by Nikita Nikolaichev
07/11/2011 | 10:02 AM
Some time ago we had the opportunity to see that the performance of contemporary SSDs was severely limited by the SATA-300 interface. So, if we needed an even faster disk sub-system, then we had either to combine several SSDs into a RAID array, or use SSDs with PCI-E interface (which internal architecture, in fact, made them similar to a RAID array of two or four drives). SSDs with SATA-600 interface looked fairly promising at that time, but unfortunately, their speed was limited by capacity of the only controller available then, Marvell 9123, and the issues in its very raw drivers. Moreover, there were not that many SSDs supporting SATA-600 interface available.
Things changed dramatically when the new Intel P67/H67 chipsets came out. One of the major innovations introduced in these chipsets was the integrated SATA-600 controller. Finally, the computer enthusiasts’ dreams are coming true. But to everyone’s greatest disappointment the new Intel P67/H67 chipset launch was ruined by the error in the B2 revision: the transistor responsible for the interface frequency of the SATA-300 ports could potentially fail much sooner than anticipated. To Intel’s credit we should say that they reacted in a timely manner and absolutely correctly: they stopped selling products with potentially faulty chipset revision and announced a massive recall.
Unfortunately, this problem has affected us, too. While we were planning to upgrade our testbed in the early spring of 2011, we were forced to postpone it until May, when mainboards with B3 chipset revision became widely available. But now that all problems are overcome, we are glad to offer you a detailed review of the new solid state drives supporting SATA-600 interface.
Crucial RealSSD C300 drive belongs to the new SSD generation with all other high-performance solid state drives launching this year. However, while it was announced in the middle of last year, it easily outperformed almost all leaders at that point, so it would be fair to consider this drive a transitional solution of some sort. In fact, this statement can be backed up by the fact that Crucial RealSSD C300 became the first SSD supporting new SATA interface with 6 Gbps bandwidth.
Although Crucial brand is used by a daughter-company of Micron, one of the semiconductor market leaders, RealSSD C300 uses not only Micron components inside. Yes, flash memory chips are only from Micron, but the actual controller is from a third-party maker. I am talking about Marvell, which offered a special chip based on a pair of processors with ARM9 microarchitecture. One of these processors is responsible for the SATA interface, while another one – for the interface connecting the NAND chips.
Another peculiarity of the Crucial RealSSD C300 is a 256 MB cache, which should not only increase the drive’s performance, but also lower the flash-memory chips load thus increasing their life-span.
However, the specifications of the Crucial RealSSD C300 shouldn’t be the object of your ultimate admiration: just like the previous-generation SSDs, this product uses 34 nm NAND chips. Moreover, Crucial already has newer and faster products, which we are going to talk about later in this article.
We had a 256 MB Crucial RealSSD C300 model participating in our today’s test session. Here are its formal specifications for your reference:
We tested this SSD twice: with different versions of SATA interfaces featuring 3 Gbps and 6 Gbps bandwidth.
Our Crucial RealSSD C300 model was tested with CTFDDAA256MAG-1G1 256GB – 0007 firmware.
The new SSD called Crucial m4 is the one that came to replace RealSSD C300. But don’t be misled by this completely new name. Crucial m4 is offered in the OEM market as Micron C400, and in reality it doesn’t differ from the predecessor as greatly as you would expect it to. In other words, Crucial m4 is a refresh of Crucial RealSSD C300 and not a completely new product.
In fact, the only completely new thing in this drive is the NAND memory. Crucial m4 uses new 25 nm Micron chips with upgraded ONFI 2.2 interface. Moreover, the developers managed to avoid replacing the hardware, i.e. the Marvell controller, and only upgraded its firmware. As a result, the 256 MB cache is still there.
Anyway, the finer manufacturing process used for the memory chips and firmware optimizations allowed to significantly improve the SSD performance. We received a 256 MB Crucial m4 model for review. Here are its official specifications:
Here you should keep in mind one peculiarity. Crucial drives offer users all available flash-memory, leaving no reserved sectors for the controller. As a result, the drives have larger storage capacity, but they may slow down when the entire drive is utilized.
Just like other SATA 6 Gbps drives, we tested this SSD twice with different versions of SATA interfaces.
