Advantages of using Silicon Oxide - Amazon Web Services



Weebit NanoWeebit Nano is a leader in the development of next generation memory technology and plans to become a key player in this market.Weebit addresses the growing need for data storage and embedded Non-Volatile Memory (NVM) technology with its Resistive Random-Access-Memory (ReRAM) technology. A unique advantage of Weebit’s ReRAM is that its switching element is Silicon Oxide.Weebit is currently seeking to deploy its next generation ReRAM technology into the global semiconductor market in two key NVM segments: Embedded – replace existing embedded flash (eFlash) technologies, with a market estimated to be worth US$23bn. Targeting first demonstration to customer by end of Q1 2021. Targeting first customer agreements within 3-4 months of demonstrationExternal (discrete) – single chip replacing NOR and NAND flash technologies and market estimated to be worth US$60bn. Targeting integrated ReRAM + selector to be demonstrated by Q3 2021.The current key priorities for Weebit are to:Continue the development of a memory module for a potential South Korean customer. The memory module work is a critical input for the production process and the work being done for the South Korean partner will largely be applicable for any further customer developments.Production fab technology transfer. This is planned to be completed by December 2020.Define the specific details of the selector work with Leti for the discrete memory market. COVID updateWeebit has been unlucky with the onset of the recent Covid 19 pandemic. It has led to unforeseen delays in progressing its 3 partnership deals, with a Korean Tier 2 company, Chinese integrated device manufacturer SiEn, and Chinese memory company XTX. The uncontrollable isolation and travel impediment delays are hopefully nearly in the past as fabricator facilities get back up and running, with both partners looking to make up lost time:?SiEn PartnershipThe fastest path to productization will be the Embedded Non Volatile Memory (NVM) market opportunity with SiEn and with the Korean partner.? It is expected that productization via a license deal will see a potential significant rerating of the share price as will updates to the market on progress which are expected in the next couple of quarters. Weebit’s very reputable leadership indicate that this product line has had the majority of its technical development risk completed and it is now a sprint to commercialisation.XTX Partnership?The partnership is aiming to integrate ReRAM into the XTX product suite to replace NOR Flash (a massive multi-Billion dollar market). Successful testing was completed by XTX 4 months ahead of schedule in December 2019. The only outstanding development item is the integration of a ‘selector’, expected to be demonstrated in Q3 2021. Leti, Weebit’s partner in France, has been designing and building a selector solution for the last 5 years. As with SiEn, updates of progress and the integration of a selector should see potential significant share price appreciation.Contents TOC \o "1-3" \h \z \u Advantages of using Silicon Oxide PAGEREF _Toc33981813 \h 3World class semiconductor team PAGEREF _Toc33981814 \h 5Weebit is currently targeting commercial agreements in the embedded memory market PAGEREF _Toc33981815 \h 6Weebit has achieved industry-leading technical parameters PAGEREF _Toc33981816 \h 6Risk level is now low PAGEREF _Toc33981817 \h 7The Mb array is not the product PAGEREF _Toc33981818 \h 7The right partner is important PAGEREF _Toc33981819 \h 7Tier two Korean partner PAGEREF _Toc33981820 \h 8Working with Weebit at its current stage of development is a massive endorsement PAGEREF _Toc33981821 \h 8The customer module will be generic PAGEREF _Toc33981822 \h 8Kitec PAGEREF _Toc33981823 \h 9Weebit is planning to accelerate its entry into the discrete, stand-alone, memory chip market PAGEREF _Toc33981824 \h 10XTX PAGEREF _Toc33981825 \h 11Reason for current focus on the China market PAGEREF _Toc33981826 \h 11Weebit already have a selector in the embedded domain. They need to develop a selector for the discrete memory domain. PAGEREF _Toc33981827 \h 12Why Weebit has accelerated the selector/discrete memory work PAGEREF _Toc33981828 \h 13Jackson Lam PAGEREF _Toc33981829 \h 13SiEn PAGEREF _Toc33981830 \h 14Silvaco PAGEREF _Toc33981831 \h 15Other activities PAGEREF _Toc33981832 \h 16300mm/28nm activity moved to a lower priority PAGEREF _Toc33981833 \h 16Transferring the technology from Leti which is a development facility to a production facility PAGEREF _Toc33981834 \h 16How commercial agreements work PAGEREF _Toc33981835 \h 17It is licence fees per customer project PAGEREF _Toc33981836 \h 17Exclusivity deals are not being considered PAGEREF _Toc33981837 \h 18Weebit has a three phase approach to the market PAGEREF _Toc33981838 \h 18Consolidation adjusted top 20 holders movement PAGEREF _Toc33981839 \h 19Advantages of using Silicon OxideThe semiconductor memory industry is highly capital-intensive. Setting up a new start-of-the-art semi conductor fabrication plant (fab) can cost billions of dollars and take 1 to 2 years to complete. Additionally, to stay competitive, fabs often need to retool every 2 to 3 years which again involves significant costs. Due to the costs and delivery times involved, few companies build their own fabs, for example, even Apple doesn’t have their own fab. The limited number of fabs means that it can be quite difficult and expensive for companies to get time in the fabs.A major thing that gets in the way of emerging memories taking over the market today is that they use new materials. Anything that uses a new material is something