WBT Discussion 2022

[Pappagallo]

Yes, I am used to these peaks from the ASX by now. The main thing is that it finds a new level this time. But I'm sure it will. The world is waiting for this next logical step in development I think.
Do you think the qualification process plays an important role in the negotiations? I think I heard that they want to be through the entire qualification programme by the end of the first half of 2023 at the latest.

Weebit have already qualified the technology themselves in conjunction with their partner CEA-Leti. I’d imagine now having physical functional wafers to demonstrate to potential fabs/customers would really help in accelerating discussions. There is also a seperate qualification process happening at Skywater in their fab and this will be completed early in 2023.

Based on the presentations I’ve watched I’m hoping for a Tier 1 fab to be signed by the end of the financial year (June 2023) and would subsequently expect WBT to enter the ASX200 at a minimum.

 

[alwaysgreen]

What a day Wee-Bitches! (is that inappropriate?).

I genuinely believe we still have 10 bags in us from here.

 

[alwaysgreen]

Announcement tomorrow? Leaky ship?

 

[Pappagallo]

What a day Wee-Bitches! (is that inappropriate?).

I genuinely believe we still have 10 bags in us from here.

10 bags easily IMO. That would equate to a MC of only $6B which is very realistic if we partner with multiple tier one fabs and form lots of those oft-mentioned triangles with their customers.

 

[cosors]

None of my three (...) leaves me calmer. I suspect that Weebit will develop excellently in 23. There is not much to report here either. It will just happen. And that turns me on

 

[cosors]

"CEA-Leti Presents RRAM’s ‘Promising Advantages’ For Neuromorphic/In-Memory Computing at IEDM 2022​

CEA-Leti

A CEA-Leti tutorial presented at IEDM 2022 highlighted promising advantages that resistive random-access memory (RRAM) technologies hold for implementing novel neuromorphic/in-memory computing systems for massively parallel, low-power and low-latency computation.
Published on 7 December 2022

In a presentation titled "Resistive Memories-Based Concepts for Neuromorphic Computing", Elisa Vianello, CEA-Leti's edge AI program manager, said RRAMs, aka memristors, offer advantages in energy efficiency and computing power when processing AI workloads. She noted, however, scientists must overcome device issues, especially variability, quantization error and limited endurance to achieve commercialization of this approach.

During the conference, CEA-Leti also reported development of the first end-to-end, gesture-recognition solution for ultralow power implementation on silicon with an estimated always-on total power consumption of 0.41 μJ/frame. This breakthrough, presented in the paper "Spike-based Beamforming Using pMUT Arrays for Ultra-Low Power Gesture Recognition", used low-power piezoelectric micromachined ultrasonic transducers (pMUTs) to emit and sense ultrasonic signals. This novel spike-based beamforming extracts spatial temporal information and a spiking recurrent neural network (SRNN) to perform simple gesture detection and classification.

Neuromorphic In-Memory Computing

In recent years, "neuromorphic" has been used to describe mixed-signal and pure digital systems that can be used to simulate spiking neural networks. As interest in the potential for this technology grew, the neuromorphic researchers were joined by material and device-physics researchers to study memristor properties and leverage their physics to implement neural and synaptic functions.

Meanwhile, artificial intelligence (AI) algorithms were being applied in healthcare, robotics, agriculture and other sectors, but those applications face power constraints. To address some of these challenges, CEA-Leti's AI research focuses on the development of novel brain-inspired technologies and processing methods. This requires researchers in multiple disciplines to combine their efforts and simultaneously co-develop technologies, circuits, processing methods and the supporting computing architectures, Vianello said.



"Spike-based Beamforming Using pMUT Arrays for Ultra-Low Power Gesture Recognition"

Explaining CEA-Leti's breakthrough end-to-end, ultralow power gesture-recognition solution, Emmanuel Hardy, lead author on the paper, said previously published systems in the literature were implemented with off-the-shelf sensors and readout electronics, so gesture recognition is always performed offline in software with full precision for the inference.

