edge where data is created, can visit https://brainchipinc.com/technology/ for more information.

 

[Build-it]

I have been hesitant to guess/speculate on the likely sales (to date) as really it is just a guessing game. The use of the word 'mass' production of chips caused a lot of angst amongst holders as to what this actually meant. It was a shame we never got any concrete figures on this!

However, I have to hope that if that number was in the vicinity of 7,000 as previously discussed that this resulted in a similar number of products i.e. Raspberry Pi and later the PCIe Boards. Also there was a mention of production slots being booked for further chips, whatever that really meant is anyone's guess? Did we produce more or were we just waiting for a customer to come on board with an order?? Why book the production slots at all if we were not anticipating a need for them???

So I want to believe that the company would not have hesitated to produce a large number of their products in anticipation of a huge uptake considering they are the first to market in this area. It is then realistic, in my opinion, that the quarterly sales will be much higher than some are thinking.

Of course I base all of this on my assumption on how our company wanted to be perceived by it's likely competitors and the market place in general. A few hundred sales is not what I am expecting, I am hoping to see thousands. The fact that they sold out of the Raspberry Pi indicates a genuine desire for such a product in the market and is encouraging that the demand is there.

If I am right then revenue for this quarter should be up to $5 million but of course you will notice that I use the word 'guess' a lot so I will just wait and see.

I agree however my crystal ball has been a little off lately so I won't speculate on revenue only to say the company has forecasted explosive growth.

When the Production Manager job was advertised back in December i read that as production was and is expected to ramp up and would be a most important role as per the job add.

Production Manager:

Location: Laguna Hills, CA

Job Description:

We are looking for a Production Manager to join our team. The ideal candidate has experience in electronics manufacturing, either in-house or by a contract manufacturer, parts sourcing, and price negotiation. We are looking for someone with previous experience as a Production Manager or as an assistant Production Manager who is eager to make the step up and assist in planning, organizing, and controlling production to ensure that goods are produced efficiently, and on time per plan.

Key Responsibilities:

Operationalize the end-to-end manufacturing and distribution of current and future product offerings, defining the criteria for parts procurement, assembly, test, and shippingContinuously identify and implement operational efficienciesParticipate in the sourcing, selection, and management of all direct and indirect supplier relationship.Authors, revises, and manages all necessary documentation, change management, and governance processes, including QA, fulfillment, and support, for delivering products to customers at scaleResponsible for materials forecasting based on production demand; ensures production materials are on-hand and ready for use for on-time order fulfillmentLeads manufacturing readiness program in support of new product introductionsAssists team with defining, quoting, and bringing new products into productionActs as liaison with both customers and suppliersWork closely with marketing, focusing on demand generation, e-commerce efforts to ensure a seamless customer experienceResponsible for in-house production of low volume development systemsDrives timely decisions and facilitates active communication and information flowEstablishes KPIs, tracks/reports monthlyOther duties as time allowsProject manage the IP delivery program to IP LicenseesMay perform other or specified duties as directed

This position is 40% production functions, 40% purchasing functions, and 20% of other operational and reporting tasks.

Qualifications and Experience:

Typically requires a minimum of 10 years of related work experience including at least 2 years of successful leadership/supervisory experience and a Bachelor’s degree or equivalent combination of education and experience.5 years experience in supply chain management within technology industriesStrong track record of leading and implementing production projectsA team player who is accurate, detailed, committed to high-quality standards, and proactive in finding solutions to achieve successful outcomes.Highly organized, self-motivated, detailed, and resourceful leader with excellent executionMust have Budget management and optimization skills

Sourcing/Procurement experience required

https://www.google.com/url?sa=t&source=web&rct=j&url=https://brainchipinc.com/production-manager/&ved=2ahUKEwi2-_-nqOr2AhUgR2wGHXm8BUsQFnoECA4QAQ&usg=AOvVaw0tYepe5qDAssuWLPXgHEEo

Would anyone be aware of who may have been the successful applicant.

Edge Compute.

 

[davidfitz]

The company aren’t trying to sell thousands of development boards, even sales of 50 dev kits could result in 20 IP deals, 20 orders of less than a million units of akd1000 and ten lost sales. And before you know it we’re getting royalties on millions of chips containing the IP

True, but the term mass production of chips has always intrigued me.

