Installation and Operation:
The Kingston SSNow mS200 was installed in a handful of systems during use, including its intended destination in a notebook computer, but for testing it was used in a desktop computer using a rather cool
4-port mSATA PCI Express controller from SYBA. This device lets me run up to four mSATA SSDs at once, and from experience I know I can max out the performance of any SATA III spec'ed mSATA drive.
The screenshot below was taken from
CrystalDiskInfo to confirm the details on the review sample.
When installed in the notebook computer it was used as the system's boot drive (and its only drive). While benchmarking was not as convenient in this setting, the real world usage was enjoyable and the system performed very well. Coming from a hybrid SSHD, there was an immediate improvement in boot times and application loading times. We weren't even comparing to an old traditional hard drive, but to a 500GB hard drive with an 8GB SSD component built in, which was intended to improve performance.
Testing:
CrystalDiskMark:
Before we get to any of the official head-to-head tests, I decided for a warm up of CrystalDiskMark 3.0.3 x64, run on just the Kingston mSATA drive. The test was actually run twice, to show the impact on compressible data versus incompressible data on a Sandforce based SSD like the SSDNow mS200, which implements a feature called DuraWrite to compress data, reduce the number of writes required, and in the long run - extend the life of the drive (
more info here). The below left screenshot shows the results when CrystalDiskMark is set to "0 Fill", which simulates compressible data being transferred. We see results that aren't quite up to the drive's published specifications, but very fast! In the below right screenshot we see results with incompressible data - which is the default manner for the test to be run. If you were to open up CrystalDiskMark and just click "Run" you would get results like you see on the right, and would mistakenly assume your drive was much slower than it should be.