Traditional beamforming directly combines the sine waves in analog or digital format after applying delays. CEA-Leti's spike-based technique simplifies the process by encoding the phase of a signal by a single spike per signal period. It then allows scientists to apply simple logic on spikes to implement beamforming. A spiking recurrent neural network (SRNN) takes the spike density as an input to perform gesture detection.

CEA-Leti's end-to-end, gesture-recognition solution is suitable for ultralow power implementation on silicon with an estimated total power consumption of 0.41 μJ/frame. The breakthrough uses low-power sensors with pMUTs and extracts and processes the minimum information with CEA-Leti's novel spike-based beamforming. It also includes classification in the spike domain with a SRNN. The institute also is working to develop an energy-efficient RRAM-based SRNN.

"Our system supports a fully integrated approach enabling ultralow-power, end-to-end operation," Hardy said. "Its primary advantage is its low manufacturing cost and easy integration, which suits its use in wearable and automotive applications. The system could be also used for acoustic ranging with interesting applications in robotics and drones.""
https://www.leti-cea.com/cea-tech/leti/english/Pages/What's-On/Press%20release/CEA-Leti-Presents-RRAM%E2%80%99s-%E2%80%98Promising-Advantages%E2%80%99-For-NeuromorphicIn-Memory-Computing-at-IEDM-2022.aspx


- Does this have anything to do with Weebit?
- And do I have it right that there are actually only two RRAM suppliers, TSMC and Weebit?


And there is already a RRAM deal:

"Infineon’s next generation Aurix microcontrollers will make use of resistive RAM (RRAM) embedded non-volatile memory instead of embedded flash.​

Foundry Taiwan Semiconductor Manufacturing Co. Ltd. (Hsinchu, Taiwan) is offering embedded RRAM as an option on its 28nm CMOS manufacturing process and Infineon said it will make use of this with samples available to customers before the end of 2023.
And a deal has already been struck with ReRAM:
..."
https://www.eenewseurope.com/en/infineons-microcontrollers-to-go-rram-with-tsmc/

 

[Slymeat]

.

…

- Does this have anything to do with Weebit?
- And do I have it right that there are actually only two ReRAM suppliers, TSMC and Weebit?

…

IMHO you are right mate.

CEA-Leti are a development partner of Weebit Nano, so it doesn’t take too much of a leap of faith to work out whose ReRAM they are using.

There are MANY companies trying to produce ReRAM (and why not? There is huge value in it.) but as you stated, as far as I understand it, Weebit is the only viable supplier to TSMC.

Fujitsu Semiconductor also has an offering, but I am not certain whether or not that is a mere repackaging of the TSMC product (as I think is the case with Infineon) or if they have developed it themselves.

Anyway—the NVM market is huge! There is room for several players. And many are talking of Weebit as a viable alternative to TSMC, so anyone wanting to stick it to the big player will come to Weebit.

And on top of that we have the Weebit advantage that they have just taped-out a 22nm FO-SOI implementation and sent it to the foundary - as reported in the Wall Street Jounal (getting huge exposure). A 22nm implementation takes it into the realm where Flash simply can’t go.

”Weebit Nano tapes-out first 22nm demo chip​

Jan. 2, 2023 5:29 pm ET

Weebit ReRAM in 22nm FD-SOI process provides cost-effective, reliable embedded NVM for IoT, edge AI and other ultra-low power applications

HOD HASHARON, Israel, Jan. 02, 2023 (GLOBE NEWSWIRE) -- Weebit Nano Limited (ASX:WBT), a leading developer of next-generation memory technologies for the global semiconductor industry, has taped-out (released to manufacturing) demonstration chips integrating its embedded Resistive Random-Access Memory (ReRAM) module in an advanced 22nm FD-SOI (fully depleted silicon on insulator) process technology. This is the first tape-out of Weebit ReRAM in 22nm, one of the industry's most common process nodes, and a geometry where embedded flash is not viable.

Weebit worked with its development partners CEA-Leti and CEA-List to successfully scale its ReRAM technology down to 22nm. The teams designed a full IP memory module that integrates a multi-megabit ReRAM block targeting the 22nm FD-SOI process which is designed to deliver outstanding performance for connected and ultra-low power applications such as IoT and edge AI.