Why use that term if you are only looking at selling 50 dev kits.

Either way the revenue should be increasing every quarter going forward so all shareholders will be better off as a result

 

[Fact Finder]

The company aren’t trying to sell thousands of development boards, even sales of 50 dev kits could result in 20 IP deals, 20 orders of less than a million units of akd1000 and ten lost sales. And before you know it we’re getting royalties on millions of chips containing the IP

I agree.

This is how I see it:

1. BRN selected about two dozen from the ‘north of 100 NDA’ potential customers and they became EAP’s and received the first boards and engineering AKD1000 chip/s at about $50,000 a pop.

2. This left 75 plus potential NDA customers and in my opinion they would be the target market for the $9,999.00 Shuttle PC’s along with the panel of small engineering firms that Rob Telson was/is putting together such as Eastronics. It is an open question whether BRN would charge Eastronics and other small engineering firms???

3. The Raspberry Pi at $4,999 would be targeted at the small engineering firms, academics/universities/research facilities - 50 to 150 would be my guess

4. The PCle Board was aimed at Academics, University Students, researchers, hobbyists, and random members of the public - CEO Sean Hehir said hundreds sold not thousands so I would say 500 or less.

You can do the numbers but it is obvious my guesses would not hit anywhere near 10 million dollars.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Fact Finder]

True, but the term mass production of chips has always intrigued me.

Why use that term if you are only looking at selling 50 dev kits.

Either way the revenue should be increasing every quarter going forward so all shareholders will be better off as a result

The term used was ‘volume’ production not ‘mass’ production.

I think it was Jesse Chapman (?) who asked TSMC what ‘volume’ was I think they said 10 million chips. However you cut the numbers the payments made by Brainchip to Socionext and TSMC would never have bought 10 million chips.

If you go back to the 7,000 chip number suggested early on and if you give a hundred chips to each of the small engineering firms and 50 to each of 20 EAP’s and put a few in the products and hand a few around to engineers at Brainchip France, India, Australia & Perth you will not have much change from 7,000 chips.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[davidfitz]

Thanks FF, when I was googling this morning I saw a couple of articles that referred to that term but as we all know journalists are not exactly accurate when reporting on our company.

Yes it was Jesse and I appreciate your logical conclusion of where those chips likely ended up.

BrainChip Share Price Up as Production of Its Neuromorphic (ASX:BRN)

The BrainChip Holdings (ASX:BRN) share price flying 22.64% higher today, at time of writing. Surging on the back of news that production of the company’s novel neuromorphic chip...

www.moneymorning.com.au

Brainchip begins mass production of its neuromorphic AI chips - Stockhead

Brainchip announced that it has commenced mass production of its Akida AKD1000 neuromorphic processor chips for use in edge AI devices.

stockhead.com.au

 

[Potato]

Are 4C's normally released in April or May?

 

[yogi]

Thats about how I worked it for the 4c

"When asked about when we will know about products containing Akida in the market Sean responded that he does not foresee a time and most
customers are pretty serious about keeping it confidential. He then added: "of course, I know (trying to stop himself from smiling), but most people could not." He later mentioned that the best way to get an idea of this will be to keep an eye on financials." from YLJ post

I think we were very lucky that Mercedes let out their connection to BRN earlier this year for the sake of the long term holders and the temporary share price.
When the Megachip deal was announced I think a lot of us immediately became aware that the likelihood of many announcements about deals were likely to be very thin on the ground. This apparently has really now been confirmed by Sean.
I can see the share price from here on as suffering until we have a significant 4C that is indicative of potential income.
For me this gives me another "in" target for buying more shares the week before each 4C. (not financial advice)
Of course it will be interesting to see if they ever give a breakdown of royalties, upfront fees and engineering fees in the future 4C's.

Where do i listen to the interview , was out and about today with work and so much is happening here dont know where to pick up thank you

 

[davidfitz]

From memory we have some solutions regarding this.

Almost $9 billion cybersecurity and intelligence package to be unveiled in federal budget​

Multi-billion-dollar cybersecurity and intelligence budget package to be unveiled as global threats rise

An almost $9 billion package to bolster Australia's cybersecurity and intelligence capabilities will be unveiled in tonight's federal budget, as concerns grow over threats from potential adversaries across the globe.

www.abc.net.au

"The multi-agency centre will ensure Australia, working with our allies, can better anticipate and capitalise on emerging technologies."

