|
|
Review: Intel 750 NVMe PCIe 1.2TB SSD Reviewed by: Wendy Robertson Provided by: Intel Firmware version: 8EV10135 |
Today I'm looking at something new. In fact this article
will feature the world's first SSD, aimed at the professional consumer, to utilise
the new NVMe (Non Volatile Memory Express)
interface. NVMe SSDs are PCIe based and are installed in a standard PCIe slot,
or via the brand new U.2 connector. PCIe SSDs are not new, and have been around
for several years. However, the PCIe SSDs of the past required a special
controller which sat between the SSD hardware and the PCIe system bus, to allow
SSD hardware and the PCIe bus to do the translation and communication between
the two interfaces. This was of course a very complex and time consuming task,
which inevitably led to increased latency.
NVMe is a native solution, with its own highly optimised
protocol, which features a very much reduced command set, much lower latency
when compared to AHCI, and is specifically optimised for Non Volatile Memory
(FLASH memory).
Intel was kind enough to send me one of their brand new 750
series NVMe SSDs for review. In this case the 1.2TB PCIe version. There is also
a 400GB version available, as well as U.2 connection versions.
So let's find out how this new SSD performs in our range of
tests.
Intel company information
Intel should need no introduction, but those of you who
would like to find out more about Intel, can do so at their website.
The Intel 750 NVMe 1.2TB SSD
The review sample provided was the PCIe version, which is
PCIe gen3, and uses 4 PCIe gen3 lanes for data transfer.
You simply install the Intel 750 NVMe SSD into a PCIe
generation 3 PCIe slot with at least 4 PCIe lanes available, but for best
results, you should install the Intel 750 NVMe SSD into a generation 3 x16
socket which is connected directly to the CPU.
Please note: To be able to boot from the Intel 750 NVMe SSD,
you will require a motherboard which allows booting from NVMe. Many motherboard
manufacturers have already updated their Z97 chipset motherboards to support booting
from NVMe. Those with X99 chipset motherboards, should find they all support booting
from NVMe.
There is apparently a workaround for motherboards which do
not support booting from NVMe, to allow booting via EFI mode, but I have not
tried this.
SSD controller.
The SSD controller used in the Intel 750 series of SSDs is
the mighty Intel CH29AE41AB0, which boasts 18 channels to the NAND array, and
is in fact the same controller found in the Intel DC-3700 enterprise class
SSDs. The firmware is of course optimised towards client use rather than the
enterprise sector.
NAND.
The NAND is Intel/Micron 20nm MLC.
Intel SSD Toolbox.
The toolbox allows the drive's firmware to be updated,
secure erased, optimised, and drive diagnostics. The toolbox also offers
utilities to tune the PC for best performance, displaying system information,
and viewing the Intel 750's S.M.A.R.T. data.
If you are running Windows 8.1 or Windows 10 then NVMe SSDs
are supported natively, and there is no need to find and install NVMe drivers.
Intel however has made available their own optimised NVMe drivers, which offer
improved performance, and allow some of the more specialised features of the
Intel SSD Toolbox, such as secure erasing the Intel 750, to function properly. The
Windows 8.1 and Windows 10 native drivers are more than enough to allow you to
do a clean install of the operating system.
Specifications.
Does the SSD support TRIM?
To allow TRIM to function you first need an SSD that
supports the TRIM command. You then need a storage stack that will allow the
TRIM command to pass-through to the SSD, and this includes the driver.
Thankfully this is now very easy to check with some degree
of reliability, using a small utility written by Vladimir Panteleev called TRIMCheck.
According to TRIMCheck, TRIM is functioning correctly on the
Intel 750 NVMe 1.2TB SSD.
Let’s head to the next page where we take a look at our
testing methods and the review PC....
Test machine
For this review I will be using a computer with the
following configuration:
Hardware:
- Motherboard: Asus Z97 Deluxe (Intel Z97 chipset)
- Processor: Intel 4th generation Core i7 4770K
- CPU cooler: BeQuiet Dark Rock Pro 2
- RAM: 16GB Samsung Green DDR3 1600MHz (dual channel)
- GFX: Onboard Intel HD 4600
- Sound: Onboard Realtek ALC1050 HD audio controller
- Hard disk OS: OCZ Vector 256GB SSD.
- Case: Antec Performance One P280
- PSU: Antec True Power modular 550W
- Display: Dell UltraSharp U2412M 24” widescreen IPS LCD (HDCP
compliant) - Operating System: Windows 8.1 Professional 64bit
- Power consumption testing equipment: Quarch Technology QTL1824-02 XLC
Programmable Power Module
AHCI mode was also selected for all drives in the UEFI of
our test PC, and all tests were carried out in this mode. The SATA 6Gbps drivers
used on our review PC were the Intel Rapid Storage Technology (RST) Version
13.2.4.1000.
The NVMe drivers used to test the Intel 750 SSD were Intel's
own NVMe driver version 1.2.0.1002.
CPU power saving states were disabled for consistency, and
all the SSDs in this article were tested with all CPU power saving states
disabled.
Test applications
To test the performance of the Intel 750 NVMe SSD, I will
be using the following test applications in this review.
- HD-Tune Pro
- ATTO
- Iometer
- AS SSD
Benchmark - CrystalDiskMark
- MyCE Reality Suite
- Anvil’s
Storage Utilities - PC
Mark 8 - Quarch Technology - Test
Monkey 2
Test procedures
I will start off our testing procedures explanation by stating
that I did not run many synthetic benchmarks on the Intel 750 NVMe SSD. You may
ask why I have run so few synthetic benchmarks?
SSD technology has moved so fast in the last couple of years,
that basic synthetic benchmarks alone are now of very limited use, as they don't
really tell us much about performance and how the drive will behave in the real
world. I have therefore decided to show some basic benchmarks of the Intel 750
NVMe SSD, and will complement this with advanced benchmarks using IOMeter and
AS SSD benchmark. I will also show how the Intel 750 NVMe SSD performs in the real
world with our Myce Reality Suite test.
The reality of SSD performance
Whilst I can easily show you which SSD is technically the
faster, when you use one of these modern SSDs as an operating system drive it
becomes very difficult to tell them apart as far as performance is concerned.