As embedded flash is unable to scale below 28nm, new non-volatile memory (NVM) technology is needed for smaller process geometries. Weebit ReRAM in 22nm FD-SOI offers a low-power, cost-effective embedded NVM solution which can withstand harsh environmental conditions.”

 

[Slymeat]

Weebit have already qualified the technology themselves in conjunction with their partner CEA-Leti. I’d imagine now having physical functional wafers to demonstrate to potential fabs/customers would really help in accelerating discussions. There is also a seperate qualification process happening at Skywater in their fab and this will be completed early in 2023.

Based on the presentations I’ve watched I’m hoping for a Tier 1 fab to be signed by the end of the financial year (June 2023) and would subsequently expect WBT to enter the ASX200 at a minimum.

You called that correctly @Pappagallo. Oh how I am looking forward to what 2023 can bring.

In Coby Hanoch’s New Years post he did in fact state that they are on the verge of signing a Tier-1 foundary. And it seems to be imminent - so may be before June 2023. I have emboldened the relevant sentence below.

Coby’s New Year message on LinkedIn:

”2022 is almost over - what an amazing year this was for Weebit Nano Ltd! A year ago when I presented ReRAM to potential partners and customers the reaction was normally “great technology… but it is a future technology...”. Today nobody says that!

During the past year we qualified our wafers and demonstrated the technology is strong and ready for mass production. We transferred our technology to our partners at SkyWater Technology Foundry, received the wafers manufactured there, and are now qualifying them. Customers are now calling on the big foundries and asking for ReRAM, and several of the top foundries are already evaluating our technology. And we will very soon be taping out the 22nm FD-SOI demo chip.

These are very exciting times for Weebit, and 2023 is shaping up to be even more exciting than 2022! We expect to sign up at least 1 tier-1 fab, sign up customers, and continue improving the technology, qualifying at more extreme levels and on more platforms.”

 

[cosors]

.

IMHO you are right mate.

CEA-Leti are a development partner of Weebit Nano, so it doesn’t take too much of a leap of faith to work out whose ReRAM they are using.

There are MANY companies trying to produce ReRAM (and why not? There is huge value in it.) but as you stated, as far as I understand it, Weebit is the only viable supplier to TSMC.

Fujitsu Semiconductor also has an offering, but I am not certain whether or not that is a mere repackaging of the TSMC product (as I think is the case with Infineon) or if they have developed it themselves.

Anyway—the NVM market is huge! There is room for several players. And many are talking of Weebit as a viable alternative to TSMC, so anyone wanting to stick it to the big player will come to Weebit.

And on top of that we have the Weebit advantage that they have just taped-out a 22nm FO-SOI implementation and sent it to the foundary - as reported in the Wall Street Jounal (getting huge exposure). A 22nm implementation takes it into the realm where Flash simply can’t go.

”Weebit Nano tapes-out first 22nm demo chip​

Jan. 2, 2023 5:29 pm ET

Weebit ReRAM in 22nm FD-SOI process provides cost-effective, reliable embedded NVM for IoT, edge AI and other ultra-low power applications

HOD HASHARON, Israel, Jan. 02, 2023 (GLOBE NEWSWIRE) -- Weebit Nano Limited (ASX:WBT), a leading developer of next-generation memory technologies for the global semiconductor industry, has taped-out (released to manufacturing) demonstration chips integrating its embedded Resistive Random-Access Memory (ReRAM) module in an advanced 22nm FD-SOI (fully depleted silicon on insulator) process technology. This is the first tape-out of Weebit ReRAM in 22nm, one of the industry's most common process nodes, and a geometry where embedded flash is not viable.

Weebit worked with its development partners CEA-Leti and CEA-List to successfully scale its ReRAM technology down to 22nm. The teams designed a full IP memory module that integrates a multi-megabit ReRAM block targeting the 22nm FD-SOI process which is designed to deliver outstanding performance for connected and ultra-low power applications such as IoT and edge AI.

