This is our testbed:

• CPU: Intel Core i7-920 Retail
• Motherboard: Asus P6T Deluxe BIOS 1201
• Memory: Crucial PC3-1066 3x1GB @ DDR3-1600 (CL8)
• Main HDD: WD 800JD
• VGA: MSI 7600GT Silent
• Windows VISTA x64 SP2 64bit with all the latest updates installed

We didn't perform any kind of optimizations for the operating system . Before running any benchmark, we made sure that the "enable advanced performance" option was selected.. We used the following benchmark software at their default options

• HDTachRW v3.0.1.0
• HD Tune v3.50 Pro
• Crystal DiskMark v2.2.2
• ATTO Disk Benchmark
• ASS SSD Benchmark 1.0345
• IOMeter v2006.07.27 with Xtreme Benchmark template

The first benchmark is the HDTachRW software with the long Bench test to give us an idea of the overall performance. The drive's reading/writing performance is almost linear with a small drop at the speed at the 37GB mark. The average reading speed is 231.7MB/sec, while writing seems to be also fast at 231.0MB/sec. The reported random access time is 0.2ms, which is really impressive:

The HD TunePro software reported a smoother graph for reading and a smaller drop in the speed at the 37GB mark. The software gave a 226.7MB/sec average reading speed:

The random reading tests were also very promising:

The HD Tune Pro software gave around 160MB/secs for writing - much lower than what the HDTach previously reported:

The Random write results continue to be very good:.

Getting PCMark Vantage x64 to work with the Torqx SSD drive was a real pleasure with all the results to be very good:

Below are more tests from ATTO Disk Benchmark, CrystalDiskMark and AS SSD Benchmark. All tests showed a great performance in both reading and writing I/O operations:

The drive scores 174 points at the AS SSD Benchmark:

Finally, we performed the IOMeter tests using the Xtreme Benchmark template and wrote down the results. Compared with the drives we had previously tested based on the JMikron controller (Warp v2 and Apex series of SSDs), the Torqx drive is a major performer managing to give a much higher I/O performance and significantly lower average access times. We think that the results speak for themselves: