Real world tests
As with benchmarking SSDs and other storage
devices, it is pointless conducting an endless amount of benchmarks on a hard
disk. Real users may conduct a few benchmarks when they first get their hard
disk, but most users simply want a hard disk that performs well in everyday use.
The benchmarks have so far shown us that
this hard disk provides very good sequential throughput for a 2.5” hard disk
and reasonably good performance with small random read and write transfers. Now
let’s see how this drive performs in the real world.
Real world single file copy test
With the vast storage capacity offered by
modern hard disks, most people keep large files stored on their internal hard
disk. When it comes to copying these files such as to an external hard disk to
backup, this can be a time consuming process. So in this test, we will measure
how long it takes to transfer a 1GB file to and from the hard disk.
For the write test, we stored the source
file on a RAM Drive to prevent the read source becoming a bottleneck. We then
used FastCopy to transfer the file to the HDD. We have FastCopy configured to
prevent written data being cached, so the moment this utility reports the copy process
as finished, there is nothing waiting in the Windows cache to be written,
unlike a copy & paste in Windows Explorer.
As already shown by the sequential tests on
the last page, this hard disk is pretty fast for a mechanical 2.5” HDD. Even at
the 50% (500GB) mark, its write speed is just shy of the Samsung M7 320GB 2.5”
HDD at the 0% mark, making it a fast HDD for bulky file transfers especially in
the first 500GB. However, as an alternative for a desktop 1TB HDD, the Samsung
F3 1TB is still significantly quicker especially towards the end of the hard
Now let’s read back this file. For this
test, we used a simple script that reads the file without storing it anywhere.
The PC was rebooted to ensure the file was not present in the Windows cache.
The Toshiba HDD reads slightly quicker than
what it writes at, but otherwise this graph is pretty much identical to the
Write Transfer graph for the comparison drives.
Now let’s see how this hard disk performs
when this file is copied from the hard disk back to the same hard disk. For all
four partition tests, the source file was stored in the first partition. We
used FastCopy in “Diff HDD mode” for this test. By choosing this mode, the hard
disk is faced with reading and writing data at the same time. Normally when
copying a large file such as in Windows Explorer, part or all of it is stored
in RAM to reduce the amount of seeking the HDD has to do. So this test
basically shows the “worst case” scenario for making a copy of a file on the
same HDD, such as when the file is considerably larger than the amount of free
As we can see here, the Toshiba hard disk
does very well in this test, significantly outperforming the Samsung M7 2.5”
HDD and just a little slower than the Samsung F3 desktop HDD.
Real world small files copy test
Not everyone uses a large hard disk for
bulky files. In fact, some may store years of photos and documents, many of which
could range from a few kilobytes to a few megabytes per file. So in this test,
we time how long it takes to copy 8,247 photos in a hierarchy of 245 folders
totalling 1GB to and from the flash drive.
Like the previous test, these files are
stored in a RAM disk as the source and we used FastCopy to measure the timing.
After the write tests, we rebooted the PC and used a script to read the full
file set from each partition. It is worth noting that this test shows the hard
disk in its “best case” scenario, as the write process results in the folder’s
files being physically stored sequentially, which means there is no
fragmentation and the files are physically lined up one file after another. We
will carry out disk fragmentation tests later in the review where the files are
physically stored in a random order.
The following are the results:
Total time taken to write the file set
Total time taken to read the file set
As we mentioned many times before in past
reviews, the write test here clearly shows why one should not rely on synthetic
tests alone, even for a mechanical hard disk. Based on this write test, this
Toshiba hard disk not perform well when writing a large number of files such as
importing photos or installing a large software package.
