by Aleksey Meyev
03/24/2010 | 10:29 AM
Benchmarking solid state drives is a highly exciting venture as it is always unpredictable and prone to produce heaps of new information. With hard disk drives, dramatic innovations occur but rarely and you often can foretell how the particular model is going to perform in general whereas each new SSD is a real mystery as yet. They vary in performance so wildly due to the current variation in controllers and firmware that the specifications can give you not a vaguest idea of whether the given model is going to challenge the best in its product category or join the outsiders. Besides, we have not yet actually found an ideal SSD. Some of them are better under random-address loads and others boast record-breaking speeds of sequential operations. There are SSDs that present a kind of compromise between the two mentioned extremities, and there are also SSDs whose main advantage is in their relatively low price.
The price factor is indeed important, especially if it’s not a wealthy corporation that wants to improve the disk subsystems of its servers but an ordinary user who is the buyer. This user wants maximum performance and high storage capacity – and both at an affordable price. But as nothing comes for free in this world, the user finds himself having to compromise anyway.
As a matter of fact, there is some solid foundation for making predictions about SSDs, too. If you know what controller is installed in the given SSD, you can predict its performance to some degree of probability. For example, our previous tests have shown that SSDs with Intel’s controllers are better than others at random-address writing. Samsung’s controller delivers high sequential speeds. The Indilinx tries to be a compromise solution, and the JMicron is downright slow at writing (and no record-breaker at reading, either). But as the manufacturers keep on improving their products, a new version of firmware can change an SSD’s performance dramatically as we’ve seen with Indilinx controllers, for example. Moreover, we have so far tested only one version of each manufacturer’s controller (Intel’s transition to a new SSD generation was accompanied with but minor changes in the controller, so we don’t count this in) but there are going to be more and more different controllers in the future. To make things even more complicated, most manufacturers do not tell us explicitly what controller the particular SSD is based on. You can only find this information on the Web, but you can’t be 100% sure about it.
Now, let’s see what new products we’ve got for our today’s tests.
Like many other companies, this well-known maker of memory modules could not help catching the opportunity of entering the new and promising market. As soon as SSDs had more or less matured, Kingston joined the ranks of SSD makers. At first, it marketed rebranded SSDs from Intel as the M and E series but then rolled out the V (obviously, this stands for “Value”) series. We’ve got the first V series product marked as SNV125 which is currently available in shops, but the company’s website already promotes the second-generation SSD of this series with a model number of SNV425. As far as we know, the first-generation model is based on the JMicron 602B controller whereas the second-generation one, on the JM618. Unfortunately, we hadn’t got a sample of the latter by the time we began our tests.
The SSD’s firmware version is B090522a.
By the way, there is some confusion in this series of Kingston’s SSDs. Besides the mentioned transition to a new controller with but a minor change in the product name, the series also includes a curious 40GB model. As opposed to its series mates, it is based on a modified Intel controller with only half the number of active access channels (5 instead of 10). We really think the company ought to separate such different products in a way that would be clearer for the end user.
And here is one more first-generation V series product from Kingston, but with a storage capacity of 128 gigabytes. It has the same firmware version of B090522a.
We want to add a couple of words about the accessories to SSDs. Most SSDs come without any, but Kingston offers its SSDs in two versions: without anything or with a set of two rails designed to install the SSD into a standard 3.5-inch drive bay. We guess most users are going to appreciate this because very few system cases provide any means of installing 2.5-inch drives and Kingston’s rails can save you some time and trouble.
The V+ series is the newest in Kingston’s line-up but it has already been promoted to a second generation, too. And we’ve got a sample of that second generation, marked as SNVP325, whereas the first generation was marked as SNVP225. As far as we know, the two generations differ with controllers: the original Samsung PB22-J we are quite familiar with was replaced in the second generation with the Toshiba T6UG1XBG controller we don’t know anything about. It’s going to be interesting to test it.
The firmware version is AGYA0201.
OCZ is quite active at creating new series of SSDs. We have already dealt with five products from that brand, and now we’ve got a sixth one. It is called Agility EX and is positioned as a Performance product by the manufacturer. It is special and highly promising because it uses the SLC (single-level cell) type of flash memory (which is generally faster at writing and has a much longer service life than the traditional multi-level cell variety) and an Indilinx controller we can expect high performance from. The only downside is that this SSD’s capacity is only 60 gigabytes. Not much by today’s standards.
