Security
Synopsys' recently acquired Intrinsic ID. Will this affect your PUF-based applications? Also, will this have an impact on similar technologies and IPs of your company?
-
After the launch of NeoPUF which is our PUF technology, many customers have switched to use our NeoPUF and OTP instead of SRAM PUF, from Intrinsic ID’s technology is based on. The main reason is that our NeoPUF has better properties than SRAM PUF, including:
1) Enhanced stability.
2) Increased reliability.
3) Eliminates the need for error correction.
4) Eliminates the need for OTP for storing helper data.
5) Demonstrates radiation-hardened capabilities
We are confident that we will become a leading company in the field of PUF, OTP, TRNG, and RoT IPs. Furthermore, Synopsys’ acquisition of Intrinsic ID is a good sign that the industry recognizes the significant potential and future demand for PUF in chip security and driving the provisioning of these IP solutions through acquisition to better penetrate the market and capitalize on these opportunities. eMemory recognized this trend in 2015 and developed the best PUF-based technology to address the problems encountered by SRAM PUF. Combined with our own OTP, we can provide customers with complete hardware security solutions. We believe that we are at the forefront of this trend and will be the first to provide one-stop-shop services to meet customer demands.
It looks like TRNG can provide a random source for generating a secret key and making obfuscation that will help resist attacks. Does this imply that having high-speed TRNG in security design is more important than having a PUF?
-
In very high security systems, the high-speed TRNG is very important as well as PUF because in our invention, the high-speed TRNG relies on the PUF as input to generate random numbers. PUF also plays an important role as unique ID, which requires unique numbers from natural randomness. By using this natural randomness, the unique ID can also create the mother security key for the system. Therefore, in very high security systems, both PUF and high-speed TRNG are very important.
What role does eMemory play in OpenTitan?
-
The OpenTitan platform sets the standard for securing data transmission between IoT devices and the cloud through hardware security. Our security IP, being a hardware security solution, can be directly applied to Google's OpenTitan platform for encryption and decryption.
Although the recent blue screen crash event at CrowdStrike seems unrelated to security, CrowdStrike's endpoint security is also touted as Zero Trust. How can eMemory compete with software companies using a hardware approach?
-
Within the Zero Trust framework, eMemory's hardware security provides a hardware root of trust and security computation capabilities. For cybersecurity software companies, hardware security can enhance the security of their applications. eMemory's security IP aims to boost the competitiveness of these software companies, not to compete with them.
What is the reason why PUF’s royalties have not increased significantly?
-
Licensing for our PUF technology has been increasing. The first customer to adopt our PUF was China’s largest chip company. However, due to US government sanctions, we were unable to continue receiving royalties from their production. Despite this, we have more than 60 tape-outs gradually entering the mass production stage and will soon contribute to our royalties.
Chairman Hsu mentioned that eMemory’s True Random Number Generator (TRNG) is the fastest in the world. If this is the case, why haven’t major HPC vendors used it?
-
Our TRNG is based on our OTP and PUF technologies. This is a hard IP and must be qualified in each process. The number of our qualified processes continues to grow, progressing towards the most advanced nodes. As more customers adopt our technology and accumulate production records, and as hackers use faster computing to launch attacks, the market will demand faster random number generators. We are working with our CPU partners to promote this technology and are very confident about its future developments.
What is the reason why PUF’s license began to increase?
-
The reason for the fast growing demand of our PUF-based security IPs is because: high security architecture can be simply implemented.
Security is now a required element to be connected to cloud services. As security becomes necessary for applications in edge and cloud devices, high- security solutions that can be easily implemented are preferred. Our PUF- based solutions provide integrated secure storage and secure Root of Trust, becoming a popular choice for security applications. In the past, creating a security system requires complicated cryptographic processing. With our perfect randomness PUF IPs, security functions can be implemented much easier and faster into chip functions without complicated crypto-processing, which typically increases chip size, sacrificing speed and power consumption.
How is eMemory and PUFsecurity’s crypto co-processor unique compared to others?
-
In the past, when people seek security solutions, they have to look around and search for a couple of providers to provide 1) security storage, 2) Root of Trust, and 3) crypto engines. After they have IPs with all three components, they still need to figure out how to integrate all into a solution. To become to a security co-processor, SoC designers need to add firmware and API such that it can be easily embedded into SoC.
Our PUFcc integrates all three elements into one security solution and is able to provide firmware and API to facilitate the design of SoC. Since our solutions is built on our existing OTP process platforms, we are able to provide our solutions quickly as we currently have more than 600 OTP process platforms ready. No other security IP companies are able to do this.
PUFsecurity recently obtained the Riscure Common Criteria Certification. Does this mean that PUFsecurity can enter the security IP application market more quickly?
-
The benefits for customers are: if customers adopt our solution for their product designs, we can help customers pass security certifications, without extra R&D and certification costs.
Rambus also has a security solution. What is the difference between Rambus and eMemory’s solution?
-
Rambus only have algorithm-related solutions. If customers adopt Rambus’ solution, they will still lack a secure OTP or PUFrt to serve as the chip’s root-of-trust.
What is the business model for working with Arm? Do you need to share royalties or licensing fees with the other parties?
-
We still charge license fees and royalties. The joint activity is mainly for promotion, we do not need to pay the other party any fees or profit shares.
Can the U.S. government restrict your security-related IPs from being licensed to Chinese companies? What is the company's patent protection policy? In addition, patent protection typically lasts 20 years, can NeoBit continue to receive royalties since its technology is 20 years old? Can other companies use NeoBit's technology without the company's authorization because the patent rights have expired?
-
Our technology is a genetic invention and there are no government restrictions. The technology patent is for 20 years, but we improve the version periodically to extend the patent period. For example, NeoBit's patent was extended to 2034 in 2014. In addition, the complete IP, including circuit design, has 50 years of copyright protection, and the royalty payment contracts we license to foundries have no termination date.
Although advanced process eFuse has the drawback of burn-in error, foundry can still provide hundreds of thousands of reliability tests, if customers do not need to test to that number, how can we convince customers to use NeoPUF?
-
Reliability is not the only consideration.