A typical use of a small capacity SSD at the moment is to
have your operating system and applications installed onto the SSD. The
performance difference compared to a traditional HDD is enormous, however when
you start to compare SSD to SSD the difference becomes almost impossible to
detect.
Let’s look at why this is the case.
Drive A can boot to the desktop in 8.11 seconds, and drive B
can boot to the desktop in 8.12 seconds, the difference in time is
milliseconds, and can one really tell the difference?
The fact is, all modern SSDs are only ticking over when they
are only running the OS and launching applications, it’s only when you get to some
of the larger capacity SSDs, with enough free space to be able to hold the
actual data that you’re going to be working with, be that video, audio or
pictures, for example, that you actually get a tangible difference in
performance. This is where the SSDs with the better sequential performance start
to pull well ahead of the SSDs which have lower sequential read/write
performance.
Small file random IOPS vs sequential performance
IOPS
This is a fairly complex subject, but I will do my best to
explain things in a manner that is easy to understand.
The term IOPS is the amount of input or output transactions
that can take place in a one second interval, so for example, if an SSD is
quoted as being able to cope with 20,000 4K random write IOPS, then the SSD should
be able to cope with 20,000 input transactions in a period of one second. If
the same SSD is said to be able to produce 20,000 4K random read IOPS, then the
same SSD should be able to produce 20,000 4K random read output transactions in
a one second interval.
Ok, now we have some figures to work with, the next question
is how many IOPS are actually required?
This will depend on your usage pattern. If you are a typical
desktop user who browses the internet, does some word processing or perhaps
some audio or video editing, and perhaps plays a few games, then in actual
fact, you don’t need to have massive 4K random read/write performance. The
actual amount of 4K random performance that is required for a fast and smooth
running system for a desktop user with a usage pattern similar to the above
will be well under 1,000 4K IOPS.
On the other hand, if the SSD is being used for running a
large and complex database server, then 4K random performance is the absolute
measurement of how fast that server will run, as this type of application does
most of its input and output transactions in the 4K domain.
So why would I need an SSD with 80,000 4K IOPS for a
desktop?
In fact you don’t need this type of performance for a
desktop, but an SSD which is capable of coping with 80,000 4K IOPS will be
faster than an SSD which can only cope with 20,000 4K IOPS.
OK, I just said if under 1,000 4K IOPS are actually required
for typical desktop usage, why is an SSD with 80,000 4K IOPS faster than an SSD
with only 20,000 4K IOPS, confused?
You may ask, if I only require 1,000 4K IOPS surely the rest
is wasted?
While you may never need 80,000 4K IOPS, IOPS is all about
latency. The reason that an SSD can cope with as much as 80,000 4K IOPS is
because latency in this domain is very low. With 4K files, even if you require
to process 500 of them at the same time, you are not talking about a huge
amount of data, it has far more to do with how long it takes the SSD to process
a single file, and the amount of time required to process a single 4K is all
about how long it takes for the SSD to access or store that data before it can
move on to the next transaction.
In other words an SSD with 80,000 4K IOPS performance will
handle those 500 files faster than the SSD with 20,000 IOPS.
So how will a desktop user even notice this faster speed if
so little 4K random IOPS and data are actually used?
Multitasking is a good example. The more tasks you run at
the same time, you more you will notice the speed difference.
Sequential performance
I have always maintained that sequential performance was
every bit as important as small random file performance for a desktop SSD. To
me this was always so obvious for a desktop user. For example, let’s say you
want to launch an application or game. Both have some fairly large files to
load, and also a great many small files, but the point is, even the smaller
files are sequential in nature. Now let’s say you’re into audio or video
editing. Video files tend to be huge, and the files are written or read
sequentially. Isn’t this how many users are using their PCs these days?
Summary
So how does this shape up in the real world? Which is
better, massive 4K IOPS or massive sequential performance?
In an ideal world you want both, as an SSD with massive
random 4K IOPS and sequential performance will always be faster than an SSD
that has high sequential performance and moderate 4K random IOPS performance,
and the same applies to an SSD that has massive 4K random performance and
moderate sequential performance. The SSD which has high performance in both
patterns will always be the faster SSD.
However, you can still have an SSD that is very fast for
desktop use that has moderate random 4K performance and massive sequential
performance, the same can be said about a drive having massive random 4K
performance and moderate sequential performance, as it is about getting the
balance right if you have to compromise on one or the other.
Test drives
- Plextor M6e PCIe 256GB SSD
- OCZ REVODrive 350 PCIe 480GB
SSD - Intel 750 PCIe NVMe 1.2TB SSD
Drive preparation for running the tests
All the SSDs used in this article were in a clean and fresh
state when the testing period started. From then on, each drive had to rely on
its own NAND cleaning effectiveness for the remainder of the tests.
The Intel 750 PCIe NVMe 750 SSD was connected to the first
generation 3 x16 PCIe socket on my test PC, and all tests were carried out the
drive connected to this socket. PCIe gen3 support was enabled in the UEFI of
the test PC, and boot from NVMe was also enabled for that PCIe socket.
- All SSDs used in this article had their partitions aligned
to the Windows 8.1 x64 defaults.
Where I use graphs in this article to display results, I
will use the following colours to make it easier, for our readers to see which drive
we are reviewing.
Intel 750 NVMe SSD
Comparison SSD
Now let's head to the next page, where I look at some
basic benchmarks...
Synthetic Benchmarks
HD Tune Pro
In this benchmark I am checking sequential reading speed.
Intel 750 NVMe 1.2TB
SSD
With an average sequential reading speed of 2519 MB/s the Intel
750 NVMe SSD gives an outstanding turn of speed. Also worth noting are the
incredibly fast access times.
Let's see how this compares to other recently tested SSDs in
the table below.
The Intel 750 NVMe SSD has performed extremely well in the
HD Tune sequential reading test.
ATTO disk benchmark
ATTO has become a standard tool for measuring the data
throughput of HDDs and SSDs. It measures the reading and writing performance,
using different file sizes and block sizes.
Intel 750 NVMe 1.2TB
SSD
The reading speed results for the Intel 750 NVMe 1.2TB SSD
are extremely impressive, topping out at over 2.68 GB/s, and writing speed is
also impressive topping out at over 1.35 GB/s.
Let's find out how this compares with other recently tested
SSDs.
ATTO Reading performance
ATTO - Reading
performance at various block sizes
The Intel 750 NVMe SSD is the fastest SSD when reading data.