As embedded flash is unable to scale below 28nm, new non-volatile memory (NVM) technology is needed for smaller process geometries. Weebit ReRAM in 22nm FD-SOI offers a low-power, cost-effective embedded NVM solution which can withstand harsh environmental conditions.”

By the way, it came to my attention in the Indian press.

 

[Slymeat]

By the way, it came to my attention in the Indian press.

The whole world will know soon!

 

[Mannic]

WBT on a great run since a low of Sept last year. Up 100% in four months whoa hoo!!!

 

[cosors]

View attachment 27216
WBT on a great run since a low of Sept last year. Up 100% in four months whoa hoo!!!

I also find this picture beautiful

 

[cosors]

 

[Mannic]

WBT shares this week are as HOT as Perth’s weather. Cocktails by the pool in celebration!!

 

[cosors]

Why Has Weebit Nano Soared +57% In 1 Month...?​

 

[cosors]

Not necessarily. I have quite a few BRN and before today, I feel like I had about 55,000 "spare" BRN shares. I was waiting for a day where BRN had a good green day to sell and buy more WBT.

I am equally bullish on both. I guess I am starting to feel there is more upside potential in WBT but who knows. I'd love for both to end up with $30 billion market caps one day.

@alwaysgreen

I am equally bullish on both. I guess I am starting to feel there is more upside potential in WBT but who knows.

How right you are

 

[Slymeat]

@alwaysgreen

How right you are

As with a lot of things, I feel we are on the same page. Weebit and BrainChip are two of my stocks that I am also the most bullish on.

I think things will happen faster (as in share price appreciation and general adoption of the technology) with Weebit in the immediate future than for BrainChip. It’s a simple fact that EVERYONE knows what RAM is and what it is used for, but spiking neuromorphic computing at the edge is something many are failing to grasp.

ReRAM has advantages over FLASH in EVERY measurable metric, it’s only obstacle to replacing ALL FLASH (in the immediate future) is the shear abundance of FLASH out there. But when we talk embedded memory, there is no such obstacle, with ReRAM’s support for sub 40nm being a real kicker.

If the current 22nm tapeout is successful I expect that will be an inflexion point for Weebit. And as with Akida, Weebit’s ReRAM is scalable below that architecture, but they first need to prove to the world that 22nm is achievable. So many people need to see in order to believe. And with the amount of money involved, I can accept that.

 

[alwaysgreen]

As with a lot of things, I feel we are on the same page. Weebit and BrainChip are two of my stocks that I am also the most bullish on.

I think things will happen faster (as in share price appreciation and general adoption of the technology) with Weebit in the immediate future than for BrainChip. It’s a simple fact that EVERYONE knows what RAM is and what it is used for, but spiking neuromorphic computing at the edge is something many are failing to grasp.

ReRAM has advantages over FLASH in EVERY measurable metric, it’s only obstacle to replacing ALL FLASH (in the immediate future) is the shear abundance of FLASH out there. But when we talk embedded memory, there is no such obstacle, with ReRAM’s support for sub 40nm being a real kicker.

If the current 22nm tapeout is successful I expect that will be an inflexion point for Weebit. And as with Akida, Weebit’s ReRAM is scalable below that architecture, but they first need to prove to the world that 22nm is achievable. So many people need to see in order to believe. And with the amount of money involved, I can accept that.

Totally agree mate.

Brainchip is now selling and my concern is that the adoption rate has been slow (as you said, people need to even grasp the concept) but my biggest fear is that nobody wants our tech. Each passing 4C with bugger all revenue raises my anxiety levels. I would never write this in the BRN forum because I would just get hammered but I believe it is a valid concern. Why are we not selling licenses? The optimist in me believes that the LDA call might have been to expedite the newer versions of Akida because companies showed interest but needed it to be upgraded. Fingers crossed.

I also agree that Weebit will be an easy sell. Swapping out one old, slow memory for a memory on steroids is a no brainer. The only concern I have is competition (TSMC has already locked up the iphone with their RERAM). I understand ours is superior but so was BETAMAX. Marketing and sales can often give a greater market share to an infrior product. I have little doubt though that Weebit will be successful. Even if we only end up in 3% of products, we will have a great company.