that is difficult to get past in manufacturing. Weebit is unique in the emerging memory market because its switching material is silicon oxide (SiOx). SiOx ReRAM has a range of competitive advantages including the technology: being easy to manufacture, having a simple structure and being cheap to produce.Being easy to manufactureInside fabs, the majority of the hardware that is used to produce chips and memory is built with silicon oxide. The fact that Weebit is using silicon oxide means that companies who partner with them don’t need to spend an enormous amount of capex on machines that can produce their chips. This is a huge advantage for Weebit and is a reason why they are going to have the ability to talk to and negotiate with the big players, when they are ready.Using standard materials (silicon oxide) will also allow Weebit to move relatively quickly from one fab to the next. When a company uses non-standard materials, getting into each of the fabs and ramping up production in the fab is a big challenge, due to the requirement for new tools and the optimization of those tools. Another significant advantage to using standard materials is that it should make it easier to achieve high yields (working parts).Having a simple structureWeebit’s ReRAM has a very simple structure which has allowed them to develop their technology much faster and with an order of magnitude lower cost than the majority of other similar companies. Weebit has quickly reached a point where they have produced a chip and are talking with and working with potential customers and partners. In May 2019, WBT announced that its SiOx ReRAM technology had achieved technical parameters that are in line with today’s market requirements, i.e. endurance and retention levels equivalent to or better than in today’s commercially sold NVM products. Simplicity is also the key to being cheap and creating products that partners can adapt to their use cases quickly. Weebit has confirmed it is possible to produce their ReRAM technology as an embedded memory with just a single added mask. If you look at how they manufacture flash today or MRAM, they have a lot masks and it is extremely complicated to manufacture.Being cheap to produceWeebit’s silicon oxide based memory will be cheap to produce. This is an extremely important advantage to using standard materials and having a simple structure. You can have the best technology in the world, but if it is too expensive to manufacture then people won’t use it. Weebit realizes that you need to have great technology, but it is also very critical for that technology to be easy to manufacture and have a low cost to manufacture. World class semiconductor teamWeebit has a world class management team. There are not many, if any, other companies on the ASX that have a management team with the same quality as Weebit. People like Atiq Raza could be making a lot more money by being in other companies, for example, in America Atiq could be getting over $100,000 a year for just being on the board. Since he has joined Weebit when he could join pretty much any other company in the semiconductor industry it shows that he believes in the goals and vision that Weebit is pursuing. If you don’t know much about the semiconductor industry, look at the people who do and the backing that they are giving Weebit.Weebit is now focused on reaching commericialisation and has some huge names on its board of directors including, but not limited to:David (Dadi) Perlmutter - there isn’t any Australian company or only a rare few that has had an executive chairman with the reputation and wealth of experience that David Perlmutter has. David was one of the founders of Intel’s microprocessor and was ex number two at Intel worldwide. It is really incredible that Weebit has been able to secure someone with David’s unbelievable background and experience. Atiq Raza - Atiq is another huge name in the industry and having a person of the quality of Atiq Raza on Weebit’s board is invaluable. It is similar to having someone of the quality of David Perlmutter as chairman. Atiq has an enormous amount of experience in the industry in both the US and worldwide. Dr Yoav Nissan-Cohen - Yoav received his PhD researching non-volatile memories under the supervision of Prof Dov Frohman the inventor of the first non-volatile memory technology. Yoav founded Tower Semiconductor, one of the top fabs worldwide and his career covers nearly 40 years of scientific research, technology development and executive management in the hi-tech industry.Weebit is currently targeting commercial agreements in the embedded memory marketMemory is at the center of the world today. It is a critical element in computing systems and the demand for memory is growing exponentially in a wide range of different domains from drones to autonomous vehicles to IOT and security. These domains are all now starting to demand something more out of the current memory technologies. For example, if you look at security with facial recognition, finger print recognition and all the different kinds of keys that are now available you want this data to be handled in the most secure way possible. Today, there isn’t an embedded non-volatile solution at the smaller geometries, so a lot of this data in kept in external memory. The bus that connects to this memory is vulnerable and this causes big issues today. Another example is if you look at autonomous vehicles which have hundreds of non volatile memories in them and since they need to operate in the engine they need to be able to operate at high temperatures. Weebit’s technology has achieved retention at 150 degrees which is a critical element in this domain.The expected growth in the embedded emerging non-volatile memory market is dramatic and there is a real demand for embedded non volatile memories that Flash just cannot answer. There are great difficulties scaling flash and the market is constantly moving to the smaller geometries. There are also great demands for energy efficiency which flash cannot answer. With IoT all the devices, for example cameras, sensors and wearables, run on batteries or even just ambient energy, e.g. sunlight. Therefore, low power consumption is a critical requirement. ReRAM operates with much lower power than flash along with having many other additional advantages. The market for embedded ReRAM, is expected to show very strong, long-term growth as it starts to complement some of today’s mainstream memory technologies, such as NOR Flash and SRAM. There is even potential to replace NOR Flash in certain applications. Weebit is currently targeting the embedded memory market as it presents many opportunities for emerging memories and is the quickest way to revenue for Weebit. Weebit has achieved industry-leading technical parametersAs a company, Weebit has made great technological progress and are now at the point where they have achieved industry-leading technical parameters and are now engaging with and collaborating with potential customers. For example:In June 2018, they produced a 40 nm 1Mbit Array. In October 2018, they reached market level retentionIn January 2019, they reached market level endurance and programming voltage levels competitive with production nonvolatile memoriesIn May 2019, they reached retention levels of over 10 years at 130-150°C which is at the forefront of ReRAM technologies. They also demonstrated endurance of a million cycles for single cells.In November 2019, they confirmed it is possible to produce Weebit’s ReRAM technology as an embedded memory using two, or possibly only one, additional mask, compared to the seven to 10 additional masks normally required for current Flash memory. In December 2019, Weebit Nano and Leti filed three new patents for improved yield and reliability of advanced ReRAM products. It was also announced that testing at XTX facilities successfully reproduced the technical parameters achieved on Weebit’s ReRAM technology at Leti.Risk level is now lowWith startups, people should always be thinking about the risk level. Before Weebit got to the kilobit array there was a huge risk level because there were technological issues that could have put a halt to everything. Once Weebit got to the kilobit and megabit array as well as the retention and endurance levels they are at now, the risk level has dropped significantly. There is still a lot of work to be done and there will be obstacles along the way, but the risk of Weebit not being able to overcome these obstacles is extremely low. The obstacles that are left are the kind of obstacles where you hit them, you understand them and then you find a way to get around them. They are not show stoppers that could kill everything. That is a very important thing to understand about where Weebit is. The risk level has dropped dramatically.The Mb array is not the productWeebit produced a 40 nm 1Mbit array back in June 2018. This array has been given to partners to work with and test with, but the eventual product that will be sold is more than just the packaged array. In the embedded domain, the product is when the array is embedded inside a system on a chip (SoC) which has a processor, sensors, etc. This requires a memory module which will act as the interface between the memory and the SoC. In the discrete memory domain, the product is the memory along with a selector that is appropriate for the requirements in this domain.The right partner is importantWhen?you have very interesting technology, there are many people who are interested. A good question to ask is: who is the best partner to start with. An Intel or a Samsung would be interested in working with Weebit, but the working relationship would not be in Weebit’s favour. If you work with a tier one company, they?are going to come to you?with a lot of demands and requests. They would?suck up all of your resources and they won’t actually be paying for anything until you have met all of their demands. Very often this leads companies to the point?where they start running out of money, but since they are so focused on the tier one project they have no room to maneuver to anything else. A company like this would either go bankrupt or be forced to take a ridiculous takeover?offer from the tier one company. It would be a strategic mistake for Weebit?to work with a tier one company as their first customer.Finding the right tier two company which is the right size and has a real need for Weebit’s technology is critical to the success of the company. The two companies that Weebit are currently working with seem like ideal partners for Weebit in its current stage of development. In parallel, Weebit, would of course, also be progressing discussions with other potential customers in other markets.Tier two Korean partnerThe tier two Korean company is a great partner for Weebit. They are hungry to move forward and are willing to put in the resources and effort to work with Weebit. They are also a great partner because: ?They are a big and well known company. Even though they are tier two, they are still big, if their name was?mentioned people in the semiconductor industry would recognize it. They have their own fab.?This is important because when you are doing an embedded deal there are three players. There is the IP provider which is Weebit. There is the customer that wants to embed the memory in their chip and then there is the fab where you need to manufacture. So, you need to get all three parties aligned. Because they own their own fab Weebit only has?one party to deal with which is a massive advantage.They?are totally focused on 200mm wafers which is what Weebit is doing today. This means that Weebit can get to a commercial agreement with them much faster as they don’t need to go and get everything working on 300mm.?They are based in South Korea. Korea is the world’s largest memory chip manufacturer, producing approximately 57 per cent of the global supply and is a key target for Weebit. It is also the second largest semiconductor industry in the world, accounting for more than 16 per cent of the global market.Working with Weebit at its current stage of development is a massive endorsementIn the semiconductor space, like with almost any other domain showing your product works in production is a key challenge. Before you can point to that first customer everyone shies away and doesn’t want to be the “guinea pig”. When the South Korean potential customer agreed to engage with Weebit, it was a massive step forward for Weebit. It meant that Weebit had found a company that was willing to invest their engineering time and internal data with them. This helps Weebit immensely in moving towards their first product release. While Weebit is not receiving any direct money, the South Korean company is still investing significant time and resources. The first module being developed by Weebit is based on their specification and can be considered a proof-of-concept, after which Weebit would then engage with them on a real project and negotiate payment terms.The first memory module is planned to be delivered in late 2020. The delivery of this memory module will be a gargantuan announcement for Weebit. This is because it will lead to Weebit’s first deal and the first deal in any business is always the hardest. Once the first deal is made, it is significantly easier to make the deals that follow. The first deal is valuable because it proves that the technology can be commercialized, it is almost irrelevant whether it is for $100,000 or $500,000 or $5,000,000.The customer module will be genericOne of Weebit’s priorities with the memory module it is developing is to make it?as generic as possible. While these modules are based on the provided?specifications, it is not a good idea to be tied to them. It is a risk that at a certain point in time the Korean company might decide that they no longer?want to work with Weebit. While this is unlikely it is still a possibility, therefore, Weebit is being careful so that if this was to happen they are not left with nothing, but wasted work.The memory?module is being built?in a way so?that it?is parameterized and very flexible. The goal is to have 95% of this work being reusable for other customers. So once one module is working, when somebody else gives Weebit their specification for the memory module they don’t need to?start from scratch. They have a base product and just need to do an adaptation. This means that the memory module for the?second, third and fourth customer will be completed much?faster. By the time the?fifth or sixth memory module is completed, the goal is to have?developed a memory compiler which you would give several parameters and it would know how to take those and build the module automatically. Creating a generic and modular?memory module does take more time, but it allows a significant reduction in the rework required to create other memory modules.The first module is by far the longest and takes the largest effort, but once it is done adapting it to fit another customer is?fairly easy. This is because it is going to be 95% similar to the next customer since everything is going to be parameterized and simple. The first memory module?also takes the longest because it is the one in which you are?the most likely to hit pitfalls, but once you hit them, fix them and get over them you don’t hit them again as you learn from them.?So with the first memory module, you hit most of the pitfalls and the second one you might hit something new, but most of the potential pitfalls you know about. After the first three memory modules, you have hit the majority of the potential pitfalls.KitecWBT announced that it had appointed leading semiconductor company Kitec Design Co. Ltd. (Kitec) as its representative in South Korea to progress ongoing discussions with potential customers and partners in this key market. Kitec Design was founded in 2000. Kitec specialises in promoting Electronic Design Automation (EDA) and IP companies into the South Korean market.Weebit is planning to accelerate its entry into the discrete, stand-alone, memory chip marketFollowing the recent third-party validation of Weebit’s memory technology by XTX Technology, achieved four months ahead of schedule, Weebit is now planning to accelerate its entry into the discrete, stand-alone, memory chip market. This broadens the work program for Weebit beyond just the “embedded” domain.Discrete memory chips contain larger memory arrays and are more technically challenging than embedded modules, requiring additional development work before reaching productisation. A key element required for discrete memory chips is called a “selector”, which helps isolate the memory cells so that only the specific cells which should be modified actually are, and all the other cells are disconnected and not impacted. The selector basically allows you to switch off all of the other bits except for the ones you want to write to. You don’t want another cell to change its value by mistake because you were writing to another one.The work required for the discrete memory chips will be performed by Weebit in co-operation with its partner Leti. Leti has been developing a selector for the discrete memory market for the past several years which will allow Weebit and Leti to save years in the research time required to develop a selector for Weebit’s ReRAM in the discrete memory market. The planned program consists of three stages: stage 1 will be 3-months long and will be used to define the specific details of the selector and based on that the commercial details. Once these details are defined, the other two stages, totaling another 12 months, will lead to a demonstration of Weebit’s ReRAM cell working with the selector. Weebit’s current goal is to demonstrate the ReRAM cell intended for the discrete memory market by mid-2021XTXShenzen based XTX is a great partner for Weebit. They are a Chinese flash producer and they are a private company that is planning to IPO in a couple of years. That is why it is so strategic for them to work with Weebit. When they want to go to IPO, they want to be able to say that they are not just old technology, but are already starting to move to the new technologies, so XTX and Weebit have a good matching of interests. XTX have a lot of customers worldwide including some of the biggest players. They are a leading provider of flash memory chips, providing a variety of different Flash-based NVM solutions to approximately 2000 semiconductor customers globally. XTX currently services the consumer electronics, industrial embedded systems, telecom and networking markets and employs traditional embedded flash (eFlash) as well as NOR and NAND flash technologies in its solutions.In December 2019, a joint testing program of Weebit’s technology at XTX facilities successfully reproduced the technical parameters previously achieved on Weebit’s ReRAM technology with Weebit’s partner Leti. The testing program was achieved four months ahead of schedule and represents an important validation from an independent leader in the semiconductor industry. This independent validation is extremely important to Weebit and is second only to a purchase order in proving that Weebit’s technology is ready for commericalisation. An important thing to realize about both the Korean tier two company and XTX is that investment is not just about money. It is also about resources and time. When a customer is a first customer, they are taking a large risk. For XTX, there is no proof that Weebit is a great company. The actions from XTX of putting together a team that will sit down and work with Weebit is a massive investment from them. For a CEO to take a team, to get them to work with you and use their facilities and machines, that is an investment. That is money that is being invested in Weebit. It is not given to Weebit, but it is money that is invested in Weebit and it has a huge significance on its own. For a CEO, allocating resources to something is just as important a decision as giving a purchase order or paying a cheque to someone. A potential customer who actually goes and allocates the resources and the lab to verify and test Weebits technology is extremely beneficial to Weebit. The sooner Weebit starts working with actual companies like XTX the sooner they will get to the market. When development is done without a customer, there are assumptions that need to be made about what needs to be done, but every company has its own different needs. The feedback that they give is extremely valuable and the validation they do, e.g. the one done by XTX, goes a long way to procuring other customers as it lowers the anxiety about the fact that Weebit’s ReRAM is a new technology.Reason for current focus on the China marketFor a very long time the top 10 market cap companies in the world were energy companies and financial companies. Until 10 years ago, only one company out of all the top 10 was a company that was based on semiconductors, either doing semiconductors or relying on them. If you look today, 7 of the top 10 are companies that are reliant on semiconductors. The world has gone digital in a big way. Everything from fridges to microwaves are starting to use embedded semiconductors. The semiconductor market is growing very rapidly. It is more than 400 billion USD today and is expected to be 800 billion in the next few years. The fastest growing sector in the semiconductor industry is memory. The non volatile market is growing and the market needs new technologies since flash is no longer able to provide what the market needs.In China, there has been a huge push to get into the semiconductor market. In the top 10 companies worldwide you can already see 2 Chinese companies Tencent and Alibaba. For years already, China has been investing billions of dollars into growing their semiconductor capabilities and becoming self sufficient. In technology it is well known that you win and lose on the transition, once a technology is entrenched it is very hard for competitors to gain an advantage due to the benefits from economies of scale, accumulated knowledge/experience and the long research/development times required for new semiconductor technologies. Therefore, it is in the transition from one generation of technology to the next that new companies often gain a foothold. China has been pushing very hard to get into the market, but it is very difficult for China to be able to manufacture and compete on Flash memories with an established company like Samsung. They have a much greater chance in becoming the market leader when the market transitions to new technologies like ReRAM.Weebit are looking for partners who are looking to transition and want that transition to happen as fast as possible. Therefore, Chinese partners like XTX are good partners as evidenced by their willingness to allocate time and resources to working with Weebit. Weebit also has benefits here due to being Isreali. Chinese companies really like that Weebit is Isreali and not American.While English speaking companies also have their share of non-ethical things like IP theft. China has gained a reputation for IP theft due to instances of it occurring in the past. Weebit is being very careful in how they structure their deals so as to protect their IP. The members on the board have between them many decades of experience working in China, so they know how to protect themselves, their knowledge and their IP. That said, China has also been changing in its stance on IP theft and have been getting better on this issue. Weebit’s legal support (Kingwood Malison) are also experts in China and have big offices in China. Weebit already have a selector in the embedded domain. They need to develop a selector for the discrete memory domain.The work for XTX is for the discrete memory domain and this domain requires some extra work as it requires the development of a selector. In the embedded domain, Weebit basically has the solution they need. They have the megabit array and the 40 nanometer wafer. They just need to finish the memory module work. In the embedded domain, the size of each memory cell is not that critical. This is because the memories are relatively small when compared to the size of the whole chip. If you save 20% of the cell size, but the size of the memory compared to chip overall is only 5% the overall saving will only be a few percent. Therefore, a smaller cell size in the embedded domain is not that critical. This means that Weebit can use a simpler solution. In the embedded domain, Weebit uses a transistor which is a simple selector. The problem with transistors is that they big. but for the embedded domain that doesn’t matter too much. Why Weebit has accelerated the selector/discrete memory work Flash technology actually has two different segments: NOR and NAND. NAND flash, with some difficultly, has been able to continue to scale, but NOR flash cannot. NOR flash has limitations and companies like XTX want solutions to these limitations as soon as possible. NOR flash is also a good middle step for Weebit because they are normally megabits, 10’s of megabits or a little bit more. This is still a relatively small size, but the size of the bits is extremely important. This is why the selector work is needed. The plan originally was for Weebit to go in the discrete memory market after they had established the company in the embedded domain. But after listening to companies like XTX and hearing their desire for discrete memory solutions, Weebit has decided to fast track the discrete memory work. Another reason for accelerating the discrete memory work is the realization that they could utilize Leti’s previous work in the development of selectors. This will save them years of research and development time. Jackson LamWeebit has appointed Jackson as vice-president Strategic Alliances, China, in an advisory capacity. Jackson is responsible for Weebit’s partnerships in China, including customers, potential industry partners, and investors.The Chinese semiconductor industry is growing faster than the rest of the world with revenues expected to reach almost US$129 billion by 2020 – nearly a third of global revenue. The Chinese government recently allocated a 300B Yuan (US$47.4 billion) fund for development of China's semiconductor industry.Jackson Lam has almost 40 years of experience in the Chinese semiconductor industry. He spent 15 years with Panasonic in business development, and established an R&D and manufacturing company providing services for many well known international semiconductor companies. In the past 12 years, Jackson Lam has focused on helping Israeli companies set-up and expand their business in China.SiEnOn the 24th of February 2020, Weebit announced that they have signed a Letter of Intent (LOI) with SiEn (QingDao) Integrated Circuits Co., Ltd. (SiEn) to jointly investigate ways in which Weebit’s technology can be used in SiEn’s products. SiEn is China’s first semiconductor company based on a Communal Integrated Device Manufacturer (CIDM) model. This model aggregates companies with similar target markets, technologies, production lines and customers to share common manufacturing technology production goals. Weebit’s ReRAM technology could enhance SiEn’s competitive position by adding significant non-volatile memory (NVM) capability to products manufactured by SiEn. SiEn was established by Dr. Richard Chang, known as the "father of the Chinese semiconductor" and founder of China-based Semiconductor Manufacturing International Corporation (SMIC). This Chinese-American entrepreneur is well known in the semiconductor industry as a pioneer with decades of experience. He worked for 20 years at Texas Instruments in the US, successfully creating and managing more than 10 wafer fabs in the US, Japan, Singapore, Italy, Taiwan, and China. He also established Worldwide Semiconductor Manufacturing Corp (WSMC), which was acquired by Taiwan Semiconductor Manufacturing Corp (TSMC). Coby Hanoch, CEO of Weebit Nano, said: “We are honored to work with an industry luminary such as Richard Chang. This is a great vote of confidence in Weebit’s technology. Today there are dozens of fabs under construction in China, but SiEn’s fab is the only one taking the CIDM approach given Richard’s vast experience and knowledge of the market, allowing SiEn to bring together the design and fabrication processes, so instead of