ATTO Writing performance
ATTO - Writing
performance at various block sizes
The Intel 750 NVMe SSD delivers excellent writing performance,
but it isn't as fast as the OCZ REVODrive 350 when writing larger block sizes.
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.
As we can see from the above screenshots, sequential reading
and writing speeds are both extremely impressive, also random reading and writing
performance, at low and high queue depths, is outstanding.
AS SSD Benchmark
AS SSD benchmark is a benchmarking tool specifically
designed to test SSDs. The application tests sequential reading and writing
performance, 4K random reading and writing performance.
AS SSD benchmark also tests 4K threaded performance. This is
very exciting, as this test is the first available test that I am aware of,
that simulates how a PC operating system actually works. A modern PC and OS,
such as Windows 7/8 does not just run a single thread at a time, it runs many
threads. The AS SSD benchmark "4K 64Thrd" tests run 64 threads
simultaneously throughout the test. If this result is good, then you can be
pretty sure the drive will perform extremely well as a system drive.
After the tests complete, AS SSD benchmark derives a total
score for the drive being tested. This is based on all aspects of the test
results, and gives an indication of how the drive is performing overall.
Now let’s look at the result from the Intel 750 NVMe SSD in
the form of a screenshot. All our other comparison drives’ results are
presented in the form of a graph.
Intel 750 NVMe 1.2TB
SSD
As we can see from the AS SSD test run, the Intel 750 NVMe
SSD has outstanding reading and writing performance, finishing in first place,
well ahead of the other SSDs in this test.
Summary:
The Intel 750 NVMe SSD has performed extremely well in the
basic synthetic benchmarks. Random reading and writing performance is extremely
impressive. Sequential reading and writing performance is outstanding.
Let's head to the next page for our IOMeter test
results.....
I/O Performance
There is little point of having an SSD drive that has
blazing sustained reading and writing speeds, if the drive can't handle reading
and writing of small random files. If you intend to use your new SSD drive to
store and run your operating system, then the drive must be able to cope with
the many small random files that Windows will write to the drive continually.
So I feel it is very important to test how many of these random files that a
drive can handle in one second. I believe that anything over 1,000 I/O’s per
second would be enough for most users running a consumer grade mainstream PC,
and should provide a smooth running system. But obviously, the more I/O's that
a drive can handle, the faster the drive will feel and leave more headroom for
those huge multitasking sessions that users sometimes engage in.
IOMeter is probably the most versatile of all the synthetic
benchmarks. Its ability to be configured to generate a multitude of different
I/O traffic is unmatched. Another great feature of IOMeter, is the capability
to test any storage metric that you can think of, providing you know how to
configure the assignments. The reviewer also has complete control over things
like queue depth, block size, whether the traffic is random, sequential, or
even a mixture of both.
Partition alignment and sector boundaries
Windows 8.1, Windows 7, and Windows Vista will automatically
align a partition to 4k boundaries during partition creation, Windows XP won’t.
It is imperative that an SSD’s partition is aligned. Windows XP is also
restricted to sector boundaries, while Windows 7 and 8 will use 4k boundaries
if they can. The Intel 750 NVMe SSD is 4k boundary aware, and will use these
boundaries if possible. Of course it will also remap LBAs for compatibility
with the sector boundaries so that the drive can be used with Windows XP.
IOMeter allows us to set the sector boundaries for
conducting the tests, and I have therefore set the sector boundaries at 4K,
which means the IOMeter tests are valid for Windows 7, Windows 8, and Windows
Vista users. XP users will not be able to obtain such results.
I will provide a screenshot of the tests on the review drive
for those of you who like to see the actual test result. All the comparison
drive results are represented in the form of graphs.
If any of you would like to see a screenshot from any
IOMeter test on a particular drive, please feel free to request one, and I’ll
post the screenshot in the forum thread.
All the IOMeter tests create a 10GB data set on the target
drive, and each test is run for a duration of 3 minutes.
IOMeter 4K random write test with repeating data.
The first test involves creating continual 4KB random files
on the target drive with IOMeter. I use a 4KB file size, as it is believed that
Windows will create and modify many of this size of file constantly in the
background during a typical Windows session. It is said that most 4K random
writes take place at a queue depth of only one, and I have been requested to
include this test in my reviews.
Queue depth 1
Intel 750 NVMe 1.2TB SSD
– 4K random write (QD 1)
At 299.25 MB/s the Intel 750 NVMe SSD's performance is
outstanding, and it finishes this test well out in front.
Our next test involves creating continual 4KB random files
on the target drive with IOMeter. I use a 4KB file size, as it is believed that
Windows will create and modify many of this size of file constantly in the
background during a typical Windows session. I will use queue depths of 4 and
32 for these tests.
Queue depth 4
Intel
750 NVMe 1.2TB SSD (QD 4)
At a queue depth of 4, the Intel 750 NVMe SSD is phenomenal,
and finishes well ahead of the other SSDs in this test.
Queue depth 32
Intel
750 NVMe 1.2TB SSD (QD 32)
At 786.1 MB/s, the Intel 750 NVMe 1.2TB SSD is once again
outstanding, and finishes this test in first place, but the OCZ REVODrive 350
at this queue depth is hot on its heels.
IOMeter 4K random write test with fully random data.
This test is exactly the same as the test above except that
the test data is fully random and is therefore much more difficult to compress.
This test was requested as SandForce based SSDs gain a lot of performance by
being able to compress data on the fly. While the above test shows the
SandForce based SSDs in a best case scenario, the following test will show the
SandForce based SSDs in a much more realistic scenario.
Queue depth 4 with fully random data
Intel 750 NVMe SSD –
4K random write (QD 4 with fully random data)
The Intel 750 NVMe SSDs pay no penalty when writing data
which is incompressible.
4K random write queue depth profile
For this test I used various queue depths from 1 – 32 to
give you an idea how this SSD performs at different queue depths. For a normal
desktop user, with lightweight multitasking, the queue depth will rarely rise
above 2. For heavy multitasking, the queue depth is unlikely to rise above a
value of 8.
The results are shown below.
As we can see, the Intel 750 NVMe SSD has outstanding
performance at low queue depths but, after reaching a queue depth of 4,
performance doesn't increase with higher queue depths. One should note, that
with this level of performance at low queue depths, the fact that the Intel 750
doesn't really scale well after a queue depth 4 is not a problem.
