Four-Way SSD Round-Up, OCZ, Super Talent, Mtron
Our Test System And SANDRA Testing
Our Test Methodologies: Under each test condition, Solid State Drives were installed as secondary volumes in our testbeds,with a standard spinning hard disk for the OS and benchmark installations. SSDs were left blank without partitions wherever possible, unless a test required them to be partitioned and formatted. as was the case with our ATTO benchmark tests. Windows firewall, automatic updates and screen savers were all disabled before testing. In all test runs, we rebooted the system and waited several minutes for drive activity to settle before invoking a test.
Why No IOMeter? Though we struggled for several days with IOMeter, in an effort to get consistent, repeatable results, in the end we were not satisfied that our IOMeter test procedures were producing an accurate measurement of solid state disk drive performance. Additionally, though IOMeter is good for establishing multi-user and multi-node performance (networking) of a drive, it has little bearing on single user measurements. Certain anomalies were observed during our testing which we believe were related to the fact that we were testing on Windows Vista. We are currently working with manufacturers to better understand our results and ways we can make use of IOMeter for testing in future SSD technology-related articles in the coming weeks.
SANDRA Physical Disk Write Performance
Why No IOMeter? Though we struggled for several days with IOMeter, in an effort to get consistent, repeatable results, in the end we were not satisfied that our IOMeter test procedures were producing an accurate measurement of solid state disk drive performance. Additionally, though IOMeter is good for establishing multi-user and multi-node performance (networking) of a drive, it has little bearing on single user measurements. Certain anomalies were observed during our testing which we believe were related to the fact that we were testing on Windows Vista. We are currently working with manufacturers to better understand our results and ways we can make use of IOMeter for testing in future SSD technology-related articles in the coming weeks.
|
Motherboard - Video Card - Memory - Audio - Hard Drives -
|
Hardware Used: Intel Core 2 Extreme QX6850 Asus Striker II Extreme (nForce 790i SLI Ultra chipset) GeForce 8800 GTX 2048MB Corsair DDR3-1333 CAS 7 Integrated on board Western Digital Velociraptor - OS Western Digital Velociraptor - Test 300GB - 10,000RPM - SATA 3Gb/s OCZ Core Series 64GB OCZ Standard 64GB Super Talent MasterDrive MX 64GB Mtron MSP 7400 32GB |
Operating System - Chipset Drivers - DirectX - Video Drivers - |
Relevant Software: Windows Vista Ultimate Intel 8.6.1006 DirectX 10 NVIDIA ForceWare v169.25 Benchmarks Used: HD Tach 3.0.1.0 ATTO ver 2.02 PCMark Vantage SiSoftware Sandra XII SP2 |
In our SiSoft SANDRA testing, we used the Physical Disk test suite. We ran the tests without formatting the drives and both write and read performance metrics are detailed below. Please forgive the use of these screen captures and thumbnails, which will require a few more clicks on your part. However, we felt it was important to show you the graph lines in each of the SANDRA test runs, so you are able to see how the drives perform over time and memory location and not just an average rated result.
SANDRA Physical Disk Read Performance
In our read performance test with SANDRA, the OCZ Core Series drive offered the fastest sustained read bandwidth at 117.11MB/sec, followed closely by the 32GB Mtron drive at 116.55MB/sec and the Super Talent 64GB drive at 110MB/sec. The OCZ 64GB standard SLC drive followed the pack some 25% slower at 93.33MB/sec. In all tests, read performance was very flat and robust across all SSD volumes. On a side note, notice how the WD VelociRaptor reference score in SANDRA (green line) maps in at about the same level as the standard SLC drive from OCZ and significantly below the OCZ Core Series, Super Talent MasterDrive MX or Mtron drives.
SANDRA Physical Disk Write Performance
As we expected, write performance characteristics were a bit more interesting. All of the SSDs, with the exception of OCZ's 64GB SLC drive, showed significant saw-tooth patterns with big dips and spikes in performance. This happens most likely due to chip boundaries and transitions as the benchmark attempts to write sequentially across the volume. The Mtron and OCZ SLC drives exhibited the best performance with respect to writes, due to their on-board DRAM and SRAM cache, which helps mitigate some of these chip transitions.