Intel Outsources Latest CPU Production to TSMC; First time in History

The compute tile will reportedly use TSMC's 3nm-class N3B process technology, contrary to Intel's initial plan of using their in-house 18A (1.8nm-class) process for Lunar Lake CPUs.

Introduction:

Intel has taken a significant step by deciding to outsource its computing cores to TSMC for Lunar Lake MX processor.

This move marks the first time Intel has opted for external manufacturing for its compute cores. This reflects the company’s strategic decision to leverage TSMC’s advanced manufacturing capabilities.

In this blog post, we’ll delve into the details of this collaboration and analyze the insights shared by renowned analyst Lu Xingzhi, shedding light on the potential implications for both Intel and the semiconductor industry as a whole.

Join Our WhatsApp News for updated information on semiconductors & AI

Background of Intel outsource TSMC

Intel’s x86 chips are venturing into uncharted territory, as leaked slides shared by prominent hardware leaker @YuuKi_AnS (quickly removed) reveal that the compute tile of the Lunar Lake MX processors will be manufactured using TSMC’s N3B fabrication technology.

This marks a significant departure for Intel, as it’s the first time the company has opted for outsourced process node technology for its top-tier x86 cores.

The Lunar Lake processors are slated to introduce a fresh microarchitecture designed to deliver breakthrough performance-per-watt efficiency, primarily targeting mobile devices.

However, the timing of the slides is uncertain, and they may represent an older plan for the series.

According to the leaked information, the Lunar Lake MX platform lineup will boast processors with up to eight general-purpose cores. This comprises four high-performance Lion Cove cores and four energy-efficient Skymont cores.

The specifications include a 12MB cache, up to eight Xe2 GPU clusters, and a six-tile NPU 4.0 AI accelerator. Depending on the power target, the platform will support 8W fanless designs as well as 17W to 30W fanned configurations.

The compute tile will reportedly use TSMC’s 3nm-class N3B process technology, contrary to Intel’s initial plan of using their in-house 18A (1.8nm-class) process for Lunar Lake CPUs.

This shift represents a departure from Intel’s usual in-house production of computing cores, showcasing the company’s acknowledgment of TSMC’s expertise in advanced semiconductor manufacturing.

Read More: 5 Key Announcements From Intel Innovation Day

Why is Intel Outsourcing to TSMC?

The utilization of TSMC’s N3B process technology by Intel for the construction of its Lunar Lake MX processors might seem unusual, but it isn’t entirely surprising.

This decision aligns with Intel’s goal of integrating CPU and GPU cores on the same silicon, making it more practical to adopt TSMC’s N3B for the entire design.

The rationale behind this choice lies in the fact that GPUs typically have larger sizes than CPUs, and adapting the Xe2 GPU for Intel’s 18A node might necessitate more time than the company is willing to allocate for its processor.

Despite the unconventional move, this marks the first instance of Intel incorporating a third-party process technology for one of its flagship CPUs. It underscores the adaptability inherent in Intel’s IDM 2.0 approach to design and manufacturing.

Leaked Slides(Check the blue line)

Read More: Intel outsources chip manufacturing to TSMC worth $19 Billion

Insights from Analyst Lu Xingzhi:

Lu Xingzhi, a respected foreign analyst, draws attention to the significance of Intel’s decision to seek TSMC’s assistance. Xingzhi likens Intel’s reliance on TSMC to a dependency, stating that once a company experiences the benefits of TSMC’s manufacturing, it becomes challenging to revert to previous practices.

He further breaks down his analysis into key points:

Capacity Expansion:

TSMC is reportedly preparing to offer 15k/m^3 nanometer capacity by the end of next yea. This increases to 30k/m^3 nanometer capacity in the following year.

This surge in capacity positions Intel as one of TSMC’s top three clients by 2025. 3 nanometer process is making up a significant portion of their collaboration.

Financial Impact on TSMC:

Xingzhi suggests that TSMC’s collaboration with Intel could contribute to 5% annual revenue growth momentum for TSMC, reaching 8% and 12% in the next two years, respectively.

Intel’s Reliance on TSMC:

Despite Intel’s outward projection of self-sufficiency, Xingzhi notes that the introduction of compute tiles in Lunar Lake indicates Intel’s growing dependence on TSMC for manufacturing.

Benefits for Intel:

Xingzhi highlights several benefits for Intel in outsourcing to TSMC. These include gaining access to advanced manufacturing capacity, cutting production costs, lowering R&D expenses, reducing capital expenditures, and achieving more competitive product pricing.

Read More: Intel Will Reclaim Foundry Leadership from TSMC by 2025: CEO

Conclusion:

Intel’s decision to outsource compute cores to TSMC is a strategic move with far-reaching implications. The collaboration not only addresses Intel’s immediate manufacturing needs but also positions TSMC as a key player in the semiconductor industry.

Xingzhi’s insights underscore the transformative impact of this partnership, with potential consequences for competitors, particularly AMD.

Reference:

  1. Tom’s Hardware
  2. Technews-Tw

Kumar Priyadarshi
Kumar Priyadarshi

Kumar Priyadarshi is a prominent figure in the world of technology and semiconductors. With a deep passion for innovation and a keen understanding of the intricacies of the semiconductor industry, Kumar has established himself as a thought leader and expert in the field. He is the founder of Techovedas, India’s first semiconductor and AI tech media company, where he shares insights, analysis, and trends related to the semiconductor and AI industries.

Kumar Joined IISER Pune after qualifying IIT-JEE in 2012. In his 5th year, he travelled to Singapore for his master’s thesis which yielded a Research Paper in ACS Nano. Kumar Joined Global Foundries as a process Engineer in Singapore working at 40 nm Process node. He couldn’t find joy working in the fab and moved to India. Working as a scientist at IIT Bombay as Senior Scientist, Kumar Led the team which built India’s 1st Memory Chip with Semiconductor Lab (SCL)

Articles: 2147