 

[Deena]

Almost $9 billion cybersecurity and intelligence package to be unveiled in federal budget . I wonder if they are astute enough to use Brainchip's cyber security offering. It could save them a lot of money and be far more effective than alternatives!​

Multi-billion-dollar cybersecurity and intelligence budget package to be unveiled as global threats rise

An almost $9 billion package to bolster Australia's cybersecurity and intelligence capabilities will be unveiled in tonight's federal budget, as concerns grow over threats from potential adversaries across the globe.

www.abc.net.au

 

[davidfitz]

Almost $9 billion cybersecurity and intelligence package to be unveiled in federal budget . I wonder if they are astute enough to use Brainchip's cyber security offering. It could save them a lot of money and be far more effective than alternatives!​

Multi-billion-dollar cybersecurity and intelligence budget package to be unveiled as global threats rise

An almost $9 billion package to bolster Australia's cybersecurity and intelligence capabilities will be unveiled in tonight's federal budget, as concerns grow over threats from potential adversaries across the globe.

www.abc.net.au

Snap

 

[Straw]

Are 4C's normally released in April or May?

Jan/Feb/March reporting period (4C end of April) - sometimes a few days earlier but usually at end of month

 

[Proga]

Well said CEO needs to earn his money not good enough to sit there hoping while shareholders suffer huge losses.

His job is not to concern himself with shareholder value in the short term. He was hired to put the building blocks in place to grow the company exponentially over the mid to long term as BRN transition out of a R&D only into commercialisation.

 

[VictorG]

When in doubt!

 

[Deleted member 118]

Well US $10 million would really shake a few shorters up they would be very nervous to hear that sort of figure.

 

[AusEire]

Hey @Fact Finder @Diogenese what you guys think of this?
https://www.zdnet.com/article/how-s...are-to-up-the-camera-game-on-its-smartphones/

@Fact Finder @Diogenese
Did you guys take a look at this article?

 

[Fact Finder]

Is Neuromorphic Computing Using Edge The Future Of AI?​

By
Victor Dey
-
March 16, 2022
https://www.facebook.com/sharer.php...orphic-computing-using-edge-the-future-of-ai/
https://twitter.com/intent/tweet?te...using-edge-the-future-of-ai/&via=DatatechVibe
https://www.linkedin.com/shareArtic...orphic+Computing+Using+Edge+The+Future+Of+AI?
https://api.whatsapp.com/send?text=...orphic-computing-using-edge-the-future-of-ai/

Neuromorphic processors aim to provide vastly more power-efficient operations by modelling the core workings of the brain​

As artificial intelligence (AI) continues to evolve, it is expected that AI at the edge will become a more significant portion of the current tech market. Known as the AI of Things or AIoT, various processor vendors like Intel and Nvidia have launched AI chips for such lower-power environments, respectively, with their Movidius and Jetson product lines.
Computing at the edge further aids in lower latency than sending information to the cloud. Ten years ago, there were questions about whether software and hardware could be made to work similar to a biological brain, including incredible power efficiency.
Today, the same question has been answered with a yes with advancement in technology, but the challenge now is for the industry to capitalise on neuromorphic technology development and answer tomorrow’s regressive computing challenges.