Below I present a table of the results in more detail.
IOMeter 4K random read test.
If there are many 4k files created, then that must also mean
that many 4k files need to be read. This test measures 4k reading performance.
It is said that most 4K random reads take place at a queue
depth of only one, and readers have requested that I include this test in my
reviews.
Queue depth 1
Intel 750 1.2TB SSD -
4K random read (QD 1)
In this test the Intel 750 NVMe SSD has outstanding performance,
and finishes in first place.
Queue depth 4
Intel 750 NVMe 1.2TB SSD
- 4K random read (QD 4)
At a queue depth of four, the Intel 750 NVMe is outstanding,
and is the fastest SSD in this test.
Queue depth 32
Intel 750 NVMe 1.2TB SSD
- 4K random read (QD 32)
At a queue depth of 32, the Intel 750 NVMe SSD is once again
showing outstanding performance, and finishes this test well ahead of the
competition.
4K random read queue depth profile.
This test shows how the review drive scales with increasing
queue depths.
Below I present a table of the results in greater detail.
IOMeter 512KB sequential write test with repeating data.
Sequential writing performance is also very important; in
this test sequential writing performance is measured.
Intel 750 NVMe 1.2TB SSD
- 512K Sequential write with repeating data
The Intel 750 NVMe SSD gave an excellent turn of speed, but
the OCZ REVODrive 350 is well out in front.
512K sequential write - Queue depth profile
While most sequential writes will rarely rise above a queue
depth of two, it has been noted from SATA analyzer traces that with more
demanding tasks, queue depths can rise very close to a queue depth of four.
This is why I now include queue depth profiles for sequential read and write.
512K sequential write
- Queue depth profile
Below I present a table of the results in more detail.
IOMeter 512KB sequential write test with fully random data.
This test is almost exactly the same as the test above
except that the test data is fully random in nature. This test was requested as
SandForce based SSDs gain a lot of performance by being able to compress data
on the fly. While the above test shows the SandForce based SSDs in a best case
scenario, the following test will show the SandForce based SSDs in a more
realistic light. In the real world, the data is neither 100% incompressible nor
100% compressible, it is somewhere in between. So please keep this in mind.
Intel 750 NVMe 1.2TB SSD
– 512K sequential write with fully random data
With data that is not so easy to compress, the SandForce SF-2282
based OCZ REVODrive 350 took a big performance hit, whilst the Intel 750 NVMe
SSD retains its writing performance, and with an outstanding 1324.08 MB/s is
easily the fastest SSD in this test.
IOMeter 512KB sequential read test QD1.
This test measures 512k sequential reading performance at
very low queue depths.
Intel 750 NVMe 1.2TB SSD
– 512K sequential reading test (QD 1)
The Intel 750 NVMe SSD has excellent sequential reading
performance at very low queue depths, finishing in second place in this test.
IOMeter 512KB sequential read test (dual threaded).
This test measures 512k sequential reading performance QD2.
Intel 750 NVMe 1.2TB
SSD – 512K sequential reading test (QD 2)
At a more realistic queue depth the Intel 750 NVMe SSD is still
showing outstanding sequential reading performance, and finishes this test in
first place.
512K sequential read - Queue depth profile
While most sequential reads will rarely rise above a queue
depth of two, it has been noted from SATA analyzer traces that with more
demanding tasks, queue depths can rise very close to a queue depth of four.
This is why I now include queue depth profiles for sequential read and write.
Please note that in the following graph, I do not have the
lowest possible score set at zero. This is purely to allow the graphs to be
easier to read, but starting with a lowest possible score other than zero,
gives the impression that there are large differences between competing SSDs with
regard to performance, so please keep this in mind.
512K sequential read
- Queue depth profile
Below I present a table of the results in greater detail.
IOMeter Workstation simulation (outstanding I/Os = 64).
When running applications you will find that there is a
mixture of small random files and larger sequential files, being created and
read. Not only that, it isn’t just one file at a time. In this test I measure a
simulated workstation pattern, with a queue depth of 64 (threaded).
Intel 750 NVMe 1.2TB
SSD – Workstation simulation
The 'workstation' simulation sorts the men out from the
boys, with its mixed reads and writes. This test shows how an SSD could behave with
a heavy workload, in a graphics, or video workstation environment. The Intel
750 NVMe SSD has phenomenal mixed read/write performance, and finishes the test
in first place, well ahead of the other SSDs.
Summary
All in all, the Intel 750 NVMe SSDs has performed extremely well
in our IOMeter tests. The Intel 750 NVMe SSD has outstanding reading and writing
performance, and in most cases it has proved to be the fastest SSD in these
tests.
Now let’s head to the next page where we will look at how
the Intel 750 NVMe SSD performs using a new benchmarking application....
Anvil’s Storage Utilities
As well as performing SSD endurance tests. Anvil’s Storage
Utilities has a very nice SSD benchmarking application. The SSD benchmark tests
many different aspects of SSD performance, including 4K random at different
queue depths, and also sequential performance, but more importantly than this,
all using real test data.
Another very nice feature of Anvil’s SSD benchmark is the
fact that you can change the compression levels of the test data. The
compression levels of the datasets used for the tests can be varied from 0%
compression right up to 100% compressed data, and there are even a few data
profiles already included, such as database (8%) compression, and also an application
profile (46%) compression, which is designed to simulate real application data
being read and written to the SSD.
I will include a screenshot of the review drive, and all
comparison results will be presented in the form of graphs. If you would like
to see screenshots of the test results obtained on the other SSDs in this
article, you can do so by following the link here.
I will also be testing three different compression profiles,
which are as follows.
- 0 fill (100% compressible data)
- Application simulation profile (46% compressed)
- 100% (incompressible data)
So let’s begin the tests.
0 fill
Intel 750 NVMe 1.2TB SSD
(0 fill)
In the 0 fill test, the Intel 750 NVMe SSD has performed extremely
well, and is the fastest SSD in this test.
Application profile
Intel 750 NVMe 1.2TB
SSD (application profile)
The application test pattern is much more realistic in terms
of the type of data that real users will employ, with the Intel 750 NVMe SSD
once again being the fastest SSD in this test.
