The memory crisis has reached a critical inflection point. With DDR5 prices increasing by triple digits year-over-year and major manufacturers like Micron abandoning consumer markets entirely, ASUS is reportedly considering an unprecedented move: manufacturing its own DRAM by the second quarter of 2026.
According to reports from Persian tech outlet Sakhtafzarmag—which has accurately predicted AMD and Intel developments in the past—ASUS is reportedly planning to enter the DRAM manufacturing market by the second quarter of 2026, establishing its own production lines to secure a stable supply for its ROG and TUF product ranges.
If accurate, this represents a fundamental shift in how PC manufacturers approach component supply. But there's substantial skepticism about whether ASUS can realistically execute this plan.
Why ASUS Would Consider This Move
The rationale is straightforward: survival. After the three memory sources shifted their focus to AI data centers, DDR4 and DDR5 RAM prices began skyrocketing, and even though Asus is a major force in the hardware industry, it's not being prioritized for a stretched supply of components.
ASUS, like Dell and Framework, has already announced laptop price increases due to memory shortages. The company faces a choice: either accept ballooning costs and pass them to consumers, or attempt to secure its own supply chain.
As established brands like Micron scale back their consumer-focused offerings, a significant gap has opened in the market that Asus could step in to fill. If ASUS produces its own RAM, it could ensure compatibility with its motherboards, optimize performance for its systems, and price its products more competitively—at least in theory.
For ASUS, it is more about survival in the current crisis scenario rather than a profit-seeking expansion into semiconductor manufacturing.
The Enormous Challenges
Building DRAM production capability from scratch is not remotely comparable to establishing a motherboard or laptop assembly line. Building a semiconductor fabrication plant is an immense undertaking. It requires billions in investment and years of technical calibration.
Modern DRAM fabs cost $10-20 billion and take 3-5 years to build from groundbreaking to production. Even if ASUS has the capital—which is questionable—the timeline doesn't align with a Q2 2026 target unless they're pursuing a very different strategy.
The first thing that comes to mind is that Asus doesn't have the means to manufacture its own DRAM chips in just a few months. This raises the question: what is ASUS actually planning?
The CXMT Partnership Angle
The most plausible explanation is that ASUS wouldn't build fabs from scratch, but instead partner with an existing manufacturer that has excess capacity and needs customers.
Enter CXMT (ChangXin Memory Technologies), a Chinese DRAM manufacturer that recently demonstrated surprising capabilities. CXMT's DDR5 chips with an 8,000 MT/s data transfer rate come in 16 Gb (2 GB) and 24 Gb (3 GB) capacities, thus enabling module makers to build both binary and non-binary DIMMs.
In November 2025, CXMT unveiled a full lineup at IC China 2025, including:
- DDR5 memory up to 8000 MT/s
- LPDDR5X mobile memory up to 10667 MT/s
- Multiple form factors: UDIMM, SODIMM, RDIMM, MRDIMM, CUDIMM, CSODIMM
The company is reported to have increased production to 6% of the world DRAM output as of Q1 2025 with aim to increase to 10% by Q4 2025.
The alternative would be for Asus to source its supplies from a challenger looking for outlets… And in this case, Chinese production is the place to be.
This makes far more sense than ASUS building its own fabs. ASUS could contract with CXMT for dedicated production capacity, potentially with ASUS branding and optimization for ASUS products. This is vastly more achievable than constructing semiconductor fabrication facilities.
CXMT's Capabilities and Limitations
CXMT has made impressive progress, but faces significant constraints. CXMT didn't disclose which fabrication process it used to produce its latest DDR5 and LPDDR5X memory devices, though it is unlikely to be a breakthrough, as the company doesn't have access to the leading-edge fab tools required to build DRAMs using sub-18nm manufacturing technologies.
Among other obstacles, it contends with strict U.S. regulations on equipment imports. As a result, it has struggled to efficiently manufacture large quantities of chips.
The U.S. has placed CXMT on various restriction lists, limiting its access to advanced lithography equipment and other critical tools. In 2022, the James M. Inhofe National Defense Authorization Act for Fiscal Year 2023 banned the U.S. federal government from buying or using chips from CXMT.
Production capacity remains a question mark. In the DRAM field, CMXT plans to increase its DDR5/LPDDR5 production capacity to 110,000 wpm by the end of 2025, accounting for 6% of the global DRAM production capacity. While 6% sounds modest, if CXMT dedicates a portion of that capacity to ASUS, it could meaningfully impact ASUS's supply chain.
In 2024, CXMT generated $2.8–$4.2 billion revenue but held only 4.9% of global DRAM market vs. 93.4% by Samsung, SK Hynix, and Micron. The company is small but growing rapidly.
Market Context: The Memory Crisis Is Getting Worse
The backdrop makes ASUS's reported plans more understandable. Chen expects availability of DRAM and NAND to worsen in the first and second quarters of 2026 once distribution stockpiles are exhausted, and he projects the current shortages to extend into late 2027 and potentially beyond.
RAM prices have reached absurd levels. The 32GB DDR5 6000 CL30 RAM I bought in late March for $99 is now going from the same vendor for $297 or exactly triple what I paid for it just 8 months ago, according to one Tom's Hardware commenter.
Memory prices are likely to rise 30% in Q4 2025 and possibly 20% more early next year, on top of 50% price increases experienced year to date, according to Counterpoint Research.