UID and Key ensure the chip's identity cannot be tampered with, copied or stolen, and the process of generating it needs to be strictly controlled. To use eFuse to for Key storage or UID, you must first write-in the key or obtain a random number from outside. The eFuse of each chip is empty at the beginning, and the data is written externally, which gives attackers opportunity to manipulate information. Using NeoPUF to generate Key or UID is from the chip's own PUF and cannot be written externally. The security of these two methods are different.
In addition to generating the UID and Key, which cannot be interfered with externally, our PUFrt and Secure OTP both contain physical, electrical anti-tampering designs (certified by Riscure), which can further prevent attacks.
NeoPUF has many applications in AI SOC/FPGA/DPU/HPC related CPUs, how do you deal with the US restrictions and how does KYC deal with it?
-
We are not affected by the U.S. requirements and do not need to provide KYC information. Our customers are mainly from Europe, US and Japan.
How will the post-quantum encryption algorithm affect the company?
-
Encryption uses crypto algorithms to protect sensitive information, including secure websites or even emails. The widely used public-key encryption system relies on algorithms difficult for even the fastest conventional computers to solve in limited time. However, quantum computers will threaten existing public-key encryption within a decade, The National Institute of Standards and Technology (NIST) is expected to announce new algorithms for encryption in 2024. The main problem with the potential post-quantum encryption algorithm is that the key size is very long, which may be thousands of times longer than current RSA and ECC keys.
The strength of eMemory's PUF-based Root of Trust lies in its flexible and high-speed key generation, which is currently the world's easiest, fastest, and safest method. It will be necessary for customers to adopt PUF-based Hardware Root of Trust to face post-quantum crypto algorithm. The last large-scale replacement of the encryption algorithm was around 2000. At that time, the United States decided to adopt the AES algorithm. It took ten years for major related companies to switch to the new encryption methods, so we expect that changing to a new encryption algorithm will also take around ten years. This is very positive for speeding up customers' adoption of our solution.
To learn more about Post-Quantum Cryptography (PQC), please visit this article.
You mentioned that AI-related applications are among the few early adopters of PUF. Why are they the first to adopt PUF?
-
Advanced AI models are typically stored in conventional commodity flash memory, and this standard storage device has no security function, so it can be stolen or modified. Our PUF-based hardware security Root of Trust IP (PUFrt) and Crypto Coprocessor (PUFcc) can encrypt and authenticate these AI models to protect them from being stolen or modified, which is why customers need PUFsecurity and eMemory solutions.
In addition, all processors have SRAM, one type of volatile memory, just like DRAM. When SRAM is manufactured, it will have bad bits that need OTP for memory repair. In the past, eFuse was used to do SRAM repair functions. However, the SRAM density in AI chips is much bigger, which eFuse cannot fulfil due to density limitations. Our OTP offers 500 to 1000 times bigger density than eFuse. Therefore, for AI with large-density/SRAM, our IP become a necessity. Based on SRAM repair and security requirements, most AI chips will convert from eFuse to our OTP.
When will you start seeing contributions from your collaboration with Arm?
-
Our cooperation with Arm has been ongoing. Some customers have already taped out our security solution through ARM's platform and contributed license fee already.
There are ongoing concerns for operations in data centers to be secure and confidential. I heard that AMD, Google, Microsoft and Nvidia recently worked together to develop a Caliptra initiative to define standards for hardware root of trust (HRoT). Are you guys a part of this standard? Follow up: what does this mean for eMemory?
-
Caliptra is an open-source standard that aims to integrate security mechanisms into chips and plays a crucial role in the Open Compute Project (OCP) reference design. Its primary goal is to establish an open-source standard for the Hardware Root of Trust (HRoT), which is essential for hardware-based security functions embedded in CPUs, GPUs, SoCs, ASICs, Network Chips, SSDs, and more.
As far as we know, this is the first time TRNG, OTP and PUF are collectively addressed within a single standard, which happens to be our strengths and expertise. This also means the HRoT of the processor for data centers must include hard macros, TRNG, OTP, and PUF, as key components. However, since hard macros are unavailable through open source, they must be obtained through licensing from us.
While Caliptra is still in its early stages, we foresee broader adoption for hardware-based security in data centers in the future, which opens up significant opportunities for our PUF-based technologies.
In your partnership with Arm, which areas are you targeting? When will you start to see real contributions from them?
-
We are working with them on Confidential Computing Architecture and Corstone Architecture, covering applications such as Smartphones, Automotive and IoT (edge computing). There are license contributions from Automotive and IoT customers already.
There have been many discussions recently regarding the use of Arm CPU in PCs. How will this affect your partnership with Arm?
-
Arm's CPU application in PC belongs to their Client Business. Similarly, CPU in mobile also belongs to Client Business. Our PUFrt will be used in the Runtime Security Engines (RSE) of Arm's mobile CPU. Once verified, it may also be used in the CPU for PC.
PUF-based technologies have seen substantial growth in licensing contributions this quarter. Who are some of your current customers/applications that have adopted PUF?
-
Other than CPU partners, we have also licensed to DPU, AI, Edge Computing, HPC, IoT and ADAS customers.
If the Arm partnership is very successful, what does this potentially mean for eMemory?
-
Arm is the leading CPU IP company with more than 95% of the market share in mobile. The successful co-development of PUF-based Root of Trust security into their CPU has the following implications :
- Our PUF-based security Root of Trust solution is the best in the market.
- We can enter the mobile market with Arm, which currently holds over 95% of the market share.
- We expect our market share in the Confidential Computing market will also be very big.
Business Model
The aggressive expansion of Chinese foundries may put severe downward pressure on wafer prices. How will this affect your company?
-
Our OTP technology is now a standard for applications, like Driver, PMIC, and various sensors, in mature processes. As foundries expand, the need to license our technology is expected to boost our licensing revenue. While the industry faces oversupply and pricing pressures due to foundry expansion, the effect it has on our company will be offset by the increase in royalties per wafer as existing customers move to more advanced processes.
In order to survive, existing foundries are focusing on developing specialty processes with high value-add, such as MTP, embedded flash, and PUF, as is happening now. Our royalties for MTP and PUF-based are much higher than OTP. Furthermore, the increasing contribution from the higher wafer prices of advanced nodes leads us to believe that average royalties per wafer will continue to rise.