The tested Crucial m4 model featured MTFDDAC256MAM-1K1 256GB – 0001 firmware.
We were impatiently looking forward to the launch of the new generation Intel SSDs supporting 6 Gbps interface. It is, actually, not surprising at all, because all previous SSDs from this manufacturer used their own controllers of proprietary design, which were considered among the best out there before SandForce controllers arrived. However, imagine how surprised we were when it turned out that the new SSD series for computer enthusiasts aka Intel 510 was using third-party controllers! And this third-party wasn’t SandForce, but Marvell.
Moreover, the controller in Intel 510 is exactly the same as the controller in Crucial m4. However, firmware matters a lot, too, and Intel’s firmware is unique. Unlike other makers who use Marvell controllers, Intel optimized their firmware for work with large files, and it will seriously affect the results, as we will see later on.
Another interesting peculiarity of the Intel product is the use of old MLC flash-memory manufactured using 34 nm process. Theoretically, this memory is faster and more reliable than the 25 nm version, that is why Intel 510 is completely different from Crucial m4 in specifications.
Intel ships two modifications of their 510-series SSDs: with 120 GB and 250 GB storage capacities. We managed to test both of them, their formal specs are listed below. Note that Intel 510 has only half the cache-memory of Crucial SSDs:
Judging by the numbers above, Intel didn’t really strive to regain the leadership in the SSD market at this time. It looks like the company was simply going to offer the enthusiasts a good 6 Gbps option with acceptable price-to-performance ratio. In order to check out this assumption we tested Intel 510 with 6-gigabit as well as 3-gigabit interface.
The tested Intel 510 SSDs had the following firmware versions at the time of tests:
A year ago OCZ launched an SSD that won the hearts of many computer enthusiasts – OCZ Vertex 2. The secret behind this product’s success was the use of a high-performance controller from a not yet very well-known SandForce Company. They introduced a technology that allowed enhancing the MLC-based products to match the performance level of SSDs using expensive SLC-memory. This technology proved great, but it has been more than a year since it became popular, so the drives equipped with old revisions of the SandForce controller started getting too out-dated.
Therefore, in the beginning of this year SandForce released new processor versions, and SSD makers hurried to update their products with them. Being SandForce’s closest long-term partner, OCZ Technology stood at the head of this process. This is how the new vertex 3 came into this world and today we are going to take a real close look at it.
Vertex 3 SSDs are built on the new SandForce SF-2200 controller. Its major innovations are the long-anticipated support of 6 Gbps interface and another performance bump resulting from firmware optimizations and compatibility with new-generation NAND-memory. According to SandForce SF-2200 specifications, this processor can deliver sequential read and write speeds of up to 500 MB/s and up to 60K IOPS random access speed.
Surprisingly, but OCZ managed to bump up these parameters even more due to their own firmware optimizations. For example, the fastest Vertex 3 modification with 240 GB storage capacity promises to deliver up to 550 MB/s sequential read speed. In other words, OCZ has every intention to become the new SSD market leader, who is offering the fastest solutions at very appealing prices.
OCZ Vertex 3 lineup includes four SSD models with storage capacities ranging from 60 to 480 GB. Frankly speaking, the junior model doesn’t look too appealing in today’s reality, that is why we decided to focus on models with 120, 240 and 480 GB capacities only. We tested each of the SSDs twice: connected to SATA-600 ports and to the “old” SATA-300 ports.
Here are the primary official specifications of the tested models for your reference:
Unfortunately, despite very impressive characteristics, OCZ Vertex 3 SSDs were not free from the problems that came with second-generation SandForce controllers. I am sure that many of you remember the recall of similar Corsair products, and OCZ also didn’t escape the ill same fate. Some Vertex 3 users complained about different issues including complete drive failure. However, the problems seem to be resolved by now. OCZ released new firmware versions, which lower the chances of failure at the expense of minimal performance drop.
We tested the “fixed” vertex 3 modification. Here are the firmware versions of all three models that participated in our tests:
Our comparative testing of the new high-speed SSDs would be incomplete without one of the previous-generation drives. It is true, many of you wonder how greatly the SSD speed has increased over the past year. Therefore, on our results diagrams you will also see the performance numbers for OCZ Vertex 2 - the legendary solid state drive based on the first-generation SandForce controller.
