BENCHMARK PROTOCOL AND THE TEST BENCH
In testing, our main objective is to obtain results as pure and as accurate as possible and we want to ensure that no anomalies slip through. Simply put, we want to provide you with the absolute best results the tested hardware can provide. Repetition in testing is standard and, if necessary, we may conduct specific tests in Windows 7 ‘safe mode’ to ensure the OS has little to no influence on the end result.
In order to validate and confirm our findings, testing is supported by industry accepted benchmark programs. All results are displayed through capture of the actual benchmark for better understanding of the testing process by the reader.
SSD COMPRESSION AND TESTING FLUCTUATIONS
All SSDs are not created equal and many new SSD enthusiasts realize that when they test their new drive to confirm specifications and ensure all is in order. LSI SandForce controlled SSDs, as in the OWC Mercury Accelsior 480GB PCIe we are testing today, use compression techniques in storage whereas many others do not. This creates a bit of confusion when enthusiasts test the drive with random data through benchmarking programs such as AS SSD and Crystal Diskmark. The results seem to be lower than the listed specifications.
Under normal circumstances one might observe a difference between the Accelsior and another ‘non-SandForce’ controlled PCIe card, however, the only other on the market is also ‘SandForce Driven. This then brings us to the Marvell 88SE9230 6Gbps controller which might also have an impact on testing and is one that we are not familiar with which makes things all that much more interesting.
The software we will be using for today’s analysis is typical of many of our reviews and consist of ATTO Disk Benchmark, Crystal DiskMark, AS SSD, Anvil Storage Utilities, HDTune Pro and PCMark Vantage. We rely on these as they each have a way of supporting one another yet, at the same time, adding a new performance benchmark to the total picture. Much of the software is free and can be downloaded simply by clicking on the linked title.
ATTO Disk Benchmark is perhaps one of the oldest benchmarks going and is definitely the main staple for manufacturer performance specifications. ATTO uses RAW or compressible data and, for our benchmarks, we use a set length of 256mb and test both the read and write performance of various transfer sizes ranging from 0.5 to 8192kb. Manufacturers prefer this method of testing as it deals with raw (compressible) data rather than random (includes incompressible data) which, although more realistic, results in lower performance results.
Initial specifications listed 780MB/s read and 648MB/s write whereas our ATTO result shows highs of 765MB/s read and 740MB/s write, write performance being just short of 100MB/s higher than listed specifications.