Foundry expansions in Europe, the US, and Japan are driving the need for local production by large customers in the future. Deploying our IPs simultaneously in foundries across each region will facilitate convenient adoption for customers, driving increased usage and expanding our business across diverse customers and applications.
Will cost constraints push your customers to use eFuse, which is free of charge, to reduce expenses?
-
Many of our customers developed their first-generation products using foundry’s eFuse. However, in response to competitive pressures and the need for improving cost and performance, they transition to our IPs. Our OTP offers improved yield and reduced chip area compared to eFuse and enables customers to protect their intellectual property (i.e. algorithms) stored in the chip. The success of these has led to the continued adoption of our IPs in subsequent generations of products.
Over the past year, discussions about the slowdown of Moore's Law have become more prevalent. Will this affect your rate of improvement in foundry penetration? Considering that many IP replacements occur during process upgrades, could the deceleration of Moore's Law potentially have a negative impact on your company?
-
Since 2014, we've been developing IPs for technologies below 16nm. Typically, the foundry provides in-house eFuse for the initial customers for the most leading processes. After the process stabilizes, third-party IPs, like ours, are permitted for technology verification, often occurring after the first batch of customers has entered mass production. This means that 3rd party providers like us need to wait for those migrating from the previous generation to adopt our technology. As Moore's Law slows down, the launch time for the next generation extends, creating a longer interval time like current 2nm and 3nm. This year, once we successfully verified 3nm, we still can catch customers seeking the most advanced process. Overall, this is positive for us.
Moreover, they are always customers migrating to more advanced process node, such as moving from 7nm to 5nm or 16/12nm to 7nm. As long as we have completed process verification, there will still be many opportunities for customer adoption. Even with mature processes, these opportunities remain unaffected by the slowdown or changes in Moore's Law.
The chairman talked about the concept of production capacity, highlighting the increasing number of foundries capacity will increase your total addressable market. Since you have been actively establishing our technology platform in these newly built facilities, can they directly transform to our royalties in the future?
-
Yes, our total addressable market will increase as the world expands foundry capacity and moves toward more advanced technology. Our technology coverage in each foundry will increase as more technology process develop and more fabs are established. With each generation of technology, there is a need for continuity to move to the next generation, therefore our technologies are increasingly available in each foundry. Historically, our royalty income for each foundry have experienced an upward trend. Consequently, as the number of fabs and technology platforms increases, we anticipate a corresponding growth in License fees, NRE, Usage, and Royalties.
In the past 10 years, your employees have increased by an average of 5% each year, while your revenue increased by an average of 20%. Additionally, your company also managed to develop many technologies. Why is this the case? Is the goal of increasing staff in the future still the same?
-
The first main reason why we can keep developing new technologies but managed to keep our employee headcount relatively the same is because our IP is reusable. Once we qualify our IPs on the foundry platforms and register it on our IP library, customers can use our IPs anytime unless they require customization. This means that our R&D team can always focus on developing the new technologies and on the more advanced nodes. The other main reason is because of our unique business model, since 70-80% of revenue is contributed by royalty. A royalty-based business drives revenue as more customers adopts our IPs without increasing the employee headcount.
In the future, our hiring plan will mainly revolve around advanced process projects and special projects such as NeoFlash, RRAM, and Security. We will maintain a similar hiring pace.
The foreign exchange losses in the fourth quarter of last year seemed to have some impact on your earnings. How do you plan to address this matter?
-
Our largest foundry partners have suggested paying royalties in Taiwanese dollars based on the exchange rate at the payment date and time. Other customers directly pay us in US dollars. We deduct necessary fees, convert the amount to NT dollars, and keep the remaining sum in US dollar time deposits to gain higher interest compared to put in NT dollar. The foreign exchange losses incurred in December will be partially offset this quarter as the NT dollar deprecated early in the year. We will review our currency accordingly.
Lately, there has been instances of M&A in Taiwan semiconductor industry. Have you considered expanding your technology portfolio through M&A to expedite your growth? If yes, what technologies or IP products are under consideration? If not, could you provide more insight into the reason behind this decision?
-
Our company has many groundbreaking NVM inventions and technologies that surpass those of other companies globally. As long as we continue to develop these inventions and technologies along our roadmap, we can accelerate our growth rate. Currently, acquiring other companies is not under consideration.
The announcement of the new board members this year include notable additions such as Dr. F.C. Tseng and Dr. Jack Sun, both well-known figures closely linked with TSMC. What is the reason behind selecting these two members and how does it benefit the company?
-
Per FSC regulations, our independent directors have served three terms, and therefore, more than half of them need to be replaced. eMemory is a leading company in embedded NVM within the IP industry. We will continue to collaborate with foundries in advanced processes to develop cutting-edge NVM and security technologies, while also expanding our business strategy and partnerships. The newly nominated directors will provide invaluable support to the company, bringing expertise in technology, operational strategy, and forging key partnerships.
TSMC has lowered its growth forecast for the global semiconductor foundry industry, indicating weak demand across industries apart from AI-related products. What is the impact on your business in 2024?
-
For the business growth in 2024, we are very confident due to the following reasons: 1) Our IP portfolios are increasingly diversified, including OTP, MTP, EE, Flash, RRAM, MRAM, PUF, SecureOTP, PUFrt (Root of Trust IP), and PUFcc (Secure Co-processor IP). This will drive significant growth in Licensing, NRE, and Usage revenue in 2024 compared to 2023. 2) Regarding royalties, according to the feedback from our customers, their inventory has decreased to a certain extent, returning to normal levels. Therefore, demand for products in mature nodes will also return to normal levels of production. Furthermore, we have accumulated over 1500 tape outs in the pipeline in the past three years, which will drive our royalty growth further.
During the 4Q2023 earnings call, you mentioned that there are three main areas in AI applications, Computing, Storage, and Sensing. Which of these three areas will contribute the fastest to your royalty revenue? Which one will have the largest addressable market in the future?
-
For AI application systems, there are three main components: data input, data/model storage, and computing (accelerators). Currently, data input will go through various sensor applications, and eMemory has already entered the field and started receiving royalties. For data storage in NAND/DRAM, there are quite a few customers in NAND SSD controllers and CXL memory interfaces using our IPs, which also results in royalty contribution. In terms of computing, mainly in advanced nodes, customers are using our Root of Trust IPs and SRAM repair IPs, and we believe it will significantly contribute to royalty in the future.