100% incompressible
Intel 750 NVMe 1.2TB
SSD (100% incompressible)
With test data that can't be compressed at all, the Intel
750 NVMe SSD is still exceptional, and finishes the test in first place.
Summary
One should keep in mind that although Anvil’s Storage
Utilities SSD benchmark is a very good benchmark, and tests many aspects of SSD
performance, ultimately it is demonstrating which SSD is technically the
fastest when reading data, and this may not be showing (for example) which
drive will be fastest in the real world with a home user's work pattern.
The Intel 750 NVMe SSD has however performed exceptionally
well in Anvil's SSD benchmark tests.
Now let's head to the next page for some real world tests....
It has become clear that simply conducting endless
benchmarks on SSD drives is pointless. Real users may run a few benchmarks when
they first fit their SSD drive, but most users just want a drive that performs
well in the real world. They want their drive to work "out of the
box" and run fast and smoothly.
Most of the latest SSD drives can deliver very fast
sustained reading and writing speeds, but these alone tell you very little
about how the drive will perform in the real world.
If you intend to use your SSD as your primary system drive,
with an operating system and applications installed and running from the drive,
real world performance becomes much more important than just fast sequential
read and write speeds.
Real world copy
tests
I will now conduct a few real world copy tests. These tests
simulate what real people do with their drives. I will be conducting writing
tests, using a large single file, and I will then round off the tests by
copying a folder of MP3 audio files, and also a folder of JPG pictures.
In past reviews I simply used Windows copy and paste to copy
the files from one drive to the target drive, and then I measured the time
taken to complete the test with a stop watch. This method was flawed in a
couple of ways. Windows employs a cache, so even when the files had been
copied, some of the data was still in the Windows cache and hadn't yet been
written to the SSD. The other flaw was that a stop watch is not a very accurate
way of measuring the time taken to complete the test.
I had also noticed that copying the small file set had
become pointless, as most modern SSDs have a rather large cache, in fact large
enough to be able to take the complete file set in this cache without having to
commit that data to NAND before the test had completed. I could have increased
the amount of data in the test, but I felt this was moving away from the real
world. For example, who would copy 2GB of data containing only very small
files?
I concluded it was perhaps better just to drop this test
completely, and just focus on the large 8GB ISO file, the folder of MP3 audio
files, and the folder of JPG picture files. I also have taken the opportunity
to increase the amount of data to be copied in the MP3 and JPG tests, to make
sure the SSD's memory cache doesn't obtain an unfair advantage.
The other change is that I now use an application to copy
the data, which also times how long it takes to complete the test. This application
also supports "cache write-through". What this basically means is,
there is now no caching of the files, and instead the data being copied must be
committed to the target SSD as it's being copied.
Obviously making such changes to the methods of testing is
not taken lightly. To make changes means a lot of extra work, as all the
comparison drives have to be re-tested with the new method. However, here at
Myce.wiki, we believe we should always try to improve our reviews, and if that
means updating the testing methods and some initial extra work, then that benefits
the Myce community as a whole.
For the reading drive, I have made the switch to a RAMDisk.
With SATA Express and NVMe PCIe SSDs just around the corner, the OCZ REVODrive
X2 would no longer be fast enough to supply data to a SATA Express or NVMe PCIe
SSDs. Because RAM has lower latency and higher transfer speeds when compared to
an SSD, this has meant having to rerun the tests on a selection of other SSDs
to make sure the results are up to date. Please note, that some SSDs which were
on loan during the review period, has meant that these SSDs still use the old
results, simply because I can't retest them.
For the tests themselves, I will show a screenshot of the
copy test for the SSD that I'm reviewing. All other results will be presented
in the form of a graph, so you can easily compare the results.
Single large file writing test (8144.6MB)
For this test I used a single DVD9 ISO file which had been
copied to the RAMDisk. The file was then copied to the Intel 750 NVMe SSDs and
our comparison drives.
Intel 750 NVMe 1.2TB
SSD
The Intel 750 NVMe 1.2TB SSD has outstanding sequential
writing performance, and finishes this test in first spot.
Write a folder of JPG picture files.
For this test I copied a folder of JPG picture files from
the RAMDisk to the Intel 750 NVMe series SSDs, and our other comparison drives.
The folder contained 7861 JPG pictures, with a total capacity of 8410.3MB.
Intel 750 NVMe 1.2TB
SSD
Once again the Intel 750 NVMe SSD is performing exceptionally
well, and is comfortably the fastest SSD in this test.
Write a folder of MP3 audio files.
For this test I copied a folder of MP3 audio files from our RAMDisk
to the Intel 750 NVMe SSD series SSD and our other comparison drives. The
folder contained 1691 MP3 audio files, with a total capacity of 9176.5MB.
Intel 750 NVMe 1.2TB
SSD
Yet again the Intel 750 NVMe 1.2TB SSD is performing exceptionally
well, and is by far the fastest SSD in this test.
Single drive copy tests
These tests are to simulate a single drive in a PC or
laptop. In other words, I will copy a series of files from one folder on the
tested drive to another folder on the same drive. This means the drive is simultaneously
reading and writing during the tests. I also want to make this a realistic test,
so I have used a folder of MP3 music files, and then repeated the test with a
folder of JPG picture files.
Single drive copy tests – 1,691 MP3 song files (9176.5MB total)
Intel 750 NVMe 1.2TB
SSD
In this test the SSD has to read and write data. We already
know that the Intel 750 NVMe 1.2TB SSD has outstanding mixed reading and
writing performance, so it's no surprise to see the Intel 750 NVMe SSD finish
this test at the top of the table.
Single drive copy tests – 7,861 JPEG picture files (8410.3MB total)
Intel 750 NVMe 1.2TB
SSD
It's the same story. The Intel 750 NVMe 1.2TB SSD is the
fastest SSD in this test.
Summary
It is quite clear from these real world copy tests that the
Intel 750 NVMe SSD is an outstanding performer, with the low latency NVMe
protocol helping things along nicely.
Installing applications
Installing applications is possibly something you don't do
that often. But should you replace your system disk, then you will most likely
have to re-install your applications. Most of the SSD drives I have tested up
until now are quite slow at installing applications, most likely because their
I/O performance was quite limited.
For these tests, we picked some popular applications and
copied the entire contents of the CD or DVD media to a RAMDisk. We did this to
make sure that the reading speed of our CD/DVD reader would not hamper the
performance of the target drive.