On the other hand, when it comes to reading
back small files, this hard disk performs very well, with the 500GB mark also
out-performing the Samsung M7 at the fastest section of the hard disk. This
means the hard disk will be fairly quick for copying or backing up small files
off of, such as to disc or external HDD, at least with little to no file
Now let’s copy this file set from the hard
disk back to the same hard disk in each partition. Like the large single file
transfer test, we used FastCopy in “Diff HDD Mode” with the source files in the
first partition for all four partition tests:
Like the write test, the Toshiba HDD
performs well behind the others when making a copy of this file set, taking
nearly 2.5 times longer than even the Samsung M7 HDD. This means that this hard
disk will feel slow when extracting file archives or performing Windows
Updates, where the files are being both read and written to the same HDD.
Real world MP3 files copy test
Many users like to store their entire music
collection on their PC, usually in the MP3 format. This file size is also typical
of JPEG images from the latest digital cameras due to their high megapixel size,
so this test would also give an idea of what the HDD is like for reading and
writing large JPEG files.
Like the previous tests, these MP3 files
are stored in a RAM disk as the source and we used FastCopy to measure the
timing. All MP3s were stored in the same folder. The PC was rebooted after the
write tests before running our read script.
The following are the results:
Total time taken to write the MP3 file set
Total time taken to read the MP3 file set
Like the small files test, the Toshiba HDD
is significantly slower than the rest when writing multiple files, even here
where the average file size is just over 5MB. This could potentially be a
problem for photographers and those interested in regularly writing files
around this size. Read performance on the other hand is excellent, so even
where a large number of MP3s and JPEGs are stored, the hard disk has no problem
reading them quickly, at least without any significant disk fragmentation.
Now for the copy test, again performed with
FastCopy in “Diff HDD Mode”:
Despite the slow writing for this file
size, the Toshiba’s faster throughput helps put it ahead of the Samsung M7
320GB, but only by a small margin. Interestingly, the Toshiba’s file copy times
are barely longer than the sum of the dedicated read & write timings, so
this copy process is unlikely to perform noticeably quicker where the data is
read and written in separate stages of the copy operation.
Real world file deletion test
One issue we noticed in the past,
especially when comes to flash drives and even external hard drives is the
amount of time it takes to delete a large folder, in some cases taking several
minutes with a slow drive.
So for this test, we simply timed the
duration it took to delete the hierarchy of 8,247 files and 245 folders from
each partition of the hard disk. The PC was rebooted before we ran this test to
clear the cache. The timing was measured by a script.
Unlike our external drive reviews, deleting
files is a fairly quick process on this hard disk and the two comparison
drives, taking just over 7 seconds to delete this large hierarchy of folders.
This also means that this hard disk should be relatively quick when performing
disk cleaning tasks, such as running CCleaner or just simply emptying the web
browser’s temporary Internet files, which often involves deleting tens of
thousands of files.
The Toshiba HDD is a pretty quick hard disk
reading files, ranging from small thumbnails to large video files. In many of
the tests, it reads quicker at the 50% (500GB) mark than what the Samsung M7
320GB reads at the 0% mark, the fastest section of the hard disk. This means
that this hard disk should perform well where fast read speeds are important,
such as writing a set of files to recordable disc (or to an external drive),
performing a scheduled virus scan or even making a backup from it.
While the Toshiba HDD’s sequential write
performance is similar to its read performance, it has poor performance when
writing small files, taking over 4 times longer to write our set of small files
than the comparison Samsung M7 2.5" hard disk. Even writing files around
5MB in size such as MP3s or large JPEGs takes nearly double the length of time
of the comparison drive. This means that this hard disk will not perform well
when importing small files such as photographs or installing software where a
large number of small files are written. Later in this review, we’ll see what
effect this has when installing Windows 7 in VirtualBox with the virtual hard
disk image placed on this hard disk.
When copying files from the hard disk back
to it, including one partition to another, the copy time is just over the
length of time it took during the dedicated read test plus the dedicated write
test. With MP3s, it performed just slightly quicker than the Samsung M7, but considerably
slower when faced with copying small files. In the real world, this hard disk
will likely take longer than competing drives when copying small files from one
partition to another or even extracting an archive containing many small files.
Let’s head on to the next page to look
at how this hard disk performs with fragmentation and multitasking …