The Crux Of Neuromorphic Computing​

Neuromorphic computing differs from a classical approach to AI, which is generally based on convolutional neural networks (CNNs), as this technology mimics the brain much more closely through spiking neural networks (SNNs).
Although neuromorphic chips are generally digital, they tend to work based on asynchronous circuits, meaning there is no global clock. Depending upon the specific application, neuromorphic can be ordered to magnitude faster and requires less power. Neuromorphic computing complements CPU, GPU, and FPGA technologies for particular tasks, such as learning, searching and sensing, with extremely low power and high efficiency.
Researchers have lauded neuromorphic computing’s potential, but the most impactful advances to date have occurred in academic, government and private R&D laboratories. That appears to be ready to change.
A report by Sheer Analytics & Insights estimates that the worldwide market for neuromorphic computing will be growing at 50.3 per cent CAGR to $780 million over the next eight years. Mordor Intelligence, on the other hand, aimed lower with $111 million and a 12 per cent CAGR to reach $366 million by 2025.
Forecasts vary, but enormous growth seems likely. The current neuromorphic computing market is majorly driven by increasing demand for AI and brain chips to be used in cognitive and brain robots. These robots can respond like a human brain.
Numerous advanced embedded system providers are developing these brain chips with the help of AI and machine learning (ML) that acts as thinks and responds as the human brain.
This increased demand for neuromorphic chips and software for signal, data, and image processing in automotive, electronics, and robotics verticals is projected to further fuel the market.
The need for potential use cases such as video analysis through machine vision and voice identification has also been projected to aid market growth. Major players for the development include Intel, Samsung, IBM and Qualcomm.
Researchers are still trying to find out where practical neuromorphic computing should go first; vision and speech recognition are the most likely candidates. Autonomous vehicles could also benefit from such human-like learning without human-like distraction or cognitive errors.
BrainChip’s Akida architecture features event-based architecture. It supports on-chip training and inference and various sensor inputs such as vision, audio, olfactory, and innovative transducer applications.
Akida is already featured in a unique product: the Mercedes EQXX concept car, displayed at the CES this year, where it was used for voice control to reduce power consumption by up to 10x. Internet of Things (IoT) and opportunities for Edge range from the factory floor to the battlefield.
Neuromorphic computing will not be directly replacing the modern CPUs and GPUs. Instead, the two types of computing approaches will be complementary, each suited for its sorts of algorithms and applications.

The Potential Underneath​

Neuromorphic computing came to existence due to the pursuit of using analogue circuits to mimic the synaptic structures found in brains.
Our brain excels at picking out patterns from noise and learning. A neuromorphic edge CPU excels at processing discrete, transparent data. For the same reason, many believe neuromorphic computing can help unlock unknown applications and solve large-scale problems that have put conventional computing systems in trouble for decades. Neuromorphic processors aim to provide vastly more power-efficient operations by modelling the core workings of the brain.
In 2011, HRL announced that it had demonstrated its first “memristor” array, a form of non-volatile memory storage that could be actively applied to neuromorphic computing. Two years later, HRL’s first neuromorphic chip, “Surfrider” was released.
As reported by the MIT Technology Review, Surfrider featured 576 neurons and functions on just 50 mW of power. Researchers tested the built chip by adding it into a sub-100-gram drone aircraft loaded with several optical, infrared, and ultrasound sensors and sent the drone into three rooms.
The drone was observed to have “learned” the entire layout and objects present in the first room through sensory input. Later, using this teaching, it could “learn on the fly”, even if it was in a new room or could recognise having been in the same room before.

Image Source: MIT
Today, most neuromorphic computing work is incorporated by using deep learning algorithms that perform processing on CPUs, GPUs, and FPGAs. None of these is optimised for neuromorphic processing. However, next-gen chips such as Intel’s Loihi were designed exactly for these tasks and can achieve similar results on a far smaller energy profile. This efficiency will prove critical for the coming generation of small devices needing AI capabilities.
Deep learning feed-forward neural networks (DNNs) underperform neuromorphic solutions like Loihi. DNNs are linear, with data moving from input to output straight. Recurrent neural networks (RNNs) are more similar to the working of a brain, using feedback loops and exhibiting more dynamic behaviour, and RNN workloads are where chips like Loihi shine.
Samsung also announced that it would expand its neuromorphic processing unit (NPU) division by 10x, growing from 200 employees to 2000 by 2030. Samsung said at the time that it expected the neuromorphic chip market to grow by 52 per cent annually through 2023.
One of the future challenges in the neuromorphic space will be defining standard workloads and methodologies for benchmarking and analysis. Benchmarking analysis applications such as 3DMark and SPECint have played a critical role to understand the technology, aiding adopters match products to their needs.
Currently, Neuromorphic computing remains deep in the R&D stage. There are virtually only a few substantial commercial offerings in the field. Still, it’s becoming clear whether specific applications are well-suited to neuromorphic computing or not. Neuromorphic processors will be faster and more power-efficient for extensive workloads than any modern, conventional alternatives.
CPU and GPU computing, on the other hand, will not be disappearing due to such developments; neuromorphic computing will be beside them to handle challenging roles better, faster, and more efficiently than anything we have seen before.