Will the excessive expansion and price-cutting competition among Chinese foundries affect the company’s royalty?
-
The impact of the excessive expansion of Chinese foundries in mature process nodes will be limited for the following reasons:
1) The leading foundries have not followed pricing cuts as reducing foundry prices does not necessarily guarantee an increase in market share.
2) Customers choose foundry partners not only based on pricing, but also on time delivery, time to market, yield, and services.
3) Furthermore, geopolitical risk pushes major US chipmakers to shift away from manufacturing in China. Also, semiconductor manufacturing localization policies in regions like Europe, the US, and Japan, all limit the effects of China’s overcapacity and downward pressure on foundry prices outside of China.
As our technologies develop towards more advanced nodes and specialty processes, such as NeoFlash and MTP, the royalty we receive will be higher than average. Therefore, we are not concerned about the impact of price-erosion of Chinese foundries. Additionally, based on our previous experience, IPs’ demand and adoption tend to rise as the prices of foundry wafer price decline, ultimately resulting in increased penetration and royalty revenue over time. Also, as foundries continue to expand their capacities, our royalty revenue will grow continuously.
There are multiple companies claiming to enter the OTP market and stating that their IPs are more competitive because their IP size is smaller than yours. Will you be losing orders because of them?
-
OTP and MTP technology have been in the market for over 20 years, and competition has never stopped. Our company continues to progress into the most leading process nodes, extending our technology portfolio into the security area. Reputation is very important in the IP industry and our customers, whether they are foundries, fabless, or even OEMs, are all satisfied with our technologies and services. All of these are reflected in our revenue and earnings. We strongly believe this trend will continue in the future.
UMC mentioned the postpone of its Singapore fab expansion. Will this affect you?
-
Foundries frequently revise their expansion plans and schedules in response to economic cycles, customer demand, and various other factors. Our collaboration with UMC is not limited to specific processes or location. There are multiple projects ongoing across all our technologies. Therefore, the delay in expanding a particular foundry location will not affect us.
Your customer mentioned in their earnings that Driver IC is facing severe price pressure, with hopes of foundry reducing wafer prices. Will this affect your royalty?
-
Our DDI customers continue to migrate towards more advanced processes, with the proportion of high-end OLED DDIs steadily increasing. Additionally, the number of OLED DDIs will increase in foldable smartphones, along with content increase on panels (such as e-paper, labels, PMIC-related and more). Overall, we believe that these factors will mitigate the impact of declining wafer prices and remain a strong cash cow for us.
Does China's semiconductor localization plan include IP? Do we face competition from local IP vendors in China? Are customers switching to domestic IPs due to government policies?
-
As we mentioned before, the NVM IP business is characterized by high technological barriers, very long periods of development, and relatively small revenue output compared to foundries and chips. Therefore, this is not the focus of localization of semiconductor policies. While we have observed some competition from local IP vendors, customers are very careful about IP adoption, which is not only cost consideration. Factors such as patent protection, quality, reliability, and technical support are much more important considerations. Thus, our licensing cases in China continue to increase. Additionally, some customers have encountered big issues after trying local IP vendors and have subsequently switched to using our IP.
Starting from 3Q2020, the average NTO (New Tape Out) number is around 140~150 for every quarter with no significant increase. Does this number mean that the company is deliberately selecting cases or your R&D resources are limited? How do you predict your future NTO trend?
-
Approximately 85% of our NTOs are from Usage cases, where customers re-use existing IPs and therefore, require minimal engineering resources. Although there hasn't been a significant increase in NTOs per quarter in recent years, our tape outs are continuously moving towards more advanced nodes. The cost of design, development and mask layers increase at more advanced nodes, resulting in fewer NTOs in advanced processes compared to mature processes. The average royalty per tape out in advanced nodes is much higher. In addition, as the proportion of tape outs in advanced node increases year over year, so does our licensing revenue.
The license fee for MTP in 2023 grew nearly 100% compared to 2022. When can we expect to see a significant royalty contribution, and in which area? Additionally, how do you view the ESL market?
-
In the past three years, we have accumulated over 200 NTOs for MTP, primarily used in applications such as MCU, PMIC, and sensors. These products have gradually entered production, contributing to royalty revenue. In the first quarter alone, we saw over 60% year-over-year growth in MTP royalty. As for ESL, we already have customers in production using our OTP for single-color and three-color driver ICs. Due to customers requiring MTP on new four-color driver ICs, we already have many customers adopting our MTP on multiple foundry process platforms over the past year, which will contribute to our royalties soon.
TSMC is accelerating their investment in advanced 3D packaging technologies such as CoWoS, InFO, and SoIC. Are there opportunities for your IPs? Can you tell us if you currently have any customers adopting these applications?
-
In 2.5D/3D packaging, chips with different functionalities are packaged together, therefore, if one chip fails, the entire package chip will fail. Our IP can help repair these failed chips. For example, memory components (such as DRAM and SRAM) and image sensors (CIS) are frequently combined with logic ICs. OTP plays an important role as a high-density memory repair and sensor repair in these packaged chips. We already have many customers in this area with continuing adoption from new customers and applications.
We noticed a significant increase in operating expenses in Q12024 compared to last year. Was there a rise in recruitment during this period? Additionally, could you provide an estimate of the growth ratio of your operating expenses for the entire year?
-
The increase in operating expenses in the first quarter is mainly due to the increase in employee and director bonuses as a result of increased net profit before tax. We allocate bonus expenses each quarter as a percentage of net profit before tax (15% for employees, 1.5% for directors), which were included in that quarter’s operating expenses. In the first quarter, there were foreign exchange gains in non-operating income, leading to an increase in net profit before tax and non-recurring increase in bonuses expenses.
The fixed operating expenses in the first quarter, excluding employee/director bonuses, were approximately NT$280 million, which is quite similar to previous quarter. We currently do not have any plans for significant recruitment, therefore, apart from the annual salary increase for employees (averaging 3-5%), changes in future operating expenses will primarily be influenced by changes in net profit before tax.
Recently, Trump said that TSMC should pay the United States for protection. Will this have an impact on eMemory?
