The global memory chip market is experiencing a fresh round of valuation volatility and debate over technology roadmaps. In capital markets, the Philadelphia Semiconductor Index and memory sector have seen notable pullbacks, while on the industry side, ChangXin Memory Technologies' (CXMT) breakthrough in 4F² DRAM architecture has sparked widespread discussion.
On October 21, Micron Technology (NASDAQ: MU) President and Chief Technology and Product Officer Scott DeBoer hosted a technology and product webinar, with one of the core topics being a response to investor concerns raised after CXMT disclosed its 4F² architecture plans.
Scott DeBoer stated plainly that although the company had conducted long-term internal evaluations and prototype validation of 4F², Micron Technology (NASDAQ: MU) will stick with its existing 6F² approach and ultimately move directly to 3D DRAM. Micron believes that from a long-term evolution standpoint, 3D DRAM is the more certain choice in terms of "scalability and cost-effectiveness."
He further explained that the key factor determining DRAM bit density and cost scaling is capacitor pitch. The vertical channel transistor (VCAT) used in the 4F² architecture has no viable scaling path beyond pitch reduction.
According to a subsequent research report from UBS analyst Timothy Arcuri's team, if Micron's judgment proves correct and the industry ultimately pivots to 3D DRAM, deposition and etch equipment suppliers stand to benefit.
As previously noted, CXMT President Cao Kanyu disclosed on October 10 that the company plans to launch a new-generation DDR5 RDIMM based on the 4F² architecture by year-end, which can reduce chip area by approximately 30% compared to the mainstream 6F² architecture, allowing more chips to be cut from a single wafer.
Micron's Core Argument: 4F² Is Not the Path to the Future
The reason CXMT's 4F² technology has drawn intense industry attention is that it directly addresses the most fundamental commercial pain point of DRAM chips—bit density per wafer and cost.
In traditional DRAM designs, the industry predominantly uses the 6F² architecture. If a successful transition to 4F² can be achieved, cell area could theoretically be reduced by more than 30% at the same process node, significantly increasing the number of effective dies yielded per wafer.
Following Micron's technology briefing, UBS noted that Micron conveyed to the market its assessment of the bottlenecks facing 4F² mass production:
First, the bottleneck lies in the capacitor, not the transistor. Micron believes the ultimate bottleneck in DRAM scaling is capacitor pitch, not merely the transistor cell structure. If capacitor pitch cannot be compressed in tandem, simply changing to a 4F² structure cannot linearly translate into sustained cost advantages.
Second, the scaling ceiling of VCAT. 4F² typically requires the introduction of vertical channel transistors (VCAT). Micron pointed out that VCAT lacks a sustainable evolutionary path beyond continued pitch compression, and imposes extremely high manufacturing complexity on processes such as ALD (atomic layer deposition) and selective epitaxy, with mass production costs potentially eroding the benefits of area reduction.
According to Deutsche Bank analyst Melissa Weathers' post-meeting report, Micron Technology (NASDAQ: MU) not only explicitly stated it does not plan to transition to the 4F² architecture, but also emphasized that the company has conducted decades of internal evaluation of 4F² solutions and developed prototypes, with the final conclusion being that 6F² remains the more attractive path before the industry moves to 3D DRAM.
Major Players Show Starkly Different Technology Considerations
UBS analysis noted that Micron's stance effectively represents the parallel pursuit of two technology paths, similar to the company's dual preparation regarding EUV adoption at the 1-beta node. The company believes 3D DRAM will eventually converge with its roadmap and views it as a "more scalable and cost-effective" long-term solution.
But Micron's logic does not represent the consensus of the entire industry.
UBS research from the same period shows that in addition to CXMT's all-out push in this direction, industry leader Samsung is also reported to be considering introducing the 4F² architecture at future 0a nm nodes to alleviate process pressure from high-aspect-ratio capacitors.
Samsung plans to complete 6F² mass production at the 1d nm node in 2027, then advance 4F² introduction at the 0a nm node by the end of 2028.
Micron's Technology Narrative
Beyond directly addressing 4F² skepticism, Micron's webinar also sought to emphasize its comprehensive technology portfolio.
AI Memory Map Expansion: Micron Technology (NASDAQ: MU) emphasized that AI's demand for memory is not limited to a single layer of training HBM, but also includes LPDDR6, DDR6, SOCAMM, and 1TB high-capacity RDIMM for inference, edge AI, and autonomous driving.
Advanced Packaging and Logic Integration: Micron Technology (NASDAQ: MU) views hybrid bonding as a core moat for future 16-High and higher-spec HBM, and is actively pursuing collaboration with foundries to develop vHBM (vertically stacked memory on compute cores). Micron aims to demonstrate that memory competition in the AI era has shifted from "per-bit cost control" to "system-level coordination of memory, packaging, and logic architecture."
NAND Value Reimagining: Micron Technology (NASDAQ: MU) also showcased its Gen11 NAND roadmap exceeding 400 layers, emphasizing high-performance storage demand for high IOPS and high bandwidth in the AI era, striving to build a comprehensive barrier combining DRAM and NAND.
The Real Impact of CXMT's 4F²: Market Reaction May Be Overinterpreted
Regarding the market concerns triggered by CXMT's 4F² announcement, several institutions had previously offered relatively cautious assessments.
Nomura Securities believes that neither 4F² nor 3D DRAM will have a "material impact" within the outlook period before 2028, and CXMT's 4F² mass production is not expected to be realized until 2027 to 2028. CITIC Securities also noted that 4F² is currently in an early validation stage, with yield, output, and customer certification all yet to be observed, and its impact on the medium-term memory industry landscape is limited.
As for CXMT's own industry position, its HBM business is still in its early stages and is unlikely to share in the high-premium dividends of this segment driven by AI computing power in the short term. Against the backdrop of a notably narrowing upward slope in general-purpose DRAM prices, this structural characteristic leaves it relatively lacking in hedging space.
TrendForce tracking data shows that the quarter-over-quarter increase in general-purpose DRAM contract prices has narrowed from 93% to 98% in the first quarter of 2026 to 13% to 18% in the third quarter, and is expected to further decline to 10% to 15% in the fourth quarter.
Comments