The future of OSATS

An article by Favier Shoo and Santosh Kumar from Yole Développement (Yole), for Chip Scale Review | Over the past few decades, the integrated circuit (IC) industry has been relying on traditional chip scaling and innovative architectures for new devices, closely following Moore’s Law. In die-level scaling, the idea is to pack ever more transistors on a monolithic die or system on a chip (SoC) with the development of each front-end process node, enabling faster chips with a lower cost per transistor. With traditional die-level scaling, however, the design cost has risen many times (e.g., 3nm design costs ~35-40X more compared to 90nm) and manufacturing has become extremely complex, leading to an increase in time to market. Therefore, it is widely acknowledged that Moore’s Law is slowing down, if not dead.

While front-end scaling remains a work in progress in the background, the industry has been diligently leveraging on packaging technologies to improve system-level interconnection density and reduce the cost and time to market by scaling the package-level pitch and integrating more functions into a single package. Packaging revenue has increased 70x from US$1 billion in 1970, to an expected US$68 billion in 2019. At the same time, the scaling gap between front end and back end, which has increased from 50x to 600x from 1970 to 2019, has been significantly narrowed down by advanced packaging.

The OSAT business model is to provide third-party IC packaging and test services while remaining in the center of the semiconductor and packaging worlds. Therefore, OSATS remain the essential pillar between the widening technology gap of front-end and back-end scaling. Especially with the growing importance of advanced packaging, OSATS’ innovation and supply chain positions are now key to on-going system-level performance increases.

Packaging market forecasts and key trends

The total IC packaging market was worth US$68 billion in 2019. The advanced packaging (AP) market was US$38 billion in 2019 and is expected to grow at a CAGR2019-2025 of 6.6% to reach US$42 billion in 2025 (Figure 2). At the same time, the traditional packaging market will grow at a CAGR2019-2025 of 1.9%, and the total packaging market will grow at a CAGR2019-2025 of 4% to reach US$43 billion and US$85 billion, respectively. Because of the COVID-19 pandemic, the AP market will decrease by 7% in 2020, while the traditional packaging market will decrease by 15%.

Growing at a CAGR2014-2025 of 6.1%, the AP market is expected to more than double its revenue from US$20 billion in 2014 to ~US$4 billion in 2025. Compared to traditional packaging, which will grow at a 2.2% CAGR2014-2025, AP will grow almost three times faster with a CAGR2014-2025 of 6.1%. In 2019, the AP share of the total packaging market was 42.6% [1]. Because of the strong momentum in the AP market driven by a slowing Moore’s Law and heterogeneous integration, along with various megatrends (5G, artificial intelligence [AI], high-performance computing [HPC], Internet of Things [IoTs], etc.), the share of AP in the total packaging market is increasing continuously, and it will reach almost 50% of the market by 2025… Full article