The firmware version is 1.31 and OCZ’s previous SSDs with such firmware delivered balanced performance.
Although still selling, this series has been moved into the EOL section on the manufacturer’s website meaning that it’s out of production.
Super Talent makes its long-anticipated debut in our SSD tests. This well-known maker of various flash storage devices offers a wide range of SSDs that were sure to get to us sooner or later. The letters SAM and the 128 megabytes of cache suggest that this model is based on Samsung’s platform.
The firmware version is VBM18C1Q and we saw the same firmware in the Corsair P128 which was based on Samsung’s S3C29RBB01 controller (PB22-J).
Super Talent’s UltraDrive series – and the GX subseries in particular – are the fastest of the company’s SSDs. Here is a 64MB model with MLC memory as is indicated by the letters FTD in the name. The controller is most likely Indilinx’s Barefoot chip.
The SSD’s firmware version is 1571.
This is one more UltraDrive SX series product, but its capacity is 128 GB. Its full model name begins with “FTM” which means it has SLC flash memory inside.
The firmware version is 1819.
Wintec makes various flash memory storage devices including SSDs. Its Filemate model has a large capacity of 256 gigabytes (there are actually no SSDs of other storage capacity in this company’s line-up). The manufacturer’s website says this model is based on SLC memory but that’s not so. There are MLC modules from Micron inside it.
The firmware version is 1819 which reminds us of Indilinx.
These are all the newcomers but we will also add the results of the second-generation Intel X25-M (G2) into this review as a kind of a reference point. Its performance is just exemplary.
The following testing utilities were used:
We installed the OS’s generic drivers for the SSDs. We formatted the drives in FAT32 and NTFS as one partition with the default cluster size. For some tests 32GB partitions were created on the drives and formatted in FAT32 and NTFS with the default cluster size, too. The drives were connected to the mainboard’s SATA port and worked in AHCI mode.
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 changed each minute so that we could see the dependence of the drive’s sequential read/write speed on the size of the processed data block. This test is indicative of the maximum speed the drive can deliver.
The numeric data can be viewed in tables by clicking the links below. We will discuss graphs and diagrams.
The JMicron-based SSDs don’t look good against their opponents: the Kingston V series drives are both only as fast as 100 MBps, which is worse than what you can get from modern HDDs. Most of the SSDs come to the finish at speeds of 185 to 215 MBps, however. Intel’s and Wintec’s drives deliver the highest top speeds.
When processing small data blocks, the Intel X25-M is ahead too, followed by the Super Talent MasterDrive SX.
When it comes to sequential writing, the overall picture looks pretty different although we’ve got the same pair of losers: the controller of the Kingston V series drives cannot deliver more than 100 MBps at the peak and is also slow at processing small data chunks. Intel’s SSD has a rather low top speed but accelerates to it quicker than the other SSDs. The Wintec slows down on large data blocks for some reason, which may indicate some problems with multichannel memory access. The best results come from the Super Talent MasterDrive SX and the Kingston V+ series: the new controller from Toshiba looks highly promising!
In this test IOMeter is sending a stream of requests to read and write 512-byte data blocks with a request queue depth of 1 for 10 minutes. The total amount of requests processed by the drive is much more than its cache buffer, so we get a sustained response time that doesn’t depend on the amount of cache memory the drive has.
Although the SSDs differ greatly in their response time, all of them are far more responsive than HDDs because of the lack of moving parts. An HDD has to move the heads to the necessary track and wait for the necessary sector to come up on the rotating platter, so even the best of today’s HDDs (15,000 rotations per minute!) cannot have a response time smaller than 2 milliseconds. SSDs deliver far better results.
Within this device category, Intel’s SSD has the smallest response while the pair of Kingston’s SSDs have the highest response time.
The numbers change dramatically when we switch to writing. Most of the SSDs – the Intel and all of the Indilinx-based models – deliver very good results, obviously thanks to buffer memory and effective firmware algorithms. The Super Talent MasterDrive SX and the Kingston V+ series are not so good. The former seems to have a large buffer but cannot effectively access flash memory for writing. Toshiba’s new controller in the V+ drive is impressive at writing, especially in comparison with the V series models which have the awfully high write response time as is typical of the JMicron 602B controller. By the way, take note that the 128GB model is almost two times as slow as its 64GB cousin within Kingston’s V series.