We have already tested a number of different modifications of OCZ Vertex 2 SSDs. Today we will talk about an E-model with 240 GB storage capacity. This model differs from the original Vertex 2 by slightly different distribution of the available flash-memory in favor of increasing the effective storage capacity and performance of the drive.
Here are the official specifications of the OCZ Vertex 2 SSD that participated in our today’s test session:
Its firmware version was OCZ Vertex 2 OCZSSD2-2VTXE240G – 1.32.
In fact the major difference from the hard disk drive testing methodology that we have been using all last year is the testbed upgrade. Our good old testbed on a Pentium 4 processor and ICH7 chipset looked completely out of place in 2011, to put it mildly. The “decrepitude” of our testbed stood out particularly badly in synthetic sequential read/write benchmarks, where CPU utilization hit 100% and the drives couldn’t reach the declared IOps. Even in WinRAR tests we saw that the performance of different drives stumbled upon the CPU speed.
Yes, the testbed was definitely due for an upgrade. But with what? We didn’t want to use an external SATA-600 controller, because this solution wouldn’t ever become mainstream. Therefore, the results obtained on a platform like that would be interesting only to a small user group.
A mainstream chipset from a leading maker could be exactly what we were looking for. AMD fans will have to forgive me here, but I am talking about a different company. It was the P67/H67 chipset from Intel that seemed like an ideal platform for the new-generation testbeds. What did we need from a mainboard? Ideally all we needed were the two PCI-E x16 slots (although we could even do with one if there were integrated graphics) and SATA-600 and USB 3.0 controller integrated into the chipset.
The free PCI-E slot would be used for external controller tests and, of course, for testing SSD drives with PCI-E interface.
As for the CPU, we didn’t need a super-powerful multi-core one: it would heat up a lot and would require an expensive cooler, if we cared for acceptable acoustics. We decided to get closer to the ordinary users and get an i3 model but with the maximum clock frequency. And no Turbo Boost!
As for the RAM, let’s use as much as necessary for the Intel NASPT and Futuremark PCMark vantage benchmarks – 1 GB. If we use more memory, then our Windows 7 operating system will get overexcited and will allocate all free memory for files caching, which will seriously affect the results of those tests that work with files (NASPT, FC-Test).
So, as a result we put together the following test platform:
The software part of our testing methodology has also been updated accordingly. We continue to use IOMeter, FC-Test and PCMark Vantage, but we also added Intel NASPT and later on will add Futuremark PCMark07 (unfortunately, we didn’t include it today).
In Intel IOMeter patters we reduced the increment for the Random Read/Write tests in order to better analyze the drives’ performance with data chunks of smaller size.
Defragmentation tests are now performed using Windows 7 built-in defrag-utility. Of course, we know that defragmentation is useless sand even harmful for SSD drives, because it wastes the drive’s writes. But we are extremely curious to see how the new SSDs cope with this type of load.
Now let’s say a few words about the succession of tests. Our primary goal is not to confirm the official maximum performance numbers declared by the manufacturer, but to obtain the performance results closest to what the ordinary user will see on his system, we run all tests as a single block without straightening out the drive between the tests. No linearization, zero fills or Secure Erase. The only thing allowed was a 10-minute break between the tests.
When we took the drive out of the box, we created a 64 GB partition, run PCMark vantage and NASPT tests. Then we delete the partition and create two 64 GB partitions for the FC-Test. Then these partitions are deleted again and we save a previously created image of a 32 GB partition with installed OS, applications and a set of test files onto the drive. After system restart we launch a test script that defragments the partition on the tested drive, restarts the system, compresses the test files using WinRAR tool, restarts the system and extracts the compressed files into a new folder. Next the partition is deleted from the drive and we start a set of IOMeter tests. The test session is rounded up by power consumption measurements.
So, let’s discuss the obtained results.
As usual we are going to start with sequential read and write tests. In this test IOMeter is sending a stream of read and write requests with a request queue depth of 4. The size of the requested data block is increased each minute, so that we can see the dependence of the SSD linear read/write speed on the size of the data block. This test is indicative of the maximum speed the tested SSD can achieve.
