However, the company has now announced that it is teaming up with TSMC to develop a 32-core ARM processor, which uses Arm’s Neoverse computing subsystem technology and claims to be “capable of operating in ultra-large-scale data centers and next-generation mobile infrastructure (including 5G and 6G). Delivering ‘scalable performance'”.

▲ Picture source Socionext official press release

After inquiries, it is learned that Neoverse computing subsystem technology is a pre-integrated and pre-verified computing platform designed to simplify the chip customization process. Chipsets are available with single or multiple “instances” within a single package, as well as for IO and specific application needs.

Socionext will be built on TSMC’s 2nm process and is said to “integrate seamlessly with customers’ existing designs.” And is expected to be compatible with other ARM chips.

Socionext also claimed that due to the reusability of the chip, it can be adapted to a variety of product platforms, and system architects can also explore new possibilities, and ARM is said to be “also happy to be a part of this partnership” a part of”.

THIS IS A SPONSOR PROMOTION: >>>>>>>>>>>>>

Geekwills is an online shop that connects consumers with millions of products and brands around the world with the mission to empower them to live their best lives. Geekwills is committed to offering the most affordable quality products to enable consumers and sellers to fulfill their dreams in an inclusive environment.

The post Socionext Announces Collaboration with TSMC to Develop 2nm ARM Processor appeared first on TechGoing.

TSMC confirmed at the press conference in April that the Kaohsiung plant had adjusted its 28nm production line to a more advanced process technology, but did not mention what process it would change to at the time. TSMC’s current 2nm production bases have planned Zhuke and Zhongke. If Kaohsiung is included in the future, the company will have three 2nm production bases.

In addition, according to Taiwan’s “Central News Agency”, TSMC’s planned 2nm process will be mass-produced in 2025, using a nanosheet transistor structure. At the same time, TSMC has developed a backside rail solution at 2nm, which is suitable for applications related to high-performance computing. It is targeted to be launched in the second half of 2025 and mass-produced in 2026.

▲ Picture source TSMC official website

According to previous reports, TSMC also announced that it will jointly invest in European Semiconductor in Dresden, Germany with Robert Bosch GmbH, Infineon Technologies AG and NXP Semiconductors NV. Manufacturing Company (ESMC), with an estimated total investment of more than EUR 10 billion, to provide advanced semiconductor manufacturing services.

The fab is scheduled to be built in the second half of 2024 and produced by the end of 2027. It will be operated by TSMC, which will hold 70% of the shares, while Bosch, Infineon and NXP will each hold 10% of the shares. After completion, it is expected to produce 40,000 300mm (12-inch) wafers per month, using TSMC’s 28/22nm planar CMOS and 16/12nm FinFET process technologies.

The post TSMC’s 2-nm factory is planned to grow to three, mass production is expected in 2025 appeared first on TechGoing.

However, it should be noted that TSMC will hold a law conference on July 20 (Note: Performance Briefing Session for Legal Person Investment Institutions), and the company is currently in a “quiet period before the conference” and will not make any comments on this matter.

Sources pointed out that TSMC’s 2nm requires more investment than the originally planned 28nm. The company has reported to the local area and expressed its hope for assistance in water supply and power supply. Because the 2nm process requires equipment that consumes more power, there is a certain chance that the environmental impact difference analysis will need to be completed again.

It is reported that TSMC’s 2nm process will be 10-15% faster than the 3nm process at the same power consumption, while power consumption can also be reduced by 20-30%. At present, the main production base of the 2nm process is located in Baoshan, Hsinchu, and plans to build 4 factories. It is estimated that risky trial production will begin in 2024 and mass production will begin in the second half of 2025.

It was previously reported that Apple has reserved 90% of TSMC’s 3nm process wafers for the production of A17 Bionic and M3 chips. However, the current yield rate for this advanced process is only 55%. TSMC expects to be able to produce 100,000 3nm wafers per month by the end of 2023 to meet Apple’s demand.

However, with a yield of only 55%, only 55,000 wafers are usable. Apple will only pay the standard wafer price when the yield rate reaches 70%, but according to reports, this is not likely to happen until the first half of 2024.

The post TSMC changes plan to build 28nm plant in Kaohsiung to mass-produce 2nm process in H2 2025 appeared first on TechGoing.

TSMC stated that it does not comment on relevant rumors, emphasizing that the development of 2nm technology is progressing smoothly, and mass production is expected in 2025.

According to sources, TSMC has also begun to dispatch engineers to the Zhuke R&D plant to prepare for the pre-preparation work for the 2nm trial production. It is expected to convene a R&D team of more than 1,000 people to lead the global mass production at Zhuke Baoshan Wafer Fab 20. 2nm. According to public information, TSMC’s most advanced 2nm production base in the future will first settle in Zhuke Baoshan Wafer 20 Factory. The plant area is planned for the fourth phase and will be expanded to Zhongke later, with a total of six phases of the project.

It is reported in the industry that TSMC will first establish a small-scale trial production line in Zhuke in the early stage of 2nm R&D. The goal is to trial produce nearly 1,000 pieces this year. The team will sprint to the risk trial production in 2024 and mass production in 2025.