-
Our IP is licensed to foundries around the world, including in the US (such as Intel and Global Foundry). With our platforms established globally, we have a widespread range of technologies and in various processes, making us less susceptible to political influence.
Price-cutting competition among Chinese foundries is very serious. Have you been affected?
-
In recent years, China has actively expanded its production capacity by building new foundries and new fab areas for mature processes, many of which are using our logic NVM technology. In terms of royalties from China, the capacity expansion and increasing penetration rate will drive overall growth, even if foundry wafer prices decline.
What is the difference between eMemory’s IP business and other IP company’s IP business?
-
There are two categories in IP companies:
1. Pure design IP companies
These companies use existing transistors to design specific functional IPs, such as standard cell, SRAM, high speed IO, etc. They license to fabless companies directly and are companies like ARM, Synopsys, and M31 and RISC-V related.
2. Technology + Design IP companies
These companies have their own device technology developed at foundry and they license the technology to foundries, such as OTP, MTP, and Flash. They also use the device they invented to design IP for fabless companies. These companies are, for example, eMemory and SST. SST’s Flash cell is only on the embedded Flash process, while eMemory’s OTP cell covers all the processes. Therefore, OTP’s market size is much bigger than Flash.
The US CHIPS Act restricts those who received subsidies from going to China. Will this affect eMemory?
-
The CHIPS Act subsidizes foundries that built fabs in the US. It is only applicable to those who build fabs with the money from the subsidies. Since we did not receive any subsidies, we’re not within the restricted scope.
Can the US Government restrict eMemory’s security-related IPs from being licensed to a Chinese company if eMemory wants to do business with the US government?
-
As mentioned before, our IPs are our original invention, we have patent rights for our IPs in major regions of the world. We also did not receive any subsidies from these governments so we’re not restricted by any national government.
Can we predict future royalties based on tape-out numbers or license fees?
-
Tape-outs vary significantly in terms of customer size, product application, and mass production scale, so it isn’t easy to use them directly to predict future revenue. However, with the increasing number of process platforms and technologies, there is a clear long-term growth trend for tape-outs and licensing revenue.
Are there any customer requests to decentralize R&D resources, does this have any impact on eMemory?
-
We have been licensing our IPs to foundries all over the world. The advantage of using our IP is that our customers can go to different locations for production. There is no such request from customers to decentralize our RD outside of Taiwan. But we did set up Japanese RD team for the reason to recruit talents.
Do Chinese companies have to obtain permission from the U.S. government to license with a company?
-
Since our IP technologies are our own invention, we don't need any US or government approval so we can license it directly to anyone.
Does the company have a backup plan if there is a war between the two sides of the Taiwan Strait? Are customers worried about this risk and has this lead to customers avoiding the company's technology?
-
Although our main R&D is in Taiwan, our IP data is stored in the cloud server. Our patents are registered in all major regions of the world and protected by international regulations. As long as the company's technology is used, regardless of the chip company or foundry, they must pay the license fee and royalty, which is the greatest protection for the company's shareholders. Currently no customers have avoided our IPs due to concerns about war.
Is the patent right held by the inventor or the company? Can the patent inventor use their invention if they leave the company?
-
All patents are held by the company, and the inventor cannot use the patent if they leave.
Does the company have any issues with accounts receivable? Are Chinese customers not receiving their money?
-
We have no accounts receivable problems. Since the company was established, we have only had one small collection problem in the early days, and for most of our Chinese customers, we take prepayment.
Many mature processes are located in China. In the past, even Arm faced the problem of collecting royalties from Chinese customers. If customers all move their mature products to Chinese foundries, will eMemory face trouble receiving royalties?
-
We have 9 foundries customers and over 900 chip customers in China, accounting for about 10-15% of our total revenue. There have been no problems with collecting royalties and other payments. This is related to the nature of IPs. ARM offers soft IP, providing customers RTL (similar to software for chip design). They collect licenses and royalties from chip companies. They must audit their customers if customers use the same RTL to other chips without being licensed.
We license the fundamental transistor to the foundries. The foundry pays royalties based on the royalty rate and wafer volume, while the chip companies pay design licensing fees only . Since there are only 9 foundries but more than 900 chip customers in China, it is much easier to collect and audit the accounts of foundries than chip companies.
In addition, we have continuously licensed our technologies to foundries for new process nodes and new applications and continuing tape-outs through end customers' adoption. The contract will be breached if the foundry fails to pay the royalties. As a result, all customers' tape- outs in the pipeline cannot go for production. The loss to the foundry will be much bigger than not paying royalties, not to mention the infringement of lawsuits penalty. The risk of embedded NVM is that there is a problem with memory reading after production, the entire chip or even the system will shut down, and the resulting penalty for damage will be much bigger, so customers don't want to risk illegally obtaining IPs.
Why is your royalty ratio of revenue much higher than other IP companies, and why has it been like this for such a long time?
-
The majority of IP companies mainly collect licensing fees but with no or little royalty. Our strategy is to charge less on licensing in the early stage but insist on collecting royalties.
What's your view on the penetration rate over the next few years?
-
We foresee an increase in foundry penetration rate due to our extensive development in technology platforms across multiple foundries. More than 100 process platforms are under development each year, with an average of 400 new product tape-outs growing to an average of 600 new product tape-outs per year, designing in our OTP, MTP and PUF-related security technologies.
China's mature manufacturing processes are severely oversupplied. Will foundry wafer prices decline in the future and affect your royalty revenue?
-
The OTP in mature processes is a very stable cash cow for us due to limited R&D resources. The leading edge and additional royalties from PUF-related security solutions will drive the ASP of future royalties. On the other hand, because of overcapacity, foundries are allocating resources to develop specialty processes to add value and differentiation, such as our MTP-related technology, which witnessed substantial licensing growth this year. The royalty rate for MTP-related technologies is double that of OTP, which will also increase the penetration rate of mature processes and the growth of royalties per wafer.
eMemory has announced the purchase of new offices. Does this mean you also have major employee expansion plans in the works?
-
We've been using our current office for over 15 years, but our employee count has increased by more than 50%, so the existing office space is not enough. We acquired a new office to provide employees with a better working environment and be prepared for expansion in the future.
As foundries begin to build outside of Taiwan, will your expenses increase?
-
Our R&D is still based in Taiwan and will not expand along with foundry expansion overseas. This is because we provide technology and design licensing, which does not require on-site execution in local foundries.
Leading foundry customer T mentioned that the 7nm revenue declined in 3Q2023 in their earnings. Will this affect eMemory's subsequent royalty contribution?
-
Our 7nm applications in leading foundry customer T are just entering mass production but are independent of the applications currently in production at leading foundry customer T. Furthermore, leading foundry customer T is actively developing mixed mode-related technologies like RF to increase the foundry utilization rate of 7nm, which will drive more applications to adopt our IPs.
What's your conversion rate (hit rate) from new tape-outs to royalty revenue?
-
The conversion rate from tape-out to mass production exceeds 95%. We have a high conversion rate because most of our tape-outs come from the mature applications of big companies. Almost all of these tape-outs will successfully move to production, resulting in royalty revenue.
Do you only license your IPs to China and Taiwan? Do you have US customers like Intel and GF?
-
Approximately 30 foundries worldwide have licensed our technologies, including Intel and GF.
Application
Charles mentioned that your IPs play an important role in memory. Since memory is a huge market, how are you progressing in this area?
-
Our DRAM royalties will increase significantly, driven by the increase in production of existing customers and new customers’ ramping up production.
You mentioned that existing customer applications are moving to more advanced process nodes. Which applications have experienced the migration to advanced nodes?
-
Taking OLED for example, the main process node in recent years has been 40nm. However, this year marks a significant move towards 28nm production, with future plans for development in 16/17nm. In the case of PMIC for smartphones, which were previously mainly in 8-inch, some will move to 55nm production. Similarly, TV SoC used to be in 12/14nm. One of our customers took the lead to move to 6nm to gain the market share from 12/14nm players. These transitions are all driven by the need for customers to incorporate additional features, achieving faster speeds, and reducing power consumption to enhance product competitiveness.
What kind of advantages does eMemory have in developing RRAM and MRAM?
-
Our strengths lie in our strong circuit design and device development team. With years of accumulated design experience, numerous patents, and a profound understanding of customer needs, we excel in quickly assisting foundries in customizing specifications for RRAM/MRAM. Our expertise extends to designing IPs tailored to meet customer requirements, and we are able to quickly complete qualification on foundry processes so that customers can use our IPs in their designs and move towards mass production. We have already completed verification at major foundries, and customers have begun to adopt our IPs in their chip designs. Currently, we are moving towards advanced processes.
Samsung is your largest end-chip customer. Does Samsung’s S24 mobile phone use your IPs? Will their memory-related products also include your IPs?
-
Currently, our IPs are adopted in ISP, OLED DDI. In the future, more applications will also adopt our IPs, such as PMIC. Memory-related has already adopted our technologies for the DDR5 module-related chips (in SPD hub and PMIC). In addition, we are developing embedded memory-related technologies in their foundries, which will gradually move its way into more products.
TSMC's latest A16 (1.6nm) process technology, featuring Super Power Rail, will relocate the power grid to the backside of the wafer. Will this have an impact on your technologies?
-
Backside power primarily aims to achieve two objectives in processes below 2nm: 1) Improving power efficiency, and 2) Reducing the number of backend layers to lower manufacturing costs. This has no impact on the use of our IPs.
Do you have cloud service customers?
-
Major cloud service providers are our customers already, using our IP across different product lines. These major players have strict requirements for IP vendors, which we have qualified for many years. Therefore, our IPs can be used in chips of different functionalities, whether in AI servers or edge devices.
Do you have any solutions for chiplet-related?
-
eMemory's IPs have been adopted across various applications, which will be integrated into a packaged module as chiplets:
1) In Digital ICs, for storing keys, ROM code, and repairing high-density SRAM.
2) In Analog ICs, for circuit repair and code storage.
3) In DRAM ICs, for DRAM repair functions.
4) In NAND modules, for storing keys for confidential computing.
In the future development of chiplets, we anticipate that more chips in the chiplets will incorporate our IPs instead of just one chip.
Do you have any solutions for AI applications?
-
There are three aspects to this question:
1) AI applications need a hardware root of trust for protecting assets and ensuring operational security. We offer several security solutions tailored to address these specific security concerns related to AI applications.
2) The computational demands of AI applications require substantial SRAM. Our comprehensive SRAM repair solutions are designed to improve the yield of AI chips and reduce production costs.
3) AI applications rely heavily on a large scale of DRAM. Our complete DRAM repair solution can help meet the demands of AI chips in high-density memory applications.
When will OLED DDI 28nm mass production begin?
-
As early as four years ago, our OTP was adopted in 28nm OLED DDI products by many customers. These products have gradually entered production and continue to contribute to royalties. The penetration rate of OLED DDI in mobile, tablets and notebooks keep on rising and moving towards more advanced processes, driving the royalty growth. Currently, some customers are moving to 22nm OLED DDI, accumulating good production records, and are developing products for the next generation of FinFET HV processes. We are optimistic about our continued growth in this market.
The chairman mentioned in previous quarters that the SRAM density is increasing, providing more opportunities for your IPs. What is your progress in this area? When can we expect to see contributions?
-
As the demand for AI and HPC increases, the density of embedded SRAM is also growing. Therefore, the demand for using OTP for SRAM repair continues to rise. We have seen an increase in the adoption of our solutions by more customers and products, resulting in more royalty, which we foresee will be a trend and the contribution will increase significantly in the future.
What is the application of your first 3nm customer? When will we see contribution?
-
Our first 3nm customers are primarily in Data Center server applications. They require not only OTP, but also PUF-based security to protect data. They will contribute to our revenue in the second half of this year.
What is your progress in DRAM?
-
Our OTP is mainly used for repair function in DRAM. It has already been implemented on multiple process platforms of several customers and is continuously being developed for more advanced processes. Recently, with the mass production of new processes of DRAM customers, the royalties have increased significantly.
MTP’s licensing fees and royalties have experienced the most significant growth. In which application is it mainly used? How do we expect MTP to contribute to the company's future?