All obtained results have been split between the two diagrams for a more illustrative picture:
As we have already mentioned above, 250 GB Intel 510 copes really well with streaming loads. Its linear read speed exceeds 500 MB/s on large data blocks! Smaller Intel SSD is a little behind its elder brother because of fewer flash-memory chips onboard.
Crucial m4 SSD with the same controller as Intel 510 yields to the latter in linear reads. Moreover, it even yields to its predecessor, Crucial C300, on smaller data blocks.
SATA-300 tests didn’t reveal anything super-exciting: all SSDs were held back by the insufficient maximum bandwidth of the interface.
Among OCZ SSDs, the leadership goes to vertex 3 on SATA-600, just as we have expected :)
And quite naturally, it was the drive with the largest storage capacity (and the biggest number of chips). Although it yields to its younger brothers on small data blocks. All three Vertex 3 drives surpassed the ultimate 500 MB/s threshold on large data blocks, but didn’t hit the declared 500 MB/s mark.
Intel 510 with 250 GB capacity shines during sequential write tests. Its younger brother falls farther behind this time than with reads. Crucial m4 works best of all on small data blocks, but yields to Intel 510 250 GB on larger blocks.
It is interesting that Vertex 3 SSDs failed to compete successfully against Intel 510 during writes. The best one of them has barely hit 300 MB/s. It is also interesting to check out the performance of models with different storage capacity. The 480 GB model takes longer to speed up and then simply jumps to its maximum speed. The 240 GB model, on the contrary, starts faster, and then little by little reaches the maximum speed.
The 120 GB model at first behaves just like the 240 GB one, but by 4 KB data chunk size it has already reached its maximum speed. It must be the few memory chips that affect this result again.
Also, I would like to draw your attention to the overwhelming superiority of Vertex 3 SSDs over the previous-generation Vertex 2 (modification E).
Now we will see how the performance of the SSDs in random read and write modes depends on the size of the requested data block. The requests queue depth in this test equals 1.
Wow! What a great difference between the “brothers-in-chip” – Crucial m4 and Intel 510! Looks like Intel bet the streaming tests, while Crucial – the IOps per second. The performance difference on small data chunks in phenomenally high. I wonder what we are going to see in file tests, which approach will prove the best?
It is interesting that when we switched Crucial m4 to SATA-300 interface the drive slowed down by about 1.5 times on small data blocks.
In the vertex 3 clan the leadership is in the hands of the 240 GB model, and the largest model suddenly yielded to the smallest one. Why did it happen? Does the 480 GB model use different memory?
Let’s see what happens with writes:
And here we see yet another surprise from Crucial. These SSDs are slow on small data blocks and then they suddenly speed up on 4 KB blocks.
I have to point out that it is not a problem of our patterns, because all requests in them are adjusted for 4 KB. So, it is most likely the result of some software “optimizations”.
Intel SSDs act much more reasonable: no sudden jumps or drops in speed…
Well, things are exactly the same here: Vertex 3 SSDs do not like small data blocks during random writes. And judging by the vertex 2 results, we can conclude that this dislike is inherited.
Now let’s see how our testing participants, who demonstrated such diverse results in synthetic tests, will perform in file benchmarks.
As you can tell from the name, this program was created by Intel professionals to test external storage devices with network interface, i.e. NAS. But it turned out that the same utility could also be used to test other storage devices, such as flash or solid state drives. The only requirement is that the tested drive has a logical drive letter assigned to it and is large enough to accommodate the test files set.
The idea behind this test is very simple: there is a set of files created on the tested drive, which is used for different scenarios. These scenarios cover all disk operations performed on the tested drive (file, offset, data size and read/write operation).
The standard testing suite includes the drive working as a streaming data source (HD Playback tests), files read/write speed tests (like FC-test) and tests like Photo Album and Content Creation, which names speak for themselves.
So, let’s start with the tests dealing with large files processing:
As we can see, the testing participants split in two clear groups according to the interface they were tested with. With SATA-300 Crucial C300 suddenly took the lead closely followed by two Intel drives.