Now we will check out the dependence between the drives’ performance in random read and write modes on the size of the processed data block.
We will discuss the results in two ways. For small-size data chunks we will draw graphs showing the dependence of the amount of operations per second on the data chunk size. For large chunks we will compare the SSDs’ performance basing on the data-transfer rate in megabytes per second.
Reading in small data blocks is consistent with the results of the read response time test you have seen above. We can only note one irregularity: the graph of the Kingston V+ series shows that this SSD is good enough with data blocks larger than 8 kilobytes but does not accelerate much with the smaller data blocks. Anyway, its performance is over 5 thousand operations per second, so that’s not a problem really.
The results of the SSDs with large data blocks resemble the standings in the sequential read test. The shape of the Kingston V+ drive’s graph is closer to the graphs of its junior cousins rather than of the opponents. Perhaps Kingston’s SSDs all share something in common.
Writing small chunks of data is the headache of all developers of multichannel controllers for SSDs. The Intel X25-M is beyond competition here. The Indilinx-based SSDs deliver consistent and good results irrespective of whether they have MLC or SLC flash memory inside.
The rest of the SSDs fail in this test, their performance being ridiculously low in comparison with the leaders and their graphs barely rising above the X-axis. You can learn from the results table that the Super Talent MasterDrive SX is faster than the Kingston models among which the V+ series is the only one to be comparable to HDDs in terms of random writing performance. The other two SSDs from Kingston are downright slow. And we can also note that the 128GB model from Kingston is two times as slow as its 64GB cousin with large data blocks but overtakes the latter with 512KB blocks and leaves it behind when processing even larger chunks of data. Perhaps the larger model accesses data in larger blocks, which results in this behavior.
There is some confusion in the ranks as the SSDs process even larger data blocks. The outsiders from the previous diagram are downright poor here, too, but the Kingston V+ series stands out among them as its speed is growing up proportionally to the data block size. The Indilinx-based SLC-memory drives are in the lead here with the exception of the OCZ Agility which tends to slow down when processing very large chunks of data.
In the Database pattern the tested drive is processing a stream of requests to read and write 8KB random-address data blocks. The ratio of read to write requests is changing from 0% to 100% with a step of 10% throughout the test while the request queue depth varies from 1 to 256.
You can click this link to view the tabled results for the IOMeter: Database pattern.
We will build diagrams to illustrate the SSDs’ performance at request queue depths of 1, 16 and 256.
We’ve got interesting results at a queue depth of 1. The Intel is unrivalled at pure reading, followed by the Super Talent MasterDrive SX which is a little faster than the Indilinx-based models. The Super Talent MasterDrive SX then slows down at 20% writes, though. The Intel X25-M holds on to its leading position until 50% and more writes when it gives way to the Wintec. Interestingly, the other two SLC-memory drives are not that fast although have the same controller inside.
Kingston’s SSDs are all on the losing side here. The V+ series model is somewhat better than the others, but still no good for server applications if the server has any writing to do.
When the queue grows longer, the Intel X25-M is first, followed by the Indilinx-based models among which the Wintec stands out. The other SSDs seem to be only any good at reading.
Winding up this part of our tests, we will show you diagrams with each SSD’s performance at five different queue depths. Such diagrams are usually so characteristic that you can easily name the controller installed in the particular SSD.
Kingston’s SSDs all behave in the same manner. Judging by the diagrams, we can suspect the Toshiba controller to trace its origin back to the JMicron. Yes, it is faster than the latter but not fundamentally different. Interestingly, Toshiba’s controller just ignores a request queue if there is any.
The 128GB V series model is somewhat faster than its series mate here.
These graphs tell us a lot indeed as they are highly typical of SSDs with Samsung’s controller inside. When there is a request queue, the SSD slows down at processing a small percentage of writes but accelerates at pure random reading.
These four SSDs all draw one and the same graph. We can see a characteristic effect of the request queue, good processing of writes, and stable performance. They even all have a performance growth in one and the same point. The minor variations in the shape of the graphs are obviously due to the flash memory modules: the MLC model is somewhat faster at reading but worse at writing. The Wintec is a little slower than its opponents under high percentages of writes.