We then installed these applications onto our comparison drives,
which were all running mirror image installations of our Windows 8 Professional
64-bit installation, and timed the amount of time taken to install the
application with a stopwatch on each of the drives.
MS Office 2007 Professional (full install)
MS Office is one of those applications that make you cringe
at the thought of re-installing it.
Let's find out how our drives coped with the MS Office 2007
full install.
The Intel 750 NVMe SSD gave an excellent turn of speed when
installing this large office suite, and finished the test in first place.
Adobe Fireworks CS3
Adobe Fireworks CS3 is another popular package. Let's find
out how our drives coped with installing this application.
There isn’t a huge margin in the amount of time taken to
install this application on our modern PCIe SSDs. However, the Intel 750 NVMe 1.2TB
SSD finishes this test in first place.
Summary
Our real world tests, though not scientific in nature, I
feel are more realistic than simply running benchmarks. What is clear from these
tests is that the Intel 750 NVMe 1.2TB SSD has outstanding performance in the
real world.
Let’s check out application and game loading performance
on the next page of this article.....
These tests are very simple tests, but very important to
some users of SSD drives.
We simply started an application or game, and measured the
time taken for the application or game to fully load and start.
Application loading times
Adobe Fireworks CS3
There is so little difference in tangible performance between
these SSDs. However, the Intel 750 NVMe SSD loaded this large application in
3.28 seconds, and finished the test in first place.
Corel PaintShop Pro 12
Again, I doubt anyone could tell difference from the fastest
to the slowest SSD, as they are all very close.
Games loading times
FAR CRY 2
Once again the results are all very close, and I highly
doubt anyone could tell the difference between the fastest and slowest SSD in
this test.
F.E.A.R. 2
It's a dead heat in this test, with all the SSDs recording
the same loading time.
Summary
By now it's is becoming very clear that the Intel 750 NVMe SSD
delivers outstanding performance, with its extremely powerful SSD controller,
and low latency NVMe interface, making sure it stays ahead of the competition.
Now let's head to the next page where we will see how the
Intel 750 NVMe SSD performs in PC Mark 8.....
PC Mark 8 - Storage Suite
Here at Myce.wiki, we only recently introduced PCMark Vantage
into our SSD testing. PCMark Vantage is a good test, but is now somewhat
outdated in the applications that it tests, even to the extent of including a
test trace on how Windows Vista booted. We could of course have opted for the
newer PCMark 7, but I personally had issues with the way it ran the HDD tests.
We have built quite a close relationship with FutureMark
software, the authors of the PCMark PC benchmarking software that we use in our
tests. I decided I would use PCMark Vantage as stopgap measure until the more
up-to-date PCMark 8 benchmarking suite became available. I'm pleased to say
that PCMark 8 is now available, and it gives me great pleasure to introduce you
all to the results obtained by this new 'real world' benchmarking suite.
I will describe the basic way that each test is carried out,
above the graph for each test.
Please note; Due to lack of time, I was not able to perform
the PC Mark 8 test on the OCZ REVODrive 350. Therefore, only the Intel 750
NVMe, and the Plextor M6e SSDs were tested.
PC Mark 8 storage suite results
Intel 750 NVMe 1.2TB
Now let’s look at the individual PC Mark 8 HDD suite scores,
in the form of tables and graphs.
PC Mark 8 storage suite: World of Warcraft
The first thing that is very noticeable is that the two
tested SSDs are remarkably close, performance wise, when loading this game.
PC Mark 8 storage suite: Battlefield 3
Once again, the results are very close between the two competing
SSDs.
PC Mark 8 storage suite: Adobe Photoshop light
This time the Intel 750 NVMe SSD manages to pull ahead of
the Plextor M6e.
PC Mark 8 storage suite: Adobe Photoshop heavy
Again, the Intel 750 NVMe SSD is faster than the Plextor
M6e, and because the workload is heavier in this test, the Intel's speed
advantage is more pronounced.
PC Mark 8 storage suite: Adobe InDesign
The Intel 750 NVMe SSD is once again the fastest SSD in this
test.
PC Mark 8 storage suite: Adobe After Effects
There is virtually no difference between the tested SSDs in
this test.
PC Mark 8 storage suite: Adobe Illustrator
Once again, there is hardly any difference between the
tested SSDs.
PC Mark 8 storage suite: Microsoft Word
With only 0.2 seconds between the fastest and the slowest
SSD in this test, I would doubt anyone could tell the difference.
PC Mark 8 storage suite: Microsoft Excel
There is only 0.1 seconds between the slowest and the
fastest SSD in this test.
PC Mark 8 storage suite: Microsoft PowerPoint
Once again, the results obtained from our test SSDs are
almost identical.
PC Mark 8 storage suite: Storage bandwidth
Storage bandwidth displays the amount of bandwidth available
from the storage device, when it is faced with requests for simultaneous reads
and writes.
According to PC Mark 8, the Intel 750 NVMe has 531.71 MB/s
of storage bandwidth, which is substantially faster than the Plextor M6e.
PC Mark 8 storage suite: Overall Score
PC Mark 8 sums all the individual times taken to run each
storage benchmark, then comes up with an overall score for each of the tested
SSDs.
As we can see from the above graph, there isn't a large
difference between the Intel 750 NVMe SSD, and the Plextor M6e. However, the Intel
750 NVMe SSD proved to be the fastest SSD in the PC Mark 8 storage suite tests.
Summary
You may well ask, if the scores are so close between the
tested SSDs, then what is the point of running PC Mark 8 storage benchmark?
Basically, most of these individual tests are very low
demand as far as storage is concerned. More or less all the traces are
lightweight. But hang on a minute, this is how real applications work, and I
and many other reviewers have been saying for years that when we have
lightweight storage traces, it becomes almost impossible to tell SSDs apart
from a performance perspective. We now have a tool that can demonstrate this to
very good effect.
It's not until we start to push SSDs very hard that the
performance differences between SSDs start to become clearer, and for that we
need much heavier workloads, such as the tests run in the Myce Reality Suite.
PC Mark 8 is still very useful, as I'm quite sure that most of you will use at
least a couple of the applications used in these tests, and now you will be
able to compare one SSD to another.
