Everyone repeats the same SSD rule: never fill your drive all the way up. Manufacturers themselves advise leaving 10-20% free. So what actually happens when you ignore that advice and push a drive to 95% full? One tester decided to find out — on her main Windows drive, no less — and the results were not what she expected.
SSD storage (Credits: How-To Geek)
Instead of jumping straight from a comfortable 64% to full, the test filled a 2TB Lexar NM790 — the system drive, with Windows and everything else on it — in stages, running a game, copying files and checking everyday tasks at each step. Temperatures were monitored with HWiNFO to make sure thermal throttling was not mistaken for a space problem.
SSD at 90% and 95% Full: Benchmark Results Surprise
At first, almost nothing changed. Even at 90% full, the drive was not suddenly struggling, and reaching 95% did not turn it into a glorified hard drive.
CrystalDiskMark results at around 90% capacity were surprisingly strong: sequential reads of 7,429 MB/s and writes of 6,639 MB/s — slightly above the drive’s advertised speeds — with respectable random 4K results of 60.77 MB/s reads and 189.76 MB/s writes. These matched results from when the drive was far emptier.
Why a Fast SSD Benchmark Does Not Tell the Whole Story
Here is the catch the test uncovered: a quick benchmark is a small workload. CrystalDiskMark’s default 1GiB test file is tiny for a 2TB drive, even a nearly full one. The Lexar also uses dynamic SLC caching, which lets it temporarily write data much faster than it otherwise could — but as free space gets scarce, that cache can shrink, and a short benchmark may never push it hard enough to reveal the problem.
That is the real lesson of the experiment. A nearly full SSD can still look fast in benchmarks and everyday use, because reads are barely affected and short writes stay inside the cache. The risk shows up in sustained, heavy writes — the kind of large file transfers where a shrinking cache and a controller with little room for garbage collection and wear-levelling finally make themselves felt.
So should you ignore the 10-20% rule? Not quite. The test suggests a full drive will not instantly ruin your PC — but the headroom exists for a reason, and it is the heavy, sustained workloads, not the benchmark scores, that will eventually expose a drive running on empty.
Also Read: 8 Free PC Upgrades Hiding in Your Junk Drawer, From Spare RAM to a Forgotten USB-C Cable
Source: How-To Geek — read the full story there.