-
Our MTP technologies include comprehensive product lines that meet different specifications, ranging from high endurance NeoEE and medium-density NeoMTP which require no additional masks, to high-density embedded flash such as NeoFlash/RRAM. The recent growth in MTP is driven by the adoption of NeoEE in PMIC and SPD ICs on DDR5, as well as NeoMTP in new four-color e-paper and electronic shelf label (ESL) driver ICs. Currently, technology licensing to foundries and the adoption across different applications are accelerating. The royalty rate of MTP is higher than OTP, so its contribution to our revenue will become increasingly significant.
What is the progress in 2nm and 3nm? Is it possible that after eMemory’s IP is ready for 3nm, customers will already start to migrate to 2nm?
-
Our development of 3nm and 2nm has always aligned with the requirements of the most advanced foundry processes and customers. Currently, some customers have already started design-in for 3nm, while 2nm is in the early stages of development, progressing much faster than previous years. Even if our technology is not adopted by the first generation, there will still be future opportunities due to process migration. For example, even in mature nodes, we still have hundreds of new products using our technologies every year.
Are you seeing a trend of customers incorporating AI capabilities into their designs? How does this impact your company?
-
As far as we know, we already have customers who have incorporated AI functions into their designs and are moving to tape out in more advanced processes for functions such as recognition and image processing. As mentioned in the last earnings, AI application systems involve data input, data/model storage and computing (accelerators). Currently, our IPs are adopted for applications related to data input through various sensors. For data storage in NAND/DRAM, customers are using our IPs for SSD controller and CXL memory interface. In terms of computing, which is mainly in advanced processes, we have customers adopting our Root of Trust IPs and SRAM repair IPs. These developments are expected to drive our future licensing and royalty growth.
Can eMemory IP work for 2 and 3nm? Why is the progress for advanced process adoption slower? Why haven't customers started adopting it? What do the most advanced processes customers currently use?
-
Foundry usually begins its development for most leading process with its own foundation IP, typically with its own OTP solution, called eFuse. Most advanced processes in the past have used eFuse for parameter setting, ID, or other functions that require OTP.
For our IPs to be provided to customers, we need to pass all the verifications foundries require, which typically takes 1-2 years. Once our IPs are qualified, it’s usually already too late for the most leading edge customers. For example, our design activities just significantly increased in 6/7nm after our IPs were qualified with customer production records. Most customer applications we engage with now migrated from 12/16nm, and most used eFuse in their previous generations. The reason customers replace eFuse with our IPs is to meet their density or security requirements because eFuse can only write 4K bits at most, while our IPs can accommodate densities a hundred times or higher. Even worse, eFuse can be easily been hacked by reverse engineering.
Due to the increasing demand for security and much higher storage density requirements, customers are constantly requesting our IP solutions in the most leading edge processes.
Moreover, our cooperation with Arm's Confidential Computing Architecture can be adopted by the most leading edge customers. We are working with customers and foundries to accelerate our qualification progress and hope we can catch up in the future.
Can you update us on the progress of MTP? What applications will require MTP?
-
The licensing of MTP dropped last year was mainly due to a one-time big license payment from China MRAM fab in 2022 and it will take a few years to commercialize before it can contribute to royalty. In terms of royalties, we have already seen significant growth in MTP due to mass production of applications over the previous years.
What is the impact of the latest US sanctions on the company?
-
The latest sanctions mainly target HPC-related products for AI applications. We currently have no Chinese customers in this area, as mostly our leading-edge customers are US and Japan-based. Also, since we develop our intellectual property rather than licensing US technology, we are not directly affected by the sanctions.The latest sanctions mainly target HPC-related products for AI applications. We currently have no Chinese customers in this area, as mostly our leading-edge customers are US and Japan-based. Also, since we develop our intellectual property rather than licensing US technology, we are not directly affected by the sanctions.
Why has eMemory’s technology become a standard for DDI and PMIC, but not for other applications? What is the alternative for other applications?
-
DDI and PMIC are analog circuit designs. Because there are variations in the production process of fabs, to control the output signal accurately and meet the specification, they need our OTP/MTP IP to perform output tuning, function setting, post-test parameter storage and code update.
These are already a standard in DDI and PMIC designs. We expect more analog and mixed mode application to follow.
In 2023, the PMIC market is quite sluggish , coupled with TI’s pricing competition, is TI a customer of the company? If not, will their price competition grab orders from existing PMIC company customers, affecting the company's future royalty income?
-
TI is not currently our customer. Despite this, we have many PMIC customers, including leading PMIC provider D and non-traditional PMIC companies, mainly using our IPs for self-made PMICs in mobile or bundled with processors. The annual wafer production volume of this type of customer is 700,000 to 800,000 pieces, increasing yearly and replacing chips from traditional IDM, such as TI.
TI’s price-cutting competition will not affect the orders of existing PMIC customers such companies. Since they are all top-tier customers of foundries, they also have many other advanced process-related products in production other than PMIC. Their wafer prices are also the lowest in the market, even as foundries raised the prices in the past few years, the prices of these customers remained the same. Furthermore, their PMIC products are primarily used in-house or in bundled sales with processors, so they are not affected by price-cutting competition.
In terms of royalties, it will mainly be affected by end-market demand and inventory adjustments and will vary depending on the situation of each customer. We believe that PMIC royalties will grow strongly. In addition to the low base due to inventory adjustment, we have successfully introduced our IPs to the PMIC of major Korean mobile phone manufacturer, DDR5 and other customers and applications such as EV cars, data center, and wearable. Our major US customer’s PMIC is also migrating from 8-inch to 12-inch 55nm, which has led to an increase in royalties per chip.
Do you have any AI processor customers?
-
The definition of AI is very broad. I think what most people care about now is the large-scale language training models made by leading GPU/CPU providers N and A, both of which have contacted us in the past. Leading GPU provider N started reaching out in 2020, and we have been discussing how to implement our OTP and PUF into their security architecture. The reason for not being adopted was that our 4/5 nm did not complete the qualification at that time.
For the key storage of Hardware Root of Trust (HRoT), they currently use eFuse. As I mentioned earlier, eFuse is not safe for key storage. OTP or PUF+OTP must be used to provide keys secure storage.
For key generation of Hardware Root of Trust (HRoT), they currently use the conventional True Random Number Generator (TRNG) without PUF, so the entropy and speed of key generation is inferior to what we provide in PUF-based True Random Number Generator (TRNG). In fact, our randomness is about 100 times theirs, they also consume about 100 times the power of ours.