Now let’s round off the performance tests with the Myce
Reality Suite on the next page.....
Myce Reality Suite revision 4.
NOTE: New in revision 4.
- Support for NVMe
- Support for SATA Express
- Support for PCIe
- Support for M.2.
So what is the Myce Reality storage test?
The Myce Reality Suite of tests is made from real everyday
applications and real data, there are no simulated tests, and everything is in
the real world. The only thing that's synthetic is that everything is automated
to make the tests fair, no matter which drive the tests are run on.
Recorded user sessions, by means of a script, are used to
launch the applications, load data, edit data, and then finally write that data
back to the target drive. The scripts do load the system much more than a human
could with these tests, as the scripts do not make mistakes, or pause to think
about what has to be done next.
Measurement system
The measuring system is part hardware and part software. The
hardware is a two part system comprising of a host PC and an external hardware
analyser which is proprietary, and runs a proprietary version of Linux with
special software.
The host PC is built around an Intel Core i7 2600 (Sandy
Bridge) CPU, and an Asrock Z68 Extreme 4 motherboard, with 8GB of 1600MHz DDR3
RAM. The interface between the host PC and the external measuring system uses a
proprietary PCIe2 x8 card, which is housed in the primary PCIe2 x16 slot on the
host PC. The analyser is calibrated before the start of the tests, and is
guaranteed to be accurate to within 0.03%.
Testing method.
Previously the test platform was Windows 7 Home Premium
64bit. The transition to Windows 8.1 Professional 64bit has now been made, and
at the same time a couple of new tests have been validated and introduced. This
has of course meant that I have had to retest a selected number of SSDs on the
new platform, and the results from SSDs that were old, or no longer available
in the test labs, have been discarded.
Building the tests and test image.
Once all the test data files and the scripts that run the
tests were complete, they were then copied to a single folder. I then fitted an
120GB SSD into the PC and did a clean install of Window 8.1 Professional x64.
The latest hardware drivers were installed and Windows update was run to
install any new updates that were available up to 24/11/2013. At this point the
applications that were to be used in the tests were installed and updated with
the latest patches.
The folder containing the application test data files and
scripts was then copied over to our fresh Windows 8.1 Professional 64bit SSD. A
drive snapshot was then taken of the complete SSD and the drive snapshot image
copied to an HDD for safe keeping.
The image is then simply restored to each of the SSDs on
test. After imaging the drive the partition is then realigned “on the fly” and the
free space is filled and then deleted to force TRIM. A 20 minute settling time
is allowed before the tests are run, then each of the 6 tests is run and the
results gathered. This process is repeated for each of the drives I am testing.
The test scenarios are as follows.
- Graphics content
- Video editing
- Audio import and compression
- Application multitasking
- Windows defender (full system drive scan)
- GRID 2 gaming test.
Let’s begin the tests.
Myce Reality Suite – Graphics content.
Using ACDSee Pro 3, 100 JPG pictures with an average size of
10MB are imported into the ACDSee library, and then 12 of these JPG files are
then selected for a batch process, of resize, compress the quality to 80%, and
finally write the edited pictures back to the drive. The test is approximately
78% read and 22% write, with an average queue depth of 1.98.
Please note that this test has become CPU tied and this will
be resolved in the next revision of the Myce Reality Suite.
Myce Reality Suite – Video editing.
Using Vegas Pro, a 14GB HD MPEG2 video stream is loaded into
the editor, from which 2 segments are then cut and pasted into new segments. There
is a lot of disc caching going on in this test, which is approximately 55% read
and 45% write, with an average queue depth of 1.89.
Once again the test has become CPU tied and will be fixed in
the next revision of these tests.
Myce Reality Suite – Audio import and compression.
Using Sony Sound Forge 10, a batch process is run consisting
of importing 30 24bit (192000 Hz sample rate) .wav files, and 100 16bit (44100
Hz sample rate) .wav files which are converted to MP3 audio files with a bit rate
of 128kbps, and the MP3s are then written back to the drive. The test is
approximately 72% read and 28% write, with an average queue depth of 2.62.
This time the Intel 750 NVMe SSD is the fastest SSD, but the
OCZ REVODrive 350 is hot on its heels.
Myce Reality Suite – Application multitasking.
For this test I used several popular applications, Microsoft
Word 2007, Microsoft Access 2007, Microsoft Excel 2007, Microsoft Outlook 2007,
Adobe reader, Adobe Photoshop CS3, uTorrent, Windows media player, and Internet
Explorer 9.
This session runs for approximately 12 minutes. The test is
started by downloading a Linux distribution via uTorrent, Windows media player
is then opened, and a 1080p video file is opened and played for the duration of
the test. Microsoft Outlook is opened and any new emails are received, read,
then replied too, a document in Adobe reader is opened and scrolled from start
to finish, 3 Microsoft Word documents with graphics content are opened, browsed
and some sections of the documents are copied and pasted into a forth document
and then saved back to the drive. The same applies to Microsoft Access and
Excel. 100 MP3 files are imported into Windows media library. Six JPG images
are loaded into Adobe Photoshop and some minor editing is done and the files
saved back to the drive.
Finally, Internet Explorer 11 is opened with 10 tabs, and
the contents of the 10 tabs refreshed, and browsed while the other applications
are busy in the background.
I would describe the multitasking pattern as moderate to
heavy.
During this test there is approximately 85% reading and 15%
writing, with an average queue depth of 6.73.
With the higher queue depths in this test the Intel 750 NVMe
SSD and the OCZ REVODrive 350 are able to show what they can really do. The
Intel 750 NVMe SSD being the faster of the two.
Windows Defender (full system drive scan)
A full system drive scan is selected on drive C: and then
run. The test is approximately 99% reading and 1% writing, with an average
queue depth of 1.2.
The Intel 750 NVMe is showing outstanding performance in
this test, and is in front by quite some margin.
GRID 2 gaming test
The game is launched and then a pre-saved level is loaded.
The test runs until the loaded level starts. The test is approximately 98% reading
and 2% writing, with an average queue depth of 1.
Once again the Intel 750 NVMe SSD has performed extremely well
in this test, and is yet again the fastest SSD.
Summary
I firmly believe that the Myce Reality Suite gives a very
good overall picture of how a drive can perform in the real world and, in this
case, the Intel 750 NVMe SSD is clearly a very capable performer.