Either way, I'm strapped in for the ride with both!

 

[cosors]

Neuromorphic Materials, Devices, Circuits and Systems (NeuMatDeCaS) that will take took place from the 23rd to the 25th of January 2023

And the first topic was:

"Topics to be covered by the conference:
Materials & Devices: ReRAM, PCM, Ferroelectric, Spintronics, Organic, OECTs, Halide Perovskites, 2D transition metal dichalcogenides, Photonics.Tutorial session: Computational Neuroscience
Circuits, Systems & Algorithms: In-sensory computing, In-memory computing, Deep neural networks (DNNs), Spiking neural networks (SNNs).

The human brain performs complex cognitive tasks with a mere 20W of power, and hence serves as an inspiration for the next generation of low-power computer systems. The co-location of computation and memory in the brain motivates for non-Von Neumann systems, subverting the memory wall observed in conventional computers. Moreover, the structure of the brain as an interconnected network of neurons and synapses as well as its operation give rise to many algorithmic advancements including deep neural networks (DNNs) and further bio-inspired spiking neural networks (SNNs). Neuromorphic computing opens up many opportunities in material discovery, device engineering, circuit design and algorithm development to tailor systems that can learn from unstructured data. The goal of this conference is to provide a forum for discussing interdisciplinary research in brain-inspired computing, with an emphasis on emerging understanding of synaptic and neuronal processes in devices and systems. The scope of this conference will cover:

(1) Employing emerging memristive and memtransistive materials and novel device physics to create neuromorphic systems and in-memory computing applications,

(2) Exploration of the principles of learning such as Hebbian learning, synaptic competition, winner-take-all mechanism, associative learning, and

(3) Investigation of computational frameworks for supervised learning, unsupervised learning, reinforcement learning, reservoir computing, one-shot learning and beyond."
https://www.nanoge.org/NeuMatDeCaS/home

 

[cosors]

"Weebit looks forward to first revenue in 2023​

News | January 31, 2023
By Peter Clarke

Coby Hanoch, CEO of resistive RAM pioneer Weebit Nano Ltd. (Hod Hasharon, Israel), has said he expects the company to make its initial sales in 2023.​

In a quarterly round up of progress at the listed startup for 2QFY23 Hanoch said: “Weebit Nano achieved key milestones in its commercialisation roadmap during the period, receiving the first silicon wafers with its embedded ReRAM from a commercial fab, commencing qualification at SkyWater, taping-out its first 22nm demo chip, ramping up discussions and evaluations with Tier-1 foundries and potential customers, and progressing longer-term development activities for the discrete memory market. Our progress in the first half of the financial year means we are well-placed to productise our embedded ReRAM technology and secure first revenues in 2023.”
He added that technology qualification is now underway with Skywater Technology on its 130nm manufacturing process and should be completed in the 1H23 enabling mass production in 2H23.

Hanoch added that TSMC’s offer of ReRAM as an alternative non-volatile memory has changed perceptions about the technology (see Infineon’s microcontrollers to go RRAM with TSMC).

“Demand for ReRAM is growing rapidly with many Tier-1 foundries now looking to add ReRAM as an alternative NVM to cater to customer requests. The use of ReRAM from the world’s leading foundry in major consumer products has been instrumental in changing industry perception of ReRAM as a future technology to one that is available right now,” Hanoch said.
Hanoch said Weebit is now engaged with multiple world-leading wafer fabs.

Meanwhile Weebit and Leti have started a qualification process for higher temperatures and for higher cycling endurance.

In addition, Hanoch said, Weebit and Leti continue to make progress on a ReRAM selector device to enable high-capacity arrays. While the initial focus of Weebit’s ReRAM selector was for discrete chips, the recent progress makes it also suitable for future embedded applications, such as edge AI and automotive, Hanoch said.
https://www.eenewseurope.com/en/weebit-looks-forward-to-first-revenue-in-2023/