Samsung S2 Portable 3 1TB USB3 2.5″ HDD review

Author

Senior Administrator & Reviewer
Article posted 04 Feb 11 12:19

Benchmarks

Crystal Disk Info


First we will take a look at the drive’s capabilities:

This gives an idea of the drive’s specifications, including a few not mentioned on the packaging, such as the buffer size, spindle RPM, firmware version and features.  TRIM is not supported as this is a hard disk, not an SSD. 

Despite the plastic hard drive casing, the temperature is reasonably low, considering that we took this screenshot immediately after we conducted a few hours of testing.  The ambient room temperature was 20C.

HD Tune Pro


We start by running the HD Tune benchmark in USB3 mode:

As we can see here, this drive performs as if it was installed internally, peaking at 95MB/s.  The seek time is nothing to get excited over, but still pretty good for a 5400RPM hard disk.  The minimum throughput of 40.2MB/s at the end is still much quicker than the quickest transfer we have seen over USB2.

Now let’s rerun this benchmark, but this time in USB2 mode:

Now it becomes clear just how much of an improvement USB3 has to offer, even with a 2.5” hard disk. 

ATTO disk benchmark


ATTO has become a standard tool for measuring the data throughput of hard drives and flash drives. It measures the performance of reading and writing, using different file sizes and block sizes.

The following is with the FAT32 file system operating in USB3 mode:

Now with the NTFS file system, again operating in USB3 mode:

In USB3 mode, we can again see the impressive read and write transfer rates, all the way down to 16KB transfer sizes.  The difference between FAT32 and NTFS is very small with NTFS giving roughly a 1MB/s improvement in a few cases.

Let’s see how it performs in legacy USB2 mode, again starting with the FAT32 file system:

Now with the NTFS file system, again connected by USB2:

Once again, we can see the dramatic difference between USB2 and USB3 performance.  In USB2 mode, the difference between FAT32 and NTFS is negligible with this benchmark.

CrystalDiskMark 3.0


Crystal Disk Mark is quite a handy benchmarking application, as it focuses on the file sizes that can cause a problem on a system drive and external drives. 

Let’s start with the FAT32 file system, operating in USB3 mode:

Now with the NTFS file system, again operating in USB3 mode:

Another run, this time in USB2 mode, again starting with the FAT32 file system:

Now with the NTFS file system, again operating in USB2 mode:

Let’s see these results more clearly as graphs:

 

USB3
FAT32

USB3
NTFS

USB2
FAT32

USB2
NTFS

Seq Read

100.1MB/s

100.1MB/s

36.24MB/s

35.96MB/s

Seq Write

99.15MB/s

98.48MB/s

37.34MB/s

37.34MB/s

512K Read

35.96MB/s

35.86MB/s

23.24MB/s

22.6MB/s

512K Write

17.14MB/s

36.03MB/s

17.71MB/s

36.25MB/s

4K Read

0.439MB/s

0.433MB/s

0.44MB/s

0.435MB/s

4K Write

0.436MB/s

0.726MB/s

0.45MB/s

0.796MB/s

4K QD32 Read

0.536MB/s

0.536MB/s

0.524MB/s

0.527MB/s

4K QD32 Write

0.425MB/s

0.724MB/s

0.444MB/s

0.758MB/s

It is interesting to see this benchmark break the 100MB/s mark with this hard disk in USB3 mode, showing once again very impressive results for an external 2.5” hard disk.  Another thing this test shows is just how much of an effect the file system has on write performance.  With the 512KB block size, the NTFS file system gave just over double the write performance of FAT32, despite everything else including the hardware being identical.  Even random 4K performance got a noticeable boost from the NTFS file system.  In fact, for small random writes, switching to the NTFS file system has a more dramatic effect than moving to USB3, at least in this benchmark.

Unsurprisingly, we see a major drop in sequential throughput moving from USB3 to USB2. 

Summary


The read and write transfer rates are excellent for a USB external hard disk, with the USB3 clearly showing its advantage in our benchmarks so far.  While smaller random transfers may not be up to par with other hard disks, it is worth nothing that this will not be used for running the operating system from.  On the other hand, we saw that simply changing the file system from FAT32 to NTFS gave us a dramatic improvement in the writing performance of random IOPs, even in USB2 mode.