It is reported that TSMC’s 2nm will adopt a new gate-around (GAA) transistor architecture for the first time. TSMC previously pointed out in a technical forum that the overall system performance of related new technologies has been greatly improved compared with 3nm, and the customer base’s willingness to invest in cooperative development in the early stage is much higher than that in the early stage of the 3nm family, and more customized solutions can be customized.

It is noticed that TSMC has previously released the 2nm family process blueprint plan, of which N2 will be mass-produced in 2025, and the backside rail design is tailored for high-speed computing (HPC) products, which will be launched in the second half of 2025; , N2P and N2X of the 2nm family are expected to be launched in 2026.

Nvidia CEO Jen-Hsun Huang previously publicly pointed out that TSMC will introduce the AI system that Nvidia has worked with other partners for nearly four years into the 2nm trial production operation, which is expected to start in June. cycle, mass production schedule, while reducing the manufacturing carbon footprint, and getting ready for 2nm and even more advanced processes.

The post TSMC has started preparatory work for 2nm trial production appeared first on TechGoing.

Since TSMC’s 3nm is not much different from 4nm under the performance of PPA, and the quotation of 3nm has risen to 20,000 US dollars, although only Apple can enjoy a 20% discount, but there are already many customers revised the process planning, adjusted the production and orders, including lengthening the 4/5nm generation cycle, slowing down the progress of N3E and N3P adoption, and waiting for the 2nm GAA process generation to re-deposit.

He believes that although TSMC’s production capacity layout may be disrupted, customers have a higher degree of stickiness and are quite confident in the 2nm GAA generation. Almost all customers who have adopted 4/3nm have 2nm production plans.

According to previous news, Apple has almost rounded up nearly 90% of TSMC’s first-generation 3-nanometer process capacity this year for future iPhone, Mac and iPad chips.

Apple’s upcoming iPhone 15 Pro series models are expected to feature the A17 Bionic processor, Apple’s first iPhone chip based on TSMC’s first-generation 3-nanometer process, also known as N3B. The 3-nanometer technology is said to be 35% more energy efficient and 15% more performant than the 4-nanometer process used to make the A16 Bionic chip in the iPhone 14 Pro and Pro Max.

Apple’s M3 chip for Macs and iPads is also expected to use a 3nm process. The first devices to feature the M3 are expected to include the new 13-inch MacBook Air and 24-inch iMac, both of which could arrive later this year. New iPad Pro models due next year may also feature the M3 chip, while Apple analyst Ming-Chi Kuo believes the 14- and 16-inch MacBook Pro models due in 2024 will feature the M3 Pro and M3 Max chips.

According to an App Store developer log obtained by Bloomberg’s Mark Gurman, Apple is currently testing a new chip with a 12-core CPU, 18-core GPU and 36GB of memory, which could be the next generation of 14-inch and The base M3 Pro chip found in the 16-inch MacBook Pro models.

According to DigiTimes, TSMC is also working on a more advanced 3nm process called N3E. Apple devices will eventually switch to the N3E process, which is expected to begin commercial production in the second half of 2023, but actual shipments will not ramp up until 2024.

The post TSMC’s almost all customers who have adopted 4/3nm have 2nm production plans appeared first on TechGoing.

TSMC Chairman Liu Deyin and President Wei Zhejia reviewed the development of TSMC in the past year in the business report to shareholders, and mentioned that TSMC’s technological development is preparing for the 2nm technology (N2), which is expected to start mass production in 2025. Technology will be produced in Hsinchu and Taichung Science Parks.

In terms of technology development, TSMC’s business report to shareholders stated that the company’s 2nm technology will use a nanosheet (Nanosheet) transistor structure and provide full-process performance and power efficiency benefits. Compared with N3E, N2 is 10%-15% faster at the same power consumption, or 25%-30% lower power consumption at the same speed, to meet the increasing demand for energy-saving computing.

TSMC also stated that the research and development of N2 technology is proceeding as planned, and it is expected to start trial production in 2024 and mass production in 2025.

The report also stated that TSMC will increase its research and development expenses to US$5.47 billion in 2022, and the 5nm family technology has entered the third year of mass production, contributing 26% to TSMC’s revenue. %.

The N4 process will start volume production in 2022. TSMC has also launched the N4P and N4X processes. The development of N4P is progressing smoothly and is expected to be mass-produced in 2023; N4X is TSMC’s first technology focused on high-performance computing (HPC) and high workloads and is expected to be designed by customers in 2023 finalized.

Following the mass production of the N3 process in 2022, N3E with better performance, power consumption and yield is expected to be mass-produced in the second half of 2023. TSMC said that N3 and N3E have very high customer participation, and the number of finalized product designs in the first and second years of mass production will be more than twice that of N5, and it is expected that the N3 family will become another large-scale and long-term demand for TSMC. process technology.

The post TSMC: The 2nm N2 process is expected to be mass-produced in 2025 appeared first on TechGoing.

According to official information from TSMC, TSMC’s 3nm process family includes four versions, namely the basic N3, the cost-optimized N3E, the performance-enhanced N3P, and the high-voltage tolerant N3X. Among them, N3E and N3P are optically scaled down versions based on N3, which can reduce complexity and cost while increasing performance and transistor density. N3X is a process specially designed for the HPC field, which can support higher voltage and frequency to achieve stronger computing power.

According to TSMC, compared with the 5nm process, N3E can reduce power consumption by 32% at the same frequency, or improve performance by 18% at the same power consumption. Compared with N3E, N3P can improve performance by 5% at the same power consumption, or reduce power consumption by 5%~10% at the same frequency. At the same time, N3P can also increase transistor density by 4%, reaching a level of 1.7 times that of the 5nm process.

The N3X is the most powerful version of TSMC’s 3nm process family. It can improve performance by 5% compared to N3P at the same power consumption, reaching a voltage level above 1.2 volts. This is very extreme for a 3nm-level process, and it also means that there will be high power consumption and heat generation problems. Therefore, this process is only suitable for processors that require extreme performance at the HPC level, and requires chip designers to take effective measures to control temperature and power consumption.

TSMC said that N3E will start mass production in the second half of 2023, while N3P and N3X will go into mass production in the second half of 2024 and 2025, respectively. These processes will use the FinFET structure, which is the metal oxide semiconductor field effect transistor (MOSFET) structure. This structure has been used by TSMC for many years, and it has mature and stable characteristics.

However, below the 2nm level, TSMC will adopt a new GAAFET structure, which is a Gate-All-Around FET structure. This structure can further increase transistor density and performance, and reduce power consumption and leakage. According to TSMC, the 2nm process is progressing well in terms of yield and component performance and will be mass-produced in 2025 as scheduled. According to TSMC, compared with N3E process technology, 2nm can increase the speed by up to 15% at the same power consumption; at the same speed, the power consumption can be reduced by up to 30%, while the chip density increases by more than 15%.

The post TSMC: 2nm process mass production in 2025, N3P mass production in the second half of 2024 appeared first on TechGoing.

TSMC’s 2nm fab is planned to be set up in Zhuche Baoshan and Zhongke respectively. TSMC President Wei Zhejia mentioned in the January meeting that TSMC’s 2nm progress is better than originally expected, maintaining the planning of risk trial production in 2024 and mass production in 2025, which is progressing well.

The second phase of TSMC’s 2nm fab expansion project in the Taichung campus of TSMC is planned for two phases and four fabs, with construction scheduled to start this year and the first 2nm fab to be put into operation by the end of 2024 at the earliest.

According to the financial data, TSMC’s consolidated revenue in February was about NT$163.174 billion (currently about RMB 36.714 billion), down 18.4% sequentially and up 11.1% year-on-year. Revenue for the first two months of the year was approximately NT$363.225 billion (currently approximately RMB81.726 billion), up 13.8% year-over-year.

The post TSMC’s Zhongke 2nm plant delivery was delayed again appeared first on TechGoing.

When questioned, TSMC CEO said that they are ready to mass produce 2nm chips in 2025. In addition, he said TSMC is considering building an automotive chip factory in Europe and is considering building a second factory in Japan.

Image source Unsplash

TSMC’s CFO revealed that research and development expenses are expected to increase by 20% in 2023; TSMC expects capital expenditures in 2023 to be US$32-36 billion, with an estimated US$34.86 billion. TSMC expects a mid- to high-single-digit decline in sales in the first half of the year, and expects year-over-year sales growth in U.S. dollars in the second half; a slight increase in U.S. dollars is expected in 2023.

TSMC’s operating income from January to December 2022 will be NT$2,263.89 billion (currently approximately RMB 502.584 billion), a year-on-year increase of 42.6%. TSMC expects first-quarter sales of US$16.7 billion to US$17.5 billion, which is currently about 113.059 billion to 118.475 billion RMB.

TSMC expects first-quarter sales of US$16.7 billion to US$17.5 billion, which is currently about 113.059 billion to 118.475 billion RMB.

According to a previous report by the Taiwan Economic Daily, the supply chain is rumored that TSMC is driven by the three major driving forces of advanced process expansion in Taiwan, investment in R&D, US and Japan expansion, and mature process upgrades due to the growing needs of the next three years. Expenditure is expected to approach 40 billion U.S. dollars (currently about 276.8 billion RMB), a new high.

The post TSMC CEO: ready for mass production of 2 nm chips in 2025 appeared first on TechGoing.

The second phase of the expansion plan of Taiwan’s Central Taiwan Science and Technology Taichung Park passed the environmental assessment panel review on the 5th of this month, the Central Taiwan Science and Technology Administration originally planned to strive to hand over the land for TSMC to build a new 2nm factory in May next year. TSMC is planning two phases and four fabs, with construction starting next year and the first 2nm process going into production by the end of 2024 at the earliest, with a total investment of NT$800 billion to NT$1 trillion.

According to TSMC’s information, the 2nm process will increase speed by 10% to 15% under the same power, or reduce power consumption by 25% to 30% under the same speed, compared to N3E. As for the more advanced 1nm process, TSMC has not yet announced the mass production schedule.

The post TSMC’s 2nm fab plans may be delayed as advanced process advancement is hampered appeared first on TechGoing.