Samsung has launched a new entry-level M.2 NVMe SSD, but its market entry defies traditional pricing logic. Priced at approximately €263 per terabyte, this new QLC (Quad-Level Cell) drive without onboard DRAM costs significantly more than the flagship 990 PRO—a high-performance TLC (Triple-Level Cell) drive with DRAM—did at its 2022 debut of €180.
The Erosion of Value in the QLC Transition
The core of this issue lies in the architectural shift from TLC to QLC NAND flash. In the hierarchy of storage, TLC has long been the gold standard for balancing endurance, speed, and cost. By moving to QLC, Samsung is packing four bits per cell, which inherently increases storage density but compromises write speeds and overall endurance—the TBW (Total Bytes Written) rating.
Normally, a shift toward QLC is the industry’s signal for a price floor reduction. Here, the inverse is occurring. Samsung is shipping a drive that is technically inferior in terms of raw IOPS (Input/Output Operations Per Second) and sustained write performance compared to its 2022 flagship, yet it carries a premium price tag. This isn’t just inflation; it’s a fundamental recalibration of what the market will tolerate.
Hardware Realities and The DRAM-less Tax
Beyond the NAND type, the lack of onboard DRAM is the most significant technical hurdle for this new drive. High-performance SSDs typically utilize a dedicated DRAM cache to store the LBA (Logical Block Addressing) map. Without it, the drive must rely on HMB (Host Memory Buffer) technology, which offloads mapping data to the system’s RAM.
While HMB is efficient for casual workloads, it introduces latency overhead in high-intensity environments. When you combine the inherently slower write cycles of QLC with the absence of a dedicated cache, the performance profile shifts from “workstation-grade” to “general consumer.” Yet, the pricing suggests a premium tier.
As noted by storage researcher and industry analyst Jim Handy of Objective Analysis, the shift in NAND pricing is often disconnected from consumer-facing MSRPs: “NAND manufacturers are under immense pressure to recoup R&D costs for high-layer count 3D NAND, often leading to pricing structures that prioritize margin over the traditional performance-per-dollar metrics users expect.”
The Broader Silicon Valley Chip War Context
This pricing anomaly is not happening in a vacuum. It is a symptom of the broader “Chip Wars,” where the supply chain for NAND flash is becoming increasingly consolidated. As firms like Samsung, SK Hynix, and Micron pivot their fabrication capacity toward high-bandwidth memory (HBM) for AI training clusters, the supply of standard consumer NAND is being squeezed.
When factory capacity is diverted to satisfy the insatiable demand for AI hardware—specifically the HBM3e modules required for NVIDIA’s H200 and Blackwell architectures—the cost of producing standard consumer storage rises. This creates an ecosystem where the average user ends up subsidizing the infrastructure required for the AI boom.
Comparative Storage Metrics
- 990 PRO (2022): TLC NAND, Dedicated DRAM, High Endurance, €180/TB launch.
- New Entry-Level (2026): QLC NAND, DRAM-less (HMB), Lower Endurance, €263/TB launch.
What This Means for Enterprise and Enthusiast IT
For the average system builder or enterprise procurement officer, this shift is a warning. The era of “cheaper by default” for entry-level hardware is effectively over. We are moving toward a market where performance tiers are no longer linearly linked to price. If you are building a machine for heavy I/O tasks, such as video rendering or localized LLM (Large Language Model) inference, the “entry-level” label on a box can no longer be trusted as an indicator of value.
The move toward QLC in entry-level drives also has implications for data integrity. As highlighted in recent technical documentation on SNIA’s Storage Networking Primer, QLC’s lower cell endurance requires sophisticated wear-leveling algorithms that add overhead, further complicating the performance profile. Developers should be wary: if your software relies on high-frequency write operations, a DRAM-less QLC drive will likely become a bottleneck far faster than a legacy TLC drive would have.
The 30-Second Verdict
Samsung is attempting to normalize higher price points for lower-tier hardware. By stripping away DRAM and opting for denser, slower QLC flash, they have created a product that is objectively less capable than their 2022 flagship, yet they have priced it as a premium offering. Unless you are constrained by specific form-factor requirements, the secondary market or remaining stock of older, superior TLC drives remains the better value proposition.
This is a market signal, not a product evolution. As production shifts to favor AI-focused silicon, expect the “price-per-gigabyte” metric to remain volatile and disconnected from performance parity for the foreseeable future. For those tracking the industry, the transition is documented in detail through the OpenZFS documentation regarding how flash storage interacts with modern file systems, where the reliability of the underlying NAND becomes a critical factor in data stability.
Ultimately, the numbers speak for themselves. When an entry-level part costs 46% more than a flagship part did four years ago, the value proposition has not just shifted—it has been fundamentally inverted.