Especially in the future, the length of the key required by post-quantum algorithm will be greatly increased, which will also increase the calculation time. Having high-speed key generation such as ours is very advantageous. We are very confident that with the progress of our 3nm and more production records, we have great opportunities in the future.
In addition to these two major chip suppliers, we also have cloud customers doing in-house AI processor in China and US this year. Furthermore, if AI, in a broad sense, also covers edge computing, then we already have many customers. Our solution is only a small area and can generate basic security functions in a simple way, which is very competitive in edge computing.
Which AI server-related applications have adopted the company's IP? Does this include CPU and GPU?
-
Currently, the applications inside the AI server adopting our IPs with tape-outs or in production are: SSD controllers, CXL memory controllers, Retimers, PMIC and SPD for DDR 5 DIMM card. The CPU and GPU have not yet adopted our IP, but we have a good chance when we complete 3nm verification, especially for those who adopt Arm's solution.
Does eMemory do 12nm ISP? In addition, if the OTP of the ISP rises from 2K to 16K, will the output value of royalty be multiplied?
-
Some customers have adopted our IP in 12nm ISPs with small volume production. Since the price of 12nm wafer is 40% higher than that of 22/28nm, if existing 22/28nm customers switch to 12nm, chip size remains the same, then royalty difference will be the increase of wafer pricing, which has nothing to do with how much density is used.
Under the CXL framework, what role do eMemory IPs play?
-
CXL is the protocol to enhance the memory capacity sharing between CPU and other devices, such as GPU. Our CXL Memory eXpander Controller customers adopt our Root of Trust to protect the security of data confidentiality and integrity during data transfer.
Other than PMIC, do other parts of DDR5 also need to use your IPs? Ex: RCD(register clock driver), DB(Data Buffer), SPD Hub, etc...
-
RCD and DB currently do not use. But SPD Hub has adopted NeoEE for 100K endurance. So, inside DDR5 DIMM card, there are two chips SPD hub and PMIC with our MTP solution.
eFuse cannot shrink below 28nm. However, in terms of the advanced manufacturing process, if the customer's eFuse is accidentally blown out, what kind of solution will be adopted if the eMemory solution has not yet been adopted?
-
If the customer uses eFuse and it blows out accidentally, they will usually increase the eFuse area, however, this will take up too much space and would be very expensive for advanced process chips. In addition, eFuse cannot accommodate circuit design that needs to burn a large current. However, since eFuse is provided by the foundry, if there is a problem, the foundry will be responsible, so eFuse is still dominating in advanced manufacturing processes. Another type of technology is anti-fuse, mainly our NeoFuse and competitor’s solution. However, the IP of competitors has experienced problems in advanced manufacturing processes. At present, not heard any customer adopt their solution in leading processes. But it also affects customers' confidence and slows down the adoption of this technology. With more production record and successfully moving into 5nm and 3nm, we are confident to replacing eFuse is only a matter of time.
Is there any progress on SRAM repair?
-
Multiple customers are working with us, and we are collaborating with the EDA tool company to incorporate our OTP into their SRAM compiler to facilitate customers’ adoption.
Are your IPs used in chiplets? What is its current proportion of revenue?
-
Customers use our OTP for multi-chip repair because traditional eFuse is unsuitable for post-packaging modifications. For example, in recent years, ISPs that need to be packaged with DRAM and CIS are driving customers to use our OTP for ISPs. As a result, ISP has contributed to over 10% of our royalties this year.
Furthermore, DRAM manufacturers have also licensed our technology for DRAM repair and after 3D or 2.5D chip packaging with logic chips. We expect more adoptions of our IPs for similar applications in the future.
Will the US government's latest ban on China's AI chips affect your business?
-
We do not have any impact since the banned chips don't use our IP yet.
As generative AI proliferates on edge devices, will this increase the adoption of eMemory IPs?
-
Edge devices all need security protection so that they can securely connect to cloud services. Particularly, with generative AI embedded in edge devices, the AI models and training data must also be protected. Our PUF-based security IPs will have a big market in edge devices.
Will changes in GAA (Gate All Around) and Backside Power Delivery Network affect your technologies?
-
No, this will not affect us. We are developing our IP in related processes.
What opportunities or threats do you face in the design of processors such as CPU/GPU/NPU?
-
Future computing processors will move toward Confidential Computing, which requires high-level security functions. As we have successfully developed PUF-based Root of Trust security IP, which provides very strong security functions for these processors, we have a great opportunity to license these IPs to these Confidential Computing processors. Besides, in order to enhance the computation performance, these processors all need high-density SRAM, our NeoFuse OTP plays an important role for SRAM repair. We expect our IPs to cover a huge market in these processor-related applications.
In the past 6 years, NeoMTP's licensing revenue was more than $400 million NTD. In 2022, it was $90 million NTD. In 2021, it even reached $140 million NTD. However, in the past 6 years, NeoMTP's royalty revenue was less than $200 million NTD. Can you explain the gap? What will NeoMTP's royalty growth look like next year, particularly with your Korean business?
-
The licensing of MTP covers five technologies: NeoEE, NeoMTP, NeoFlash, RRAM and MRAM. Over the past few years, MTP's licensing revenue mainly came from foundries. MTP's development and verification period is relatively long, typically exceeding two years. As a result, it takes at least five years before customers even move from tape-out, to production, to royalty contribution. However, since these technologies have already been adopted into mainstream applications, we anticipate accelerated growth in royalty in the future.
Regarding Charles' talk, have you seen any customer adoption for SRAM repair?
-
We have AI-related tape-out, which uses our OTP for SRAM repair due to high SRAM density. With the increasing SRAM density in AI and HPC processors, we foresee increasing demand of our OTP for SRAM repair.
What is your progress on 3nm and 5nm?
-
This year, 5nm technology will be adopted in self-driving, data center, and AI-related applications. Meanwhile, 3nm technology is undergoing verification at major foundries. In addition to our collaboration with CPU partners, we already have ADAS and processor customers, as well as a US cloud service provider requesting our IP for 3nm. We are confident about the verification process and anticipate customer design-ins next year.