PCIe 5.0 Storage Speed Depends on Which CPU Is in the Socket
PCIe 5.0 SSDs have been on shelves long enough that their headline transfer speeds – often north of 12 GB/s sequential read – no longer feel like science fiction. What still surprises builders, though, is how much the host CPU influences whether those speeds actually show up in practice. AMD’s Ryzen 7 9800X3D and Intel’s Core Ultra 7 265K both support PCIe 5.0 x4 for M.2 storage, but their platform architectures handle that bandwidth differently, and the gap between them becomes visible the moment you start pushing a Gen5 drive under load.
The comparison matters most for PC gaming builders who are choosing a platform in 2024 and 2025, not just buying a CPU. Pairing a flagship PCIe 5.0 SSD with either processor costs real money, and the assumption that both platforms deliver identical storage performance is worth questioning before the credit card comes out.
Platform Architecture: Where the Differences Start
The Ryzen 7 9800X3D sits on AMD’s AM5 platform using the X670E or B650E chipset. Its PCIe 5.0 lanes for M.2 storage route directly through the CPU die, which keeps latency tight and avoids chipset bottlenecks. AMD’s Zen 5 architecture, paired with that 96MB 3D V-Cache, was optimized for gaming workloads first, but the memory subsystem also feeds storage I/O efficiently. The result is a platform where a Gen5 SSD like the Samsung 9100 Pro or Crucial T705 operates close to its rated envelope without the CPU becoming a limiting factor in queue-depth reads.
Intel’s Core Ultra 265K runs on the LGA1851 socket with Z890 motherboards. Intel also routes its primary M.2 PCIe 5.0 slot directly through the CPU, so on paper the setup is comparable. Where Intel’s platform introduces complexity is in how the memory controller and cache hierarchy interact with high-throughput storage I/O. The Core Ultra 200S series shifted to a disaggregated tile design, and while that architecture pays dividends in multi-threaded workloads, it can introduce slightly higher latency at low queue depths – the kind of reads and writes that game asset streaming actually produces.
Neither platform throttles PCIe 5.0 storage in any catastrophic way. Both can sustain the full rated sequential throughput of a top-tier Gen5 drive. The differences appear in the details: random 4K IOPS, access latency under mixed read/write loads, and CPU overhead during sustained sequential transfers.
Sequential Throughput: The Easy Part
In sequential benchmarks run through CrystalDiskMark and AS SSD, both CPUs allow Gen5 SSDs to hit their rated read peaks. A Crucial T705 2TB, for example, reaches around 14.9 GB/s sequential read on both platforms without meaningful variance between them. This is where most marketing material stops, and understandably so – the numbers look identical and the story feels complete.
Sequential write throughput follows the same pattern. Neither the 9800X3D nor the 265K restricts write bandwidth in ways that show up in standard synthetic workloads. If sequential speed is the only metric on your checklist, both CPUs get full marks, and the SSD choice matters more than the CPU choice at this tier.
Where the 9800X3D Pulls Ahead in Real-World Storage Load
Random 4K performance at low queue depths is where platform differences start separating the two processors. Game engines, particularly those running open-world titles on DirectStorage, generate thousands of small, randomized read requests rather than clean sequential streams. At QD1 and QD4 – the access patterns that represent actual game load scenarios – the 9800X3D platform has shown lower average latency than the 265K in multiple third-party evaluations. The margin is not dramatic. We are talking about single-digit microsecond differences in average read latency, but those differences compound when a game is pulling hundreds of assets simultaneously during a scene transition or fast-travel sequence.
The reason traces back to the 9800X3D’s cache architecture. The 3D V-Cache is purpose-built for gaming workloads, and AMD’s fabric is tuned to minimize round-trip latency between the CPU and connected peripherals. Storage controllers communicate with the processor through that same fabric, so the cache efficiency that benefits game rendering also benefits storage I/O indirectly. The 265K’s tile-based design prioritizes throughput scaling for workstation and productivity tasks, which is a legitimate design choice – just not one that optimizes for the low-latency random access patterns that matter most in gaming.
CPU overhead during sustained storage transfers is another measurable difference. When a PCIe 5.0 SSD is saturating its interface with a large file copy or a game installation in the background, the 265K can see CPU utilization tick up more noticeably than the 9800X3D under the same conditions. This matters in a gaming context because background storage activity – downloads, shader compilation, auto-saves – does not stop while a game is running. Higher CPU overhead from storage management competes with frame-generation threads and physics workloads for CPU time, and the 9800X3D handles that competition more gracefully because of how Zen 5’s scheduler isolates workloads across cores.
The 265K is not a poor platform for PCIe 5.0 storage. Its sequential performance is indistinguishable from the 9800X3D, and in workstation scenarios – video editing timelines, large dataset transfers, 3D asset rendering – it performs well and sometimes edges ahead in throughput-heavy tasks thanks to its higher sustained all-core frequency. But for a gaming PC where the SSD’s job is to feed the GPU as fast as possible with the lowest possible CPU involvement, the 9800X3D’s platform architecture suits that goal more directly. A builder spending money on a Gen5 drive expecting gaming benefits should weigh that platform advantage alongside the CPU price and motherboard costs.
One practical consideration the benchmarks do not capture: PCIe 5.0 SSDs run hot. Both platforms require the drive to be thermally managed, and neither CPU changes that equation. A Gen5 drive without an adequate heatsink will throttle regardless of whether a 9800X3D or 265K is in the socket – and once throttling kicks in, the platform latency advantages the 9800X3D offers disappear entirely behind thermal limits the SSD imposes on itself.
