It used to be said that technological development followed Moore’s Law — as transistors shrank, costs decreased. But today, this law seems to be “faltering” in reality.
The recently revealed costs of TSMC’s 2nm and 1.4nm wafers have once again shown the industry what it means to “burn money.”
According to reports, TSMC officially began taking orders for its 2nm process on April 1 this year. Leading clients like Apple, MediaTek, and Qualcomm were already lining up to “secure production capacity.” However, the ticket price for this “technology train” is not cheap — each 2nm wafer is quoted at $30,000 (approximately 220,000 RMB).
And that’s just the beginning. The cost of 1.4nm wafers, expected to go into mass production in 2028, is projected to rise another 50%, reaching $45,000 per wafer (around 323,000 RMB).
Despite the staggering prices, chip giants like Apple, NVIDIA, AMD, and Qualcomm remain on TSMC’s production roadmap.
The reason is simple: the performance gap is too significant. Industry data shows that each new generation of process can improve chip speed by over 30%, while also reducing power consumption by about 20%. Take mobile processors for example — compared to 3nm, a 2nm chip can process much faster under the same power conditions. To stay competitive, manufacturers are compelled to chase the latest technology.
A look at price trends over the past decade also offers clues. Back in 2013, during the A7 chip (28nm) era, each wafer cost about $5,000. By the time of the A14 (5nm) in 2020, the price had risen to $16,000. The current A18 Pro (3nm) costs $18,000 — and now 2nm jumps straight to $30,000.
Although chip performance, features, and core counts are rising rapidly, manufacturing costs are growing even faster. The cost per square millimeter has increased from $0.07 to $0.25 — driven by across-the-board upgrades in equipment, materials, and process complexity.
At the same time, transistor density growth is gradually hitting bottlenecks. In the early days, from 28nm to 16nm, density improved rapidly, but entering the 3nm generation — especially during the N3E phase — performance gains per unit area have slowed, while costs continue to rise. SRAM has encountered especially significant physical limitations.
TSMC’s pricing is not arbitrary. Reportedly, building a 2nm fab could cost over $700 million, not to mention the use of gate-all-around (GAA) structures and the upcoming integration of backside power delivery technology. These technological leaps require extremely complex and expensive equipment — such as ASML’s High-NA EUV lithography machines, each priced close to $400 million.
For consumers, these costs will eventually trickle down into end products. Analysts estimate that if the 2026 iPhone uses a 2nm chip, the cost of a single processor could be $50 higher than now. Following Apple’s usual triple-markup pricing model, the overall price of the device could rise by $150 (around 1,000 RMB).
Note: Data in this article is from reports by Industrial & Commercial Times, Economic Daily, and Wccftech.
Disclaimer:
- This channel does not make any representations or warranties regarding the availability, accuracy, timeliness, effectiveness, or completeness of any information posted. It hereby disclaims any liability or consequences arising from the use of the information.
- This channel is non-commercial and non-profit. The re-posted content does not signify endorsement of its views or responsibility for its authenticity. It does not intend to constitute any other guidance. This channel is not liable for any inaccuracies or errors in the re-posted or published information, directly or indirectly.
- Some data, materials, text, images, etc., used in this channel are sourced from the internet, and all reposts are duly credited to their sources. If you discover any work that infringes on your intellectual property rights or personal legal interests, please contact us, and we will promptly modify or remove it.
