AIDA64 Random Access Time
Version and / or Patch Used: 1.60
Developer Homepage: http://www.aida64.com
Product Homepage: http://www.aida64.com
AIDA64 offers several different benchmarks for testing and optimizing your system, or network. The Random Access test is one of very few (if not the only), that will measure hard drives random access times in hundredths of milliseconds, as oppose to tens of milliseconds.
Drives with only one or two tests displayed in the write test mean that they have failed the test, and their Maximum, and possibly their Average Scores, were very high after the cache fills. This usually only happens with controllers manufactured by JMicron and Toshiba.
As expected, the read latency was very good on the Samsung 840 EVO mSATA 1TB. The TLC flash still manages to produce Marvell level read latency, as shown here with the Plextor PX-M5M.
The write latency is a roller coaster of peaks though, with varying degrees of latency while moving through the cache stages of the flash.
Here we see the actual latency test, measured in 64KB block size. We didn't run today's tests with RAPID enabled, so you can see the drive, and not the software performance. Samsung only mentions one cache stage, Turbo Write, but it appears there are two (if not three) layers before reaching TLC flash on its own.
What saves the 840 EVO in real-world use are flushes of the buffer. In this test, we write to 100% of the user available space, so we're writing nearly 1TB of data, but you won't do that under normal use. Even in extreme circumstances, you may write a 50GB Blu-Ray ISO file; less than half of the first buffer stage on an empty drive.
I've used a 750GB EVO 2.5" in my daily use notebook for the last six months, and have never run into a situation where the drive felt sluggish from write latency, either with, or without RAPID enabled.