Huawei overturns Moore's Law

Moores are aging, and the Tao King should rise.

Author: Zhang Bo, Duan Mingzhu, Miaotou APP

Tao (τ) Law has "exploded."

On May 25th, at the 2026 International Circuit and System Seminar, Huawei's board director and semiconductor business unit president He Tingbo officially announced the Tao (τ) Law, proposing to replace "geometric scaling" with "time scaling." In simple terms, the chip competition is no longer about who "makes smaller," but about who makes signals "run faster". This is the first time China has proposed a new guiding principle for industry development in the global semiconductor field.

The capital market is most敏锐, with Huahong company's 20cm wafer foundry concept hitting the daily limit-up in the afternoon, SMIC approaching the limit-up; advanced packaging concepts like Changdian Technology, Tongfu Microelectronics, and Huatian Technology all hitting the limit-up; a surge in tech stocks, driving the STAR 50 index up nearly 6%, reaching a new high.

In fact, the Tao Law cannot be seen as mere hype; it represents a profound restructuring of industry logic, which may lead to China's semiconductor industry "overtaking on a curve."

So, how should we truly understand the Tao Law? Which segments of technology are being re-evaluated first? Where can profits be more effectively accumulated?

Moores are aging, τ King should rise

The most valuable aspect of the Tao Law lies in its innovative solution to the choke points in China's chip competition.

Over the past decades, the global semiconductor industry has mainly developed along Moore's Law, centered on "geometric scaling," continuously shrinking transistors to improve chip performance and reduce costs.

But now, this law is reaching its boundary. When components are already approaching physical limits, further miniaturization not only becomes technically more difficult but also sharply increases costs. At this stage, relying solely on "making parts smaller" yields diminishing returns.

For China’s semiconductors, this problem is even more complex. Because advanced process nodes involve not just technical challenges but also constraints related to equipment, materials, processes, supply chains, and the international environment. In other words, the issue isn't just "can we continue shrinking," but also "at what cost" and "can we do so stably."

After Moore's Law fails, the industry has been seeking alternatives. Huawei's proposal of the Tao Law is essentially a different path.

τ is the time constant in circuit theory, representing the time needed for a signal to switch. The smaller τ is, the faster the chip runs. Moore's Law reduces τ by shrinking transistors; Tao Law, on the other hand, reduces τ by optimizing architecture, packaging, interconnects, software, and system collaboration.

To use an analogy, a chip is like a city, transistors are buildings, and signals are vehicles. Moore's Law is like building narrower roads and closer buildings to shorten distances; τ Law is not shrinking the buildings but redesigning the entire traffic system—building overpasses, setting up express lanes, optimizing traffic lights. Vehicles run faster, and city operation efficiency improves accordingly.

In simple terms, the Tao Law does not just focus on "more refined parts," but on making the "whole system smarter."

Huawei discussed at the conference that through logical folding, 3D integration, advanced packaging, and hardware-software collaboration, it is possible to achieve system performance close to advanced nodes based on mature processes, aiming to reach an "equivalent 1.4nm performance density" by 2031.

The most market-amplified figure here is "1.4nm." But what investors should really pay attention to is the word "equivalent."

Huawei's move essentially finds a system-level optimization solution outside the traditional process path, using more complex and efficient system design to approach performance, without relying solely on the most advanced process nodes. Advanced process technology remains important, but it doesn't mean performance can only be improved through that route.

This approach isn't something Huawei invented out of thin air; the global semiconductor industry has been exploring similar paths in recent years. Increasingly, people realize that performance gains may not only come from the chip itself but also from how chips are assembled, stacked, interconnected, and coordinated.

Many functions that previously had to be integrated into a single large chip can now be separated and assembled more efficiently; signals that used to travel long distances within a system can now be shortened; designing a chip is no longer just an independent engineering task but increasingly resembles designing a complex system, considering devices, packaging, cooling, software, and applications simultaneously.

Huawei's approach is to integrate these previously separate directions into a more explicit expression, providing a complete set of new underlying rules.

The first segments to be re-evaluated may not be the currently popular ones

For investors, rather than obsessing over whether a concept is "new," it’s more important to analyze: if industry logic is truly changing, where will value migrate? That is the starting point for all judgments.

Following the logic of the Tao Law, the companies that tell the most compelling stories about advanced process nodes may not be the first to be revalued. The real early beneficiaries of value re-assessment are likely those that can truly deliver "system efficiency."

The most typical example is advanced packaging.

In the past, many viewed packaging and testing as later-stage processes in the semiconductor supply chain, more akin to manufacturing, with less technical and valuation imagination compared to front-end processes. But if future chip performance increasingly depends on tighter integration between modules, shorter interconnects, and higher density, then packaging is no longer just about encasing chips; it begins to participate in "performance creation" in a certain sense.

This is why companies like Changdian Technology can no longer be simply summarized as "packaging and testing leaders." What truly matters is whether they can evolve from traditional manufacturing roles to platform capabilities in the era of advanced packaging.

Similarly, Tongfu Microelectronics's value isn't just about capacity expansion but whether it can continuously position itself in higher-end packaging demands and collaborate with top-tier clients.

As for Huatian Technology, the market is more concerned whether it can keep pace with a new wave of technological upgrades and transition from traditional packaging/testing to higher value-added capabilities.

If packaging addresses "how to produce more complex chip systems," another segment likely to be re-evaluated is EDA (Electronic Design Automation).

This field has often appeared somewhat "aloof" in the capital markets. Everyone recognizes its importance and its status as a shortcoming, but much of the narrative remains on "domestic substitution." However, if chips increasingly resemble complex systems, the importance of EDA is not just substitution but foundational infrastructure.

Because stacking chips in 3D and doing hybrid bonding require EDA to coordinate and simulate multiple physical interactions—thermal effects, mechanical stress, signal delays, etc. Without advanced EDA, 3D chips simply cannot be "stacked" properly, nor can their performance be guaranteed.

The more complex the system, the less it can be managed by experience and manual effort.

From this perspective, the significance of companies like BGI Jiutian is not just as "domestic EDA leaders," but whether they can become the underlying tool platform for China's era of complex chip design. Golong Electronics, with its expertise in device modeling and simulation, is worth watching because as systems grow more complex, the importance of underlying models increases. Similarly, Guangli Micro, positioned between design and manufacturing, which was once seen as a niche, may gain higher strategic value as system complexity rises.

Profit zones beyond the spotlight

Another trap to avoid is the misconception that emphasizing system collaboration diminishes the importance of front-end equipment, materials, and testing.

In fact, the opposite is true.

Any so-called "equivalent advanced" process ultimately still relies on real manufacturing capabilities. Without equipment, materials, processes, and testing infrastructure, even the most beautiful system design remains on paper. The difference is that in the future, the benefits from these segments may not just come from simple process node upgrades but from upgrades in more complex structures, denser interconnections, and higher verification standards.

Equipment companies like North China Innovation and AMEC will not see their long-term value diminish because of the "τ Law." As China's semiconductor industry advances toward high-end, the foundational equipment capability remains an indispensable base. In fact, as systems become more complex, the requirements for underlying process and equipment capabilities only increase.

The same applies to materials. Companies like Anji Technology may not always be at the conceptual forefront, but they are often among the most stable beneficiaries of industry upgrades. As processes become more complex, demands on polishing, cleaning, interface treatment, and other steps grow higher, with longer customer validation cycles and stronger stickiness. Often, the most sustainable profits come from these high-technical-threshold, hard-to-replace foundational segments.

Another often underestimated area is testing and validation.

As systems grow more complex, simply "making" chips is far from enough for success. Once modules are assembled, performance stability, thermal management, connection reliability, and long-term operation quality all depend on rigorous testing and validation. Companies like Jingce Electronics may not be at the forefront of hype, but as the industry deepens, the importance of these segments will be re-recognized.

Final words

It’s important to note that "new paths" do not necessarily mean "all stocks will rise." The real focus should be on which segments are gaining industry status and which companies have the technical, customer validation, and mass production capabilities.

Based on this logic, the several directions mentioned above are worth long-term tracking:

• Advanced Packaging: Changdian Technology, Tongfu Microelectronics, Huatian Technology;

• EDA: BGI Jiutian, Golong Electronics, Guangli Micro;

• Equipment: North China Innovation, AMEC, Anji Technology;

• Testing & Validation: Jingce Electronics.

They may not always be the fastest risers in every market sentiment cycle, but they are more likely to benefit continuously when industry upgrades truly take hold.

Deeper change may be that the Tao Law signifies a shift in competitive logic. In the post-Moore era, chip competition is moving from single-point process breakthroughs to system-level innovation involving architecture, packaging, design tools, equipment, materials, software, and application scenarios—which fundamentally depends on system organization capabilities.

For many years, China’s semiconductor mainline has been about completing the chain, strengthening the chain, substituting, and catching up. This path is necessary; catching up can narrow gaps but makes it hard to define the rules.

From this perspective, the value of the Tao Law lies more in its indication of a new winning approach—shifting from "how to catch up" to "whether we can change the game."

This is the true significance of "overturning Moore's Law."

For China’s semiconductors, it’s a transition from chasing to defining; for investors, it means valuation anchors should shift from single-point technology to system-level capabilities.