It is interesting but the same Crucial C300 with SATA 600 interface turned out the slowest. Although it has an excuse: after all, it was the first drive with SATA-600 interface and it was released much earlier than any of other today’s testing participants.
Intel products took over the lead in SATA-600 tests. Vertex 3 SSDs, however, yielded a little to Intel 510, although they all stayed very close together.
Crucial m4 worked a little faster than its predecessor.
Now let’s see how the drives cope with file reading in two threads:
Once again we clearly see the separation: SATA-600 interface still gives the drives more chances to shine.
Under two-thread load Vertex 3 SSDs not only challenged Intel products, but even outperformed them. Only the 120 GB Vertex 3 model let the larger Intel 510 get ahead. Other than that we have every right to state that Vertex 3 SSDs are faster in this test.
Things are a little more complicated with Crucial SSDs: two-thread load not prevented the m4 model from getting ahead of the C300, but made them extremely close.
Now let’s increase the number of threads to 4:
With four threads Intel 510 250 GB SSD stepped up a line, but still failed to take over the leadership. Vertex 3 SSD with maximum capacity remains the king of the hill. Crucial m4 finally broke free from C300.
Now let’s see how the drives will cope with writing a large file. I suspect that Intel 510 will be ahead again…
And I was right: Intel 510 250 GB does cope with sequential writing better than everyone else. I would also like to say that the interface has the lowest effect on the results in this test compared with the reads. Despite all software optimizations flash memory still limits the write speed. It is most visible on vertex 2.
Now let’s check out simultaneous reading and writing of large files:
Well, Crucial m4 is finally among the leaders, with Intel 510 still being the No.1.
Content creation tests are the most confusing, but let’s assume that they do have something to do with our everyday work. Although in reality no one is capable of creating content as fast as the SSD is capable of saving it :)
Remarkably, Vertex 3 SSDs are the ultimate winners here. Their performance advantage percentage is outstanding. Great job, Vertex 3! Well done, OCZ!
Another test with intuitively understandable name – Office Productivity:
How much depends on the drive in this test!
But anyway, if the office employees had Vertex 3 SSDs in their systems (this is the winner of this test), then by the end of the day they would undoubtedly get considerably more work done.
Now we move on to not such an understandable test - we are going to measure how fast a set of files can be written onto a drive:
Intel 510 is again the winner, with Vertex 3 SSDs following closely behind. Crucial m4 also did pretty well.
Now let’s check out the opposite operation – reading a set of files off the drive:
The picture changes: junior Intel 510 model outperformed the senior one. And the two of them outperformed everyone else. The second surprise is the old Crucial c300 being faster than the new m4 one.
I wonder if the situation changes when we work with a folder?
And it did! Intel SSDs suddenly gave in to Vertex 3 as well as Crucial C300. Vertex 3 SSDs are the leaders here.
Now let’s put the test folder back on the drive:
No, Intel doesn’t give in that easily. Only one Vertex 3 SSD managed to withstand Intel during file reading. Crucial M4 is convincingly faster than C300 this time.
Finally, the last test – photographs processing speed:
The leading group includes Crucial and Intel, and Vertex 3 SSDs take a step back.
Summing up all the obtained results, we would like to single out Intel SSDs that proved exceptional when working with large files and Vertex 3 SSDs that worked brilliantly in threaded tasks and aren’t afraid of complex loads.
Time has come for our favorite FC-test. Unlike NASPT, it is a much simpler one. It works with preliminarily created file sets, with clearly defines number and size of files. According to the testing scenario, a file-set is created on the drive (we measure the write time), then it is read from the drive (we measure the read time), then it is copied within the same drive but into a different folder. The file sets (Install, ISO, MP3, Programs and Windows) differ by the number and size of files, which allows testing the drives in different conditions.
Today we decided to use only three file sets: Install, ISO and Programs.
Let’s start with writing the files onto the drives:
Install set contains few files of medium size – these are typical application distributives, which we now carry on our flash-drives. As we can see, Vertex 3 and Intel 510 SSDs work with these files of this size best of all. Crucial m4 outperforms its younger brother, but, unfortunately, yields to the newcomers.
