Samsung’s P9 USB4 portable SSD arrived with the kind of sequential read and write figures that get posted in large type on product pages. The device targets the professional portable-storage market, where the theoretical ceiling of USB4 (40 Gbps on compliant hosts) finally makes an external drive feel less like a compromise. Whether that throughput actually reshapes a real batch photo workflow is a more careful question than the spec sheet answers.
What USB4 Changes — and What It Doesn’t
USB4 at 40 Gbps roughly doubles the practical headroom available compared to USB 3.2 Gen 2×2 (20 Gbps), and it’s substantially faster than the older USB 3.2 Gen 2 single-lane standard that most previous portable SSDs max out at. The P9 is designed to approach that ceiling on a host that properly supports USB4’s full bandwidth — something that requires checking your laptop or desktop’s actual controller, not just the presence of a USB-C port.
That caveat matters more than it sounds. USB4 is a specification that nests several capability tiers under one connector and one branding umbrella. A host machine can have a USB-C port that negotiates USB4 but only at 20 Gbps, or a Thunderbolt 3 port that provides 40 Gbps with different protocol overhead. On a USB4 40 Gbps host, you get the full benefit. On anything less, the drive scales back — silently, without a warning dialog. Checking the host controller’s specification before building a workflow around this drive is not optional; it’s the first step.
Assuming a genuinely capable host, what does the throughput gain buy in practice? The honest answer is: it depends almost entirely on which part of the workflow is the actual constraint.
The Bottlenecks in a Real Batch Workflow
A typical batch scenario — ingesting RAW files from card, running conversions or exports, archiving the output — passes through several choke points that aren’t all storage I/O:
Card reader speed. CFexpress Type B cards can sustain writes in excess of 1,500 MB/s on the camera side, but a USB 3.2 Gen 2 card reader tops out around 1,000 MB/s theoretical and often less in sustained transfer. If the card reader connects via USB 3.2 Gen 2 and the destination drive connects via USB4, the card reader is the binding constraint for ingest. The faster drive sits waiting.
CPU and GPU decode time. Processing RAW files — particularly the demosaicing, denoising, and color profile application stages — is handled by the processor and, increasingly, by GPU acceleration. A batch export from Lightroom or Capture One spends far more wall-clock time in decode and render than in file I/O for most file sizes. Storage speed stops mattering once the CPU is the limiter.
File size and count. Large individual files (high-megapixel bodies shooting multi-hundred-megabyte RAWs) are where sequential read performance of a fast external SSD becomes genuinely valuable — loading those files into an editing application, or writing export batches of tiff stacks or layered PSDs. Smaller JPEG exports or web-sized outputs are often too short to saturate any modern interface.
Software I/O pattern. Sequential read/write is where USB4 and the P9 shine. Lots of small random reads — the access pattern that comes from navigating a catalog, loading previews, or reading sidecar files — are dominated by latency more than raw throughput, and a drive’s random IOPS matter more there than sequential bandwidth.
This isn’t a reason to dismiss the P9’s performance. It’s a reason to audit your workflow before concluding that storage speed is the constraint you need to address.
Where the Speed Genuinely Applies
For workflows that are actually I/O-bound, the difference between a USB 3.2 Gen 2 drive and a USB4 drive running at full bandwidth is not subtle. Sustained sequential throughput in the range the P9 targets means:
- Moving several hundred gigabytes of overnight tethered-capture files from one external drive to another — a pure sequential write-to-read scenario — completes in a meaningfully shorter window.
- Writing large export batches of uncompressed TIFFs or layered PSD files from an internal NVMe scratch location to a portable archive drive is faster, because the destination no longer outpaces what it can absorb.
- Scrubbing through proxies or cache files on the external drive, when the editing application keeps those sequentially arranged, benefits from higher sustained reads.
Workflows built around conversion-heavy pipelines — batch RAW-to-DNG conversions at volume, or any scenario where the software is feeding the encoder faster than it can drain through a slower interface — are the clearest beneficiaries. That’s the scenario where USB4’s headroom closes a real gap rather than solving a bottleneck that doesn’t exist. Our Conversion Workflows coverage explores a range of those pipeline scenarios in more depth.
Practical Setup Considerations
A few concrete details that affect whether you capture the rated performance:
The cable is not incidental. USB4 40 Gbps requires a cable certified for that bandwidth, and the specification allows passive cables at shorter lengths but mandates active cables for anything beyond roughly 0.8 meters. Using a generic USB-C cable — even a high-quality one that handles charging at full wattage — may silently negotiate a lower speed tier.
Thermal behavior matters on sustained transfers. Portable SSDs in compact enclosures throttle when the NAND reaches temperature limits. A sustained multi-gigabyte transfer may start at rated speed and step down partway through. How the P9 behaves thermally in extended batch sessions depends on ambient conditions and airflow; a chassis sitting flat on a desk in a warm room will throttle sooner than one elevated with airflow underneath.
File system overhead is real but small. APFS on macOS handles large sequential I/O without significant penalty; NTFS on Windows is comparable for sequential operations. ExFAT, common on freshly formatted portable drives, has higher overhead on small-file random access — another reason to match the format to the access pattern.
Before You Reorganize Your Workflow Around This Drive
The P9 represents a genuine step forward in portable SSD throughput, and USB4 40 Gbps is a meaningful ceiling to be pushing against rather than a theoretical limit that never appears in practice. But the right way to use that speed is to first identify where your workflow’s actual wall is.
If you’re exporting 45-megapixel RAW files and Lightroom’s export progress bar stalls in the processing phase rather than the writing phase, a faster drive won’t change that. If it stalls at the write bar — or if large archive transfers between drives are where time goes — the P9 is directly addressing the constraint.
Run a simple transfer of a large, representative batch between your current drive and your working location, time it, then check where CPU usage lands during the same operation. If CPU pegs at 100% while the drive idles, the bottleneck is the processor. If CPU has headroom and the drive is sustaining near its rated speed, you’re I/O-bound and a USB4 drive earns its place in the setup. That test takes minutes and tells you more than any spec comparison does.