The root cause is well-established: DRAM makers reallocate their production capacities to HBM (which uses larger DRAM dies than commodity types of memory) that is consumed by AI accelerators, like Nvidia's B300 or custom accelerators by large cloud service providers, such as AWS, Google, and Microsoft.
These hyperscalers book supply years in advance, and memory manufacturers prioritize them because profit margins are substantially higher than consumer DRAM.
What This Could Mean for Consumers
If ASUS successfully establishes a relationship with CXMT or another supplier, several outcomes are possible:
Best Case Scenario:
- ASUS secures dedicated memory supply at more stable pricing
- ASUS-branded memory modules appear at competitive prices
- Other PC manufacturers follow suit, reducing dependence on the Samsung/SK Hynix/Micron oligopoly
- Increased competition moderates price increases
Realistic Scenario:
- ASUS secures enough supply for internal use (laptops, pre-built desktops)
- Limited retail availability of ASUS-branded modules
- Prices remain high but stabilize relative to continued increases elsewhere
- ASUS products become relatively better value compared to competitors still dependent on expensive spot market pricing
Worst Case Scenario:
- Partnership negotiations fail or CXMT cannot scale production adequately
- ASUS still faces severe allocation constraints
- The company raises prices anyway and the DRAM manufacturing plan fizzles
- Or ASUS successfully sources memory but charges premium prices anyway, pocketing the cost advantage rather than passing savings to consumers
Of course, Asus could also choose to sell its DDR5 RAM at high markups. That would not be the outcome consumers are hoping for.
The Geopolitical Complication
Partnering with CXMT introduces geopolitical complexity. U.S.-China tensions over semiconductor technology mean that ASUS would need to navigate export controls, potential sanctions, and political sensitivities.
Products using CXMT memory might face restrictions in certain markets or customer segments. Government and enterprise customers may refuse systems with Chinese-sourced memory due to security concerns or procurement regulations.
This could fragment ASUS's product line, with some SKUs using CXMT memory for cost-sensitive consumer markets and others using Samsung/Micron/SK Hynix for enterprise and regulated markets.
Is This Real or Negotiating Leverage?
One possibility that can't be dismissed: Is this a bluff attempt to put pressure on those who have been supplying it with RAM and the like for years for its notebooks, servers, complete PCs and so on?
If ASUS publicly threatens to establish alternative supply sources, Samsung, SK Hynix, and Micron might prioritize ASUS's allocation to avoid losing a major customer. The mere rumor of ASUS partnering with CXMT could improve ASUS's negotiating position without the company needing to actually execute the plan.
However, given the severity of the memory crisis and the track record of Sakhtafzarmag's predictions, it seems more likely that ASUS is genuinely exploring this option as a contingency.
The Broader Industry Impact
If ASUS moves forward, other PC manufacturers will face pressure to secure alternative supply chains. If ASUS does end up opening up a DRAM front, then we can see other PC businesses also benefiting from them if they can meet their own demand and are left with additional capacity.
Dell, Lenovo, HP, and others could follow with similar strategies, either partnering with CXMT or seeking arrangements with other second-tier memory manufacturers.
This fragmentation could actually benefit consumers by reducing market concentration. Currently, three companies control 93.4% of global DRAM production. Any increase in viable alternatives reduces oligopoly power and should theoretically moderate pricing.
If China meets more of its own demand, global supply could finally ease. The well known trio of SK Hynix Samsung and Micron would then be able to serve the rest of the world more freely.
Timeline Reality Check
The Q2 2026 timeline appears extremely ambitious if ASUS plans to physically manufacture memory. However, if the plan involves contracting with CXMT for dedicated production runs using existing capacity, Q2 2026 becomes plausible for initial product launches.
Current reports expect memory shortages to last till the end of 2027 and even into 2028. This multi-year crisis provides ASUS sufficient runway to establish alternative supply chains even if execution takes longer than initially hoped.
What PC Builders Should Do
For consumers watching this situation:
Don't wait for ASUS memory to arrive before upgrading. If you need RAM now and find reasonable pricing, purchase it. The memory situation is more likely to worsen than improve in early 2026.
Watch for ASUS product pricing. If ASUS secures alternative memory supply, its laptops and pre-builts may offer better value relative to competitors still paying inflated spot market prices.
Consider CXMT-based products if they appear. While CXMT memory won't match the absolute cutting-edge from Samsung or SK Hynix, DDR5-8000 specifications are competitive for most consumer applications. The primary concerns are reliability/RMA rates and long-term availability.
Monitor geopolitical developments. U.S.-China semiconductor policy could dramatically impact the viability of CXMT partnerships.
The Bottom Line
ASUS producing its own DRAM by Q2 2026 seems implausible if interpreted as building semiconductor fabs from scratch. However, ASUS partnering with CXMT or another manufacturer for dedicated production capacity is both feasible and strategic given the current market conditions.
Whether this materializes, and whether it benefits consumers or simply stabilizes ASUS's internal supply chain at elevated prices, remains to be seen. The memory crisis is real, worsening, and projected to continue through at least 2027-2028. Companies dependent on memory supply are exploring every option to avoid being entirely at the mercy of a three-company oligopoly prioritizing AI customers.
If nothing else, the rumor highlights how desperate the situation has become for PC manufacturers. When building your own semiconductor supply chain starts looking like a reasonable business strategy, you know the market has fundamentally broken.
We'll be watching to see if ASUS actually follows through—and what other PC manufacturers do in response.