PCIe trouble with 4TB Crucial T500 NVMe SSD for >1 power cycle on MSI PRO X670-P

Re:PCIe trouble with 4TB Crucial T500 NVMe SSD for >1 power ... by userbinator on Saturday December 28, 2024 @08:07PM (#42536451)

That sounds like NAND degradation (retention failures) which can only be partially worked around in firmware (and causing more write cycles on already-marginal QLC). Unfortunately the real solution is "use better NAND", which is unlikely to happen unless enough people demand it.

[ Parent | on HN ]

Re:PCIe trouble with 4TB Crucial T500 NVMe SSD for >1 power ... by ciupicri on Saturday December 28, 2024 @08:32PM (#42536595)

Kingston KC3000 supposedly uses Micron 176L TLC memory [1].

The Seagate Firecuda 530 datasheet clearly says "Built with a Seagate-validated E18 controller and the latest 3D TLC NAND". A review is more precise: "Phison PS5018-E18" & "Micron B47R 176-layer 3D TLC NAND" [2].

[1]: https://www.tomshardware.com/reviews/kingston-kc3000-m2-ssd-...

[2]: https://www.kitguru.net/components/ssd-drives/simon-crisp/se...

[ Parent | on HN ]

Re:PCIe trouble with 4TB Crucial T500 NVMe SSD for >1 power ... by userbinator on Saturday December 28, 2024 @10:12PM (#42537139)

B47R is indeed TLC, rated for only 1000 cycles (and 35k in SLC mode, at 1/3 the capacity.) There's also the question of whether this is "true" Micron NAND, or SpecTek which is basically Micron's rejects (and rated for even fewer cycles; only 300 in the case of their B16A.)

[ Parent | on HN ]