Buried deep within ASMPT's recent earnings call transcript lies a revelation. It's not a flashy new product, a sudden surge in orders, or a bold prediction about the semiconductor market. It's a subtle shift in language, a change in tone that hints at a potentially explosive growth trajectory for one of their core technologies: Thermal Compression Bonding (TCB).

ASMPT, a global leader in semiconductor and electronics assembly equipment, has been cautiously optimistic about TCB's potential for years. The technology, which offers a precise and cost-effective way to connect chips to substrates or wafers, has long been seen as a key enabler for advanced packaging, especially in high-performance computing (HPC) and artificial intelligence (AI) applications.

However, in this latest earnings call, there's a palpable sense of excitement, a newfound confidence in TCB's ability to not just participate in, but dominate the advanced packaging landscape. The language has shifted from "potential" to "momentum," from "engaging with customers" to "winning orders." The narrative, once focused on future possibilities, is now grounded in present realities.

This shift in tone isn't unfounded. ASMPT has secured meaningful TCB orders from a leading foundry for chip-to-substrate applications, a testament to the technology's ability to meet the demanding requirements of 2.5D packaging for generative AI. But it's the discussion around chip-to-wafer applications that truly reveals the magnitude of the opportunity ahead.

Currently, chip-to-wafer bonding relies primarily on mass reflow (MR) technology, a process that faces limitations as the demand for finer pitch and higher placement accuracy intensifies. ASMPT believes TCB, with its inherent advantages in accuracy and thin die handling, is poised to supplant MR as the go-to solution for chip-to-wafer integration in high-end AI chips.

This isn't just a technological leap; it's a potential seismic shift in market dynamics. The number of TCB tools required at the wafer level far surpasses the demand for chip-to-substrate applications. Robin Ng, ASMPT's Group CEO, highlighted this point, stating that "if you compare just chip to wafer and chip-to-substrate, the number of TCB tools in [our] opinion will have to be higher at a chip to wafer level compared to chip-to-substrate."

The implication is clear: ASMPT, already a market leader in TCB with the largest installed base globally, is positioned to ride a wave of demand that could dwarf its existing business. The company has stated that they expect TCB revenue from 2022 to 2024 to equal the total revenue generated from the technology in the preceding decade. But the recent wins and the clear articulation of the chip-to-wafer opportunity suggest a potential for even greater acceleration.

The excitement doesn't end with logic chips. ASMPT's TCB tools are already in production with a leading High Bandwidth Memory (HBM) player, and the company anticipates significant order flow from multiple HBM players over the next few quarters. As HBM stacking requirements escalate, moving from 8-high to 12-high and beyond, the need for TCB's precision and thin die handling capabilities becomes paramount.

This potential inflection point for TCB adoption aligns perfectly with the projected growth trajectory of the advanced packaging market. ASMPT estimates that the addressable market for AP will surge from $1.7 billion in 2024 to $3.3 billion in 2028, representing a CAGR of 18%. If TCB, as ASMPT believes, becomes the dominant interconnect technology for high-end AI and HPC chips, capturing a significant share of this market could propel the company's revenue to unprecedented heights.

While other analysts may focus on the immediate recovery of the semiconductor market or the near-term order momentum for AP solutions, the astute investor will recognize the significance of ASMPT's changing narrative around TCB. The subtle shift in language, the tangible order wins, and the explicit focus on the chip-to-wafer opportunity all point to a technology poised for explosive growth.

ASMPT, founded in 1975 as a small trading company in Hong Kong, has a history of innovation and adapting to industry shifts. Could their mastery of TCB be the key to unlocking a 10x return for investors? Only time will tell, but the signs are compelling, and the opportunity is undeniable.

Hypothesis:

Based on the transcript and ASMPT's projections, we hypothesize that TCB revenue will grow at a CAGR exceeding the 18% projected for the overall advanced packaging market. We estimate that TCB could account for over 30% of ASMPT's total revenue by 2028, driven by the rapid adoption in chip-to-wafer applications for both logic and HBM.

Supporting Numbers:

This combination of factors suggests a strong likelihood that TCB will outperform the overall AP market and become a major revenue driver for ASMPT, potentially driving significant shareholder value in the years to come.

TCB Revenue Projection

The following chart represents a hypothetical projection of ASMPT's TCB revenue based on the company's statements and market projections.

ASMPT Financial Snapshot

"Fun Fact: The world's first microprocessor, the Intel 4004, was released in 1971 and contained 2,300 transistors. Today's advanced processors can have over 100 billion transistors, showcasing the incredible advancements in semiconductor technology that companies like ASMPT enable."