working with two separate entities, we can just focus on SiEn.” Dr. Richard Chang, Chairman of SiEn, said: “We look forward to our cooperation with Weebit. SiEn will finish constructing two fabs of 200mm and 300mm wafers this year, and we believe Weebit’s technology can be a perfect fit into our offering. Weebit has a very impressive Board and management team and I had the honor of working with Yoav Nissan-Cohen when I was setting up SMIC. Our production line will be based on STMicroelectronic technology, similar to that used by Leti, making the technology transfer simpler, and is expected to bring forward the time required to bring products to market.”SilvacoIn February 2020, Weebit and Silvaco announced that they had developed a Technology Computer-Aided Design (TCAD) solution to accurately model the electrical behaviour of Weebit’s ReRAM devices.?The models are based on Weebit’s physical samples and data. This new simulation model capability in Silvaco’s advanced Technology Computer-Aided Design (TCAD) solution, Victory Device?, accelerates the incorporation of Weebit’s ReRAM modules into advanced semiconductor designs used in the discrete and embedded memory chip markets and speeds Original Equipment Manufacturer (OEM) product development schedules by reducing technology testing stages.Victory Device analyzes the electrical, optical, chemical, and thermal behavior of semiconductor devices. Its physics-based simulations dramatically speed up the design, fabrication, and commercial use of semiconductor device technology through the elimination of expensive and time-consuming experimental wafers during technology adoption by OEMs.Coby Hanoch, Weebit CEO, said: “Silvaco provides state-of-the-art device models and TCAD tools that are widely used by semiconductor companies. Our successful joint development provides customers with early access to Weebit’s ReRAM technology on the most powerful simulation tools in the industry. The newly developed TCAD model will enable customers to fast-track the development and release of their next-generation products, while saving expensive and time-consuming runs in fabrication. Costly learning, development and testing cycles that take several months have now been eliminated. This is another key milestone step by Weebit to accelerate the adoption of its ReRAM technology by a larger number of global OEM customers.”Other activities300mm/28nm activity moved to a lower priorityAfter meeting with and beginning work with potential customers and partners, e.g. the tier two Korean company and XTX, Weebit has changed the priority of the 300mm/28nm activities. Originally a year ago the plan was to put a focus on these activities. However, after meeting with and talking to the potential customers Weebit realized that the companies that are the most favorable to cooperate with them don’t need this at all. As these first potential customers don’t require 300mm/28nm wafers, it has been moved to a lower priority. Due to limited time and resources, if the customers don’t need the 28nm geometries then it is an inefficient use of Weebit’s current resources to focus on that. It is something that is important and something that Weebit will get back to, but it is just a matter of priority with the resources that they have.The 300mm/28nm activities are longer term goals that would be required for someone like a tier one customer. The second tier companies, e.g. the Korean group and XTX, have shown a whole lot of interest to take up the current technology. If and when Weebit manages to strike up a commercial agreement with one, two or five of those second tier companies, that’s when they would start getting back to the 300mm arrays and start getting down to the requirements that a tier one, like a Google or Intel, would need. An ideal scenario for Weebit would be to get an Intel like company to take up their technology, but these types of companies would want Weebit to first commercialize with a second tier. If they see that and see that it is working, then they will start to get involved. Getting to the tier one level is a long term goal of Weebit, but they want to consolidate and commercialize first. Transferring the technology from Leti which is a development facility to a production facilityWeebit has focused development at 200mm because the cost and availability were much more favourable. It is much easier to progress at 200mm and since every minute on a fabrication line is very expensive it is never easy to secure time. However, Weebit believes that they are talking with the right partners and fabricators to ensure that they can transfer their technology to production in line with their published timelines. While Weebit is currently working on 200mm wafers, and some competitors are working on 300mm wafers, it should be noted that Weebit’s 200mm wafers have been developed using 300mm equipment, so moving to 300mm wafers is mostly a matter of money and focus, with manageable technical risk.The good thing is that Leti has world class facilities. In fact, they have what could be called a superset of the different machines that you see in fabrication places, so Weebit has progressed well on the path to productisation, but each fab has its differences and required optimizations, so that’s where the extra work would be. Working these out is also part of what the memory module milestone for the Korean tier two company is about.How commercial agreements workWBT is expecting to generate revenues from three separate, but related streams, which are paid per customer project/product non-recurring engineering (NRE) work related to specific integration needsone-off licencing paymentsroyalties from customers’ sales of semiconductors that use WBT’s ReRAM technologyFigure 1 Source Pitt Street ResearchIt is standard in the industry for IP licencing fees to be between $500,000 and $2,000,000. It is expected that Weebit’s first deals would be more at the $500,000 levels than the higher levels. The royalties are typically 1% to 1.5%. 