ISO file set, as you can tell from its name, contains ISO images of a CD disk. The total pattern size is 1.5 TB, so the victory of Intel 510 250 GB SSD wasn’t a surprise. Vertex 3 drives are right behind the leader, while Crucial m4 again fell behind.
Intel 510 shines when working with smaller files. Even the junior model outperformed all the others. Crucial m4 did pretty well here, while Vertex 3 drives suddenly fell short.
Let’s check out the reading:
Vertex 3 SSDs are the winner in Install pattern. Both Intel 510 SSDs fell a little behind, with two Crucial drives closing the ranks.
Vertex 3 SSDs maintained the first two positions in patterns with large files, but Intel products managed to get very close to the leaders here. As we can see, the clear breakdown of the results shows how beneficial it is to use the new SATA-600 interface.
Unlike writing, Vertex 3 is again at the top here, and with a very convincing advantage. Crucial m4 did pretty well, too and challenged Intel 510.
Now we need to check the copy speed within the drive.
Vertex 3 SSDs win in the Install pattern. Although Intel 510 250 GB also did very well, and so did Crucial m4.
The results during large files copy test are very similar to what we have just seen in NASPT: Intel 510 250 GB is the leader, and Vertex 3 follow closely after.
Now let’s see if Intel 510 manages to keep its leadership on the small files:
Nope, it doesn’t. Although the results of the first three leaders in this tests are so close that we could probably call them equal in speed.
Let’s sum up the results of our FC-Test: Vertex 3 is the number 1 winner. Even though these drives yielded a little to Intel products during file writes, they rehabilitated themselves completely during read and copy tests.
Now let’s check out the performance of our SSDs using the latest version of PCMark – the Vantage suite. Compared with the previous versions, the Vantage version of PCMark is more up-to-date and advanced in its selection of subtests as well as Windows Vista orientation. Each subtest runs ten times and the results of the ten runs are averaged for further analysis.
Here is a brief description of each subtest:
Basing on these subtests, the drive’s overall performance rating is calculated.
Empire strikes back (c)! One thing I didn’t expect was the indisputable leadership of the Crucial m4. Well done! OCZ Vertex 3 SSDs are coming closely behind the leader, with the junior model being the fastest of the three this time. Also, I would like to point out a total fiasco of the Intel 510. Even with a SATA-600 controller these SSDs were completely destroyed by OCZ Vertex 2, not to mention the Vertex 3…
What a surprise! Crucial m4 wins the second sub-test in a row. Frankly speaking, taking the win from Vertex 3 in PCMark is an achievement. Intel SSDs are again a disappointment: they are slower than Vertex 2.
Now it looks like we’ve been praising Crucial m4 way too much: here it yielded one step to the largest Vertex 3 SSD. Intel drives improved their performance a little, although they are still inferior to all latest-generation products in the race.
And again Crucial m4 is ahead of everyone else, although the competitors are breathing down its neck. Intel, however, is again in the very end.
Vertex 3 SSDs are becoming a bigger part of the leading group, even though Crucial m4 still looks great.
Here Vertex 3 products take over the leadership pushing Crucial m4 to the fourth place.
The situation in this sub-test is very similar to what we have just seen in the Movie Maker scenario: the leaders have changed, while Crucial m4 still holds the third place. Intel SSDs remain at the bottom of the ranks.
The test for application loading time appoints Vertex 3 the winner. Crucial m4 remains in the fourth place, while its younger brother, C300, moved to the very bottom of the diagram. The only consolation for it is being next to Intel SSDs.
Final rendering test is won by Vertex 3, although its advantage over Crucial m4 is minimal. These two are in fact equally fast in this test, which is an obviously victory for Crucial.
As we have already mentioned in the description of our testing methodology, SSDs do not need defragmentation and may even suffer from it. But why would we give up a good benchmark? You will see in just a minute that this test is indeed great.
As we see, the results are extremely interesting. The defragmentation time of the leader and the outsider differs by more than 1.5 times. The leader is Intel 510 250 GB (remember that this SSD won in the NASPT: HD Playback and Record test) and the outsider – OCZ Vertex 2.
Vertex 3 drives also did very well in this test. In fact, they alternate with Intel SSDs at the top of the diagram. Crucial m4 also did well.
