Why you can trust Tom's Hardware Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.
Four-Corner Performance Testing
Sequential Read
To read about our storage tests in-depth, please check out How We Test HDDs And SSDs. Four-corner testing is covered on page six.
The Plextor M6V lands at the top of our sequential read chart. Though, when it comes to reading data back from flash, even the low-cost TLC-based SSDs perform really well. Writing data for more than a few seconds is what creates problems for triple-level-cell NAND. This is countered with an emulated SLC buffer that quickly fills, leaving you with dismal native TLC speeds that often trail mechanical disks.
Sequential Write
With most modern SATA-attached SSDs, we can usually link the price directly to write performance. This applies to entry-level drives armed with TLC flash. Samsung's 850 EVO and SanDisk's Ultra II are both TLC-based, though you'd have a hard time guessing it since their emulated SLC caching algorithms are so good. It doesn't surprise us to see both competitors writing sequential data faster than the MLC-based M6V.
Random Read
The M6V comes close to the 10,000 random read IOPS mark at a queue depth of one. That's the magic number distinguishing good SSDs from great ones. Very few client drives make it over the hump. In fact, it's rare that you'd actually need 10,000 random IOPS. But when a drive is capable of processing that many operations per second, you can expect low latency, which in turns makes your PC feel more responsive.
Random Write
Plextor's M6V lands in the top group for random writes as well. The MX200, the only other MLC-equipped model in our group, and Samsung's 850 EVO round out the top tier at all queue depths. SanDisk's Ultra II starts out strong but quickly levels out with a hard ceiling just over 45,000 IOPS.
We didn't expect the M6V 256GB to perform this well, but it's proving itself to be a solid low-cost solution. Onward with the benchmarking...