3% is possible, but would only be done in cases where there were more dependencies on Weebit. Looking forward, it is important to note that in the semiconductor industry everything is measured in years. However, revenues will occur earlier than this through IP licencing fees and NRE. NRE will happen when they make strategic partnerships with other companies in order to develop versions of the memory that are suitable to their needs. NRE is an upfront payment that will be made to Weebit in order to adapt their memory to the companies needs. Once production begins the company will then begin paying royalties to Weebit. It is licence fees per customer projectIn general, the business model in the semiconductor industry is based on a licensing fee at the beginning of the project and then royalties when they get into production. That’s the basic business model. An important point to note is that it is license fees per customer project, not per customer and that customers often have many different projects. So for each one of these projects, they would be paying a license fee. Now, when a customer starts using Weebit’s technology on tens of projects, then they would want a global licence fee. Hopefully, Weebit can get to that stage and quickly. But, in the beginning it is licence fees per customer project.After the licence fee is paid, the company would embed Weebit’s technology into their design and it takes them a year to a year and a half for them to finish doing their design and get into the fab to start manufacturing. So there is a delay, but when it gets into the fab that is when Weebit would start getting royalties. Non-recurring engineering (NRE) deals are also possible and once the first memory module has progressed it will be easier to make these kinds of deals.Exclusivity deals are not being consideredAlthough there is always a price for everything, Weebit would not be interested in an exclusivity deal or the amount would be so large that other companies would not be willing to pay it. An exclusivity deal would be bad for Weebit because it would shut off the rest of the market, so if it was to happen it would need to be for a large enough amount that it would compensate Weebit for losing the rest of the market and most companies wouldn’t be willing to pay that.Weebit has a three phase approach to the marketThe first phase is the embedded market which is what they are targeting with the Korean customer. For the embedded market, they largely have what they need and just need to get the module work completed and then they will start working on commercial deals. The second phase is the discrete components or what some people call stand alone modules which are memory chips. This is a more advanced market and it is where the work with XTX is. The verification and analysis of Weebit’s technology by XTX also involved a deep analysis of what is still needed in order to make the technology work for discrete components. Discrete components split into two markets: NOR flash and NAND flash. The first market would be NOR. The third phase would be the Neuromorphic computing market. Consolidation adjusted top 20 holders movementHolder Name16 August 201925 September 201822 August 2017Citicorp Nominees Pty Ltd3,589,897 735,456 349,478 IBI Trust Management (James Tour A/C)1,509,462 1,805,806 1,805,806 Silver Horizon Pty Ltd1,429,150 1,149,150 1,149,150 IBI Trust Management (David Perlmutter A/C)1,153,500 1,153,100 1,153,500 BNP Paribas Nominees Pty Ltd (IB AU noms Retail client DRP)1,132,404 907,625 0Droxford International?Ltd1,044,000 957,192 0IBI Trust Management (Yossi Kerret A/C) 753,500 1,153,100 1,153,500 Two accounts -Mr Fabian Seibold + Dr Silvia Maria Seibold (Seibold family A/C) (377,832) - Mr Fabian Seibold + Mrs Silvia Maria Seibold (The Seibold family (333,333) 711,165 265,362 0IBI Trust Management (Limor Levy Kindler A/C) 688,355 677,834 753,456 IBI Trust Management (Rami Hadar A/C) 576,874 743,086 913,196 Mrs Anne Le 531,489 00Mr Eric Mark Caspary 523,731 247,228 0Everbloom Trading Pty Ltd 466,000 00Mr Eli Staub 425,144 00Shinewarm Resources?(HK) Group Ltd 400,000 320,689 400,000 IBI Trust Management (Hagai Arbel A/C) 388,728 1,044,000 1,128,728 BT Portfolio Services Ltd (Michael Burnell S/F A/C)340,222 00BT Portfolio Services Ltd (Horseland Saddlery S/F A/C) 330,222 00HSBC Custory Nominees (Australia Ltd 315,458 00Previous top 20 share holders who are no longer in the top 20IBI Trust Management (CTW Changing the World A/C)05,593,088 5,663,664 IBI Trust Management (Borenstein ltd A/C)05,593,088 5,663,664 IBI Trust Management (Motti Gross A/C)02,154,557 2,154,557 Buddy Platform ltd0793,196 0IBI Trust Management (Amir Regev A/C)0400,000 677,834 Um Anne0261,642 0Um Tim Tchun0240,041 0102 Capital Management (Gabriel Hewitt A/C)00 1,280,151 Victory Mining Pty Ltd00 1,044,000 Potash Minerals Ltd00 907,625 102 Capital Management (Oran Dorel A/C)00 560,016 Celtic Capital Pty Ltd00 516,902 102 Capital Management (Tzipora Silver)00 410,411 Kloss Steve00 410,411 *Most 102 Capital Management accounts moved to IBI Trust Management in 2018-2019. Both were grouped under IBI Trust Management* Units adjusted for 25 to 1 consolidation in February 2019. Calculation in 2018 and 2017 was total units divided by 25 and rounded up to the nearest number. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download