Now let’s head to the next page, and see how well the
drive performs after heavy use....
Myce Sustainable Performance Test
Over the last few months I have been studying countless
analyzer traces of real computing workloads, and also developing a test that
would accurately emulate and measure how performance is sustained over a period
of time. For obvious reasons, it is not possible to test an SSD review sample
over several months before publishing a review. The solution was to condense
this down to a manageable test, that doesn't take too long to run.
I will make it clear right from the outset that this is not
a torture test. Bringing any SSD to its knees is not helpful in the least, as I
for one would not use any SSD that had slowed down to crawl, just to prove a
point. The Myce Sustainable Performance test, I believe is a tough, but
sensible test pattern to use for measuring how an SSD will be behave once it's pushed
hard over a period of time.
The test pattern is "workstation" based, and
closely emulates a typical video or graphics workstation environment. The
results are measured using the same hardware I use for the Myce Reality Suite
tests, however, the test data and measuring system use a different method.
The SSD is first filled to 80% of its stated capacity.
Adding to the data that is already there, the "Sustainable
Performance" test data is added. This test data is approximately 20GB is
size, so once this is added the SSD is pretty full.
The test is then run for a period of 20 minutes. 60
performance measurements are taken for every minute of the test, and an average
performance figure is generated after each minute. At the end of the test I
have 20 performance measurements which are then used to generate the graph
below.
The faster SSDs will obviously sustain more writes then the
slower SSDs. For the fastest SSD in this test, the test pattern generated 173GB
of writes, and 193GB of data was read from the SSD during the test.
When reading the graph, you should not pay too much
attention to which drive is the fastest, but instead look at the sustainable
performance curve of each SSD, as this is what this test is all about.
For the SSD that I am reviewing, I will also add a second
graph which looks at the result in more detail.
So let's look at the results.
Sustainable
Performance test
Detailed results for the
review drive
The Intel 750 NVMe 1.2TB SSD has exceptionally good
sustainable performance when pushed extremely hard. There is a slowdown, but
it's only down by 8 MB/s from peak performance. I highly doubt anyone would be
able to notice this slowdown in the real world.
This concludes our review. To read the final thoughts and
conclusion, click the link below....
Final thoughts and the conclusion
User experience
A modern operating system such as Windows 8 rarely does one
thing at time; it processes hundreds of threads at once. Just take a look at
the processes and services that are running in task manager for an idea of how
much is going on, even with the PC idling at the desktop. When you start
running applications on top of this, the workload increases in line with the number
and type of applications you are running. It’s also fair to say that many of
these processes are already loaded into system RAM, but many are also loaded into
and unloaded from RAM to the system drive as and when they are required.
If we look at the 4 basic requirements for a really fast
SSD, they are as follows.
- Small file threaded performance needs to be high.
- Small random file performance needs to be high.
- Sequential read and write speeds need to be high.
- Fast access times.
The Intel 750 NVMe series SSDs have all of these attributes
in abundance, and feel very snappy in use as system drive.
Stability
I have only had the Intel 750 NVMe 1.2TB SSD for a few weeks,
so it’s not possible to comment on the drive's long term reliability. However,
during the testing period, this SSD has been 100% stable and has caused no
issues whatsoever.
The Intel 750 NVMe SSDs are as “plug n play” as it gets,
providing you are running Windows 8.1 or later, and you have a motherboard
which supports 'boot from NVMe'. If you don't have a motherboard which supports
'boot from NVMe' then you may need to find a workaround to allow you to boot
the operating system from the Intel 750 NVMe SSD. If you can't find a
workaround to boot the drive, then you can still use the Intel 750 NVMe SSD as
perhaps a scratch disk for something like Photoshop, and to be honest, the
Intel 750 NVMe, and the OCZ REVODrive 350 are much better suited to this task
than as a system drive.
If you have an operating system older than Windows 8.1, then
you will require the NVMe drivers which are supplied by Intel. You will also
have to inject these drivers, at say for example, during the installation of
the operating system if it is older than Windows 8.1.
The 750 NVMe 1.2TB PCIe SSD is a little slow at booting
Windows 8.1. It took some seven seconds longer to boot Windows 8.1, than many
of the SATA SSDs that I have recently tested. Once Windows 8.1 has booted to
the desktop, the Intel 750 NVMe SSD is blindingly fast. It just takes a little
bit longer to reach the desktop than some other drives.
Conclusion:
Let us summarise the most important positive and negative
points below:
Positive:
- Silky smooth operation as system drives.
- Outstanding sequential reading and writing performance,
even at very low queue depths. - Outstanding 4K random writing performance, at low and high
queue depths. - Outstanding 4k random reading performance at very low, and
very high queue depths. - TRIM support under Windows 7 and Windows 8.
- Completely silent operation.
- Fast operating system start-up and shutdown times.
- Very fast in 'real world scenarios'.
- Low power consumption considering the amount of grunt the
drive has.
Negative:
- Price. The Intel 750 NVMe SSD is expensive.
- A little bit slow to boot Windows 8.1
To sum up, this is what I
would say:
As operating system drive, the Intel 750 NVMe is pretty hard
to fault. Performance is outstanding and this SSD proved to be very stable
during the testing period. However, make no mistake, the Intel 750 NVMe range
of SSDs is aimed squarely at the professional user, rather than the casual PC
user, who only uses the PC for lightweight tasks.
If you can utilise all the performance that the Intel 750
NVMe has on tap, then you're going to love using this SSD. It is by far the
fastest SSD I have ever tested. The down side is, the Intel 750 NVMe SSD is
expensive, and if you can't utilise all its power, then there are more sensible
alternatives out there, for users who mainly do lightweight computing tasks.
Price and availability
The Intel 750 NVMe 1.2TB SSD is available now, and I found
one at Scan
computers in the UK for £806.54 including VAT.
The parting sentence is:
“If you really must have the fastest consumer grade SSD
currently available, then the Intel 750 NVMe SSD is the one to have. It's an astonishingly
fast performer”.
Rating system
The editor rating is based on the following key factors.
- Performance
- Stability (is the device stable?)
- Price
- Warranty
- Supplied accessories (what is included in the package)
Thanks to:
|
|
EFD Software for |
|
|
Alex |
|
|
|
|
|
FutureMark for |
You may comment on this review below.
