The following are the specifications of the computer this product has been tested on.
- Crucial 2GB x 2 kit 240pin DDR3 PC3-8500
- Gigabyte S1156 Intel P55 motherboard - GA-P55-UD4
- Intel Core i5 Quad 750 2.66GHz 8MB cache
- Antec Three Hundred Black Case
- Sapphire HD4350 512MB PCIe DVI graphics card
- Samsung Black DVD+/-RW SH-S223B SATA
- Lite-On DVD+/-RW SOHW-1693S ATA
- OCZ Vertex 2 120GB SSD FW v1.23
- Samsung F3 2TB 5400RPM HDD
- ASUS U3S6 PCIe x4 USB3.0 / SATA 6Gbps
- Corsair 450W ATX PSU
- Windows 7 64-bit Enterprise Edition
- Panasonic DMC-LX3
- Nikon D60
These two cameras were used to take pictures for preparing our file sets in the file transfer benchmarks.
To test the performance of the Samsung HDD, we will be using the following test applications in this review:
- HD Tune Pro
- CrystalDiskMark x64
- (Various in-house batch file scripts)
Unlike USB pen drives where most people leave the drive formatted in the default FAT/FAT32 file system, when it comes to external hard disks, some users will reformat the drive as NTFS to be able to store files greater than 4GB. As we have noticed before, switching from FAT32 to NTFS also has a very noticeable difference in write performance due to the way Windows handles each file system.
While many users will make use of the USB3 connectivity with this drive, there are times where they will need to connect this hard disk to a USB2 port.
So to cover USB2 and USB3 connectivity as well as the two popular file systems, we will rerun each benchmark to cover each scenario:
- USB3, FAT32 formatted
- USB3, NTFS formatted
- USB2, FAT32 formatted
- USB2, NTFS formatted
Due to the variation in performance form one USB controller to another, we will conduct all tests in this review using the same USB port on the review PC’s ASUS U3S6 PCIex4 controller. For the USB2 tests, we simply connect the hard disk with a regular USB2 cable to force it to operate in USB2 mode.
For the real-world testing, we will test the following:
- Copy a large 1GB file from the RAM drive to the HDD.
- Read the 1GB file.
- Copy a large 1GB folder consisting of 8,247 JPEG files in a hierarchy of 245 folders.
- Read this entire folder’s content.
- Delete this folder.
We chose a RAM drive as the file copying source to minimise the latency as much as possible by the read source. For the read tests, the files are simply read by our script without being stored anywhere. All timings are measured by script with accuracy to 1/100th second. The computer was rebooted after each write test and prior to the delete test.
The purpose of the delete test is to show how long it takes to delete a large folder, as this task is usually quite tedious with a FAT32 file system.
Note that as all the real-world tests are conducted with a blank drive, the performance will vary depending on the amount of data the drive contains and its fragmentation.
Original File System
Before we conduct any tests, we will first take a look at the file system this drive came pre-formatted with:
As with any other modern hard disk, manufacturers rate the capacity in multiplies of 1,000 bytes, so 1TB means 1,000,000,000,000 bytes. From the above screenshot, we can see that the drive is literally just shy of this mark, but right next to this mark, we see 931GB. You may wonder, where’s the other 69GB? Well, the way operating systems measure capacity is in multiples of 1,024 bytes, where 1KB has 1,024 bytes, 1MB has 1,024KB and so on. Computers operate in binary, so 1,024 is easier to work with in binary as 2 to power of 10 gives 1,024. So if we calculate the capacity in the way computers measure a Gigabytes, we get 1,000,000,000,000 / 1024 / 1024 / 1024, which gives ~931GB, the capacity we see here.
The 234MB of data on the drive is taken up by the Samsung software that comes pre-installed on the hard disk, as shown in the following screenshot:
The following ‘CHKDSK’ screen gives some more technical information about the file system, such as the allocation unit size:
All FAT32 tests are conducted with this factory formatted file system and preinstalled data. Only when we complete these tests do we format the drive in NTFS to carry out all the tests again under NTFS.
Before the NTFS tests are conducted, we copy off the preinstalled data, format the hard disk as NTFS and copy this data back such that we are testing with the same conditions apart from the file system being NTFS.
Now let’s head to the next page, where we carry out some benchmarks…