The next test measures archiving speed – a large number of small files are being read from the drive and compressed into one larger file:
Since archiving is very CPU-dependent (we use default archiving settings), the results of different SSDs do not differ as much as in the previous test. But this is exactly what happens in real life. There are drive-dependent tasks and there are the opposite. For example, in this test the performance difference between the fastest and the slowest drive is 6.5%, which is not too much at all. And the winner is Vertex 3!
And what is going to happen when we will be extracting files from the archive? In this case the SSD utilization will be much higher, because a lot of smaller files will be written on it:
Vertex 3 SSDs are winning and in this case the interface influence on the results is minimal. The performance difference between the fastest and slowest disk in this test is 24%, although in absolute terms 5 seconds is pennies.
In conclusion we are going to discuss the power consumption of our today’s testing participants. As usual, we measured the power consumption in five different modes: random reads and writes, sequential reads and writes, and idle mode.
Let’s start with random reading:
As we see, the most energy-efficient in this case are Intel SSDs, while Vertex 3, on the contrary, consume 1 watt more than their competitors. And as we recall, there was nothing outrageous about Vertex 3 speed in this mode.
Now let’s check out the SSDs power consumption during random writing:
The picture changed: Intel 510 250 GB, which was so energy-efficient during reading, has the power appetite of two. :) Vertex 3 SSDs aren’t modest either, while Crucial ones seem to be dieting…
Now let’s move on to sequential operations.
During sequential reading Intel SSDs are the best, while Vertex 3 again need 1 watt more.
History repeats itself during writes, although the energy-efficient crown goes to Crucial m4.
The last test mode left is idle mode. In this mode no external requests are being sent to the drive and it is either in sleep or idle garbage collection mode.
Intel drives consume considerably less than their competitors here, and the most power-hungry ones are again OCZ Vertex 3.
I would like to urge all our readers not to take the results of the Power Consumption tests too close to heart, because in reality it doesn’t matter if the drive consumes 1 watt or power or 2 watts. When we consider the power consumption of an entire system 1 watt will be within the error margin anyway. So it is important to remember only one thing: the today’s fastest 2.5-inch SSDs consume no more than 2 W of power.
Our test session ahs come to an end, so let’s sum up the obtained results and hand out awards.
But before we do that we have to say a few words in general. Yes, contemporary SSDs need SATA-600 interface really badly. Of course, they will still be faster than the previous generation drives even on SATA-300, but if we let them shine by connecting them to high-speed SATA-600 ports, their performance will get a significant boost.
In fact, the new SSDs in 2.5-inch form-factor working with SATA-600 interface have caught up with first-generation PCI-E drives. This is a true breakthrough in a new dimension, nothing less. The only thing that upsets me is that it is the second generation of SATA-600 drives that has got dangerously close to the interface bandwidth limitation. So how will the drives continue to improve?
Well, first of all, there is a tremendous reserve in increasing the write speed. Secondly, the tests showed that firmware optimizations can sometimes work wonders, but there is rarely a drive that shows brilliant results in all tests.
And if we get to the specific solid state drives, we can state with all certainty that our personal favorites are the OCZ Technology Vertex 3 SSDs. They proved phenomenally fast in all synthetic benchmarks, performed very well in NASPT and FC-Test file tests. They showed outstanding results in PCMark Vantage and data compression tests. We can definitely call these SSDs the drives with no bottlenecks. Therefore, it is our pleasure to award OCZ Vertex 3 SSDs with our Editor’s Choice title.
As for Intel 510, they were great in synthetic tests, coped very well with file tests, but their failure in PCMark Vantage spoilt the picture. Therefore Intel 510 SSDs is only the second best.
The third major participant of our today’s test session as Crucial m4. This SSD left a very diverse impression. On the one hand, we can’t forget its remarkable performance in Random Read test, where it completely destroyed all the competitors. But on the other hand, it didn’t really shine in any other benchmarks. Except for the PCMark Vantage, of course, where it performed unexpectedly brilliantly. We believe that this SSD has a lot of uncovered potential: we tested the drive with the very first firmware version, so maybe Crucial developers will eventually optimize it even better.