The drives are now going to be tested under loads typical of servers and workstations.
The names of the patterns are self-explanatory. The Workstation pattern is used with the full capacity of the drive as well as with a 32GB partition created on it. The request queue is limited to 32 requests in the Workstation pattern.
The results are presented as performance ratings. For the File-Server and Web-Server patterns the performance rating is the average speed of the drive under every load. For the Workstation pattern we use the following formula:
Rating (Workstation) = Total I/O (queue=1)/1 + Total I/O (queue=2)/2 + Total I/O (queue=4)/4 + Total I/O (queue=8)/8 + Total I/O (queue=16)/16.
It is easy to make a brief comment on this read-only test. Everything depends on the drive’s response time and ability to process a request queue.
The standings do not change much when there are some write requests in the load although the SSDs differ more here. The Samsung-based Super Talent falls behind, being effectively “killed” by the request queue (quite expectedly so, considering its graph from the IOMeter: Database test). Kingston’s SSDs are all very slow. Even the best of them, the one based on the Toshiba controller, is but slightly better than ordinary desktop HDDs, judging by its performance rating.
The standings do not change under the workstation load, either. There are but minor exchanges of places among the Indilinx-based models. The different type of load makes one important difference for the outsiders: the JMicron-based SSDs are just downright slow whereas the Toshiba and Samsung controllers deliver higher performance and are superior to HDDs. The Toshiba looks surprisingly good under this load, the V+ series drive outperforming the same-brand models.
The multithreaded tests simulate a situation when there are one to four clients accessing the disk subsystem all at the same time – the clients’ address zones do not overlap. The number of simultaneous requests from each of them varies from 1 to 8, but we will discuss diagrams for a request queue of 1 as the most illustrative ones. When the queue is 2 or more requests long, the disk subsystem’s performance doesn’t depend much on the number of applications. You can also click the following links for the full results:
When it comes to multithreaded reading from flash-memory drives, you cannot expect exciting results because with such storage devices, there is no difference what address to read data from – all disk addresses are virtually equivalent. You can only expect unusual results from some poorly designed controller. Anyway, it is only the SSD from Intel that can handle a second data thread without a significant performance hit. The other SSDs slow down considerably. Funnily, the JMicron-based outsiders lose less speed – only about 10% as compared to the other drives’ performance hit of 30% or something. It looks like the block-based access together with a multi-channel controller produces an effect at multithreaded reading, and that effect is not positive.
When there are four data threads to be processed, the Indilinx-based SSDs wake up and accelerate. That’s nice on their part, yet we don’t think this can save the day for those drives. No user will deliberately run four disk access threads just because the drive prefers it that way.
The multithreaded writing is not so interesting. The SSDs all slow down somewhat when there are more data threads than one to write. The only exception is the SSD from Intel which does not slow down and takes second place when processing multiple write threads. The SLC-memory drives are surprisingly slow here. SLC memory is supposed to be fast at writing, but we don’t see that in practical applications.
For this test two 32GB partitions are created on the tested SSD and formatted in NTFS and then in FAT32. A file-set is then created, read from the SSD, copied within the same partition and copied into another partition. The time taken to perform these operations is measured and the speed of the SSD is calculated. The Windows and Programs file-sets consist of a large number of small files whereas the other three patterns (ISO, MP3, and Install) include a few large files each. The ISO pattern has the largest files.
We’d like to note that the copying test is indicative of the drive’s behavior under complex load. In fact, the SSD is working with two threads (one for reading and one for writing) when copying files.
This test produces too much data, so we will only discuss the results achieved in NTFS. You can use the links below to view the full results:
The SLC-memory models show their best when creating files: the OCZ Agility EX and Super Talent UltraDrive GX get the top places. The Wintec is a disappointment, though. It is not so fast for some reason. The Kingston V+ series drive should be singled out among the slower models: as opposed to its plus-less cousins, it delivers a normal speed rather than that of a storage device with USB 2.0-interface.
Most of the SSDs are just perfect at reading (and you should already be able to name the losers), especially when reading the small files of the Programs pattern. Their speed is an amazing 150 MBps then! Still, we want to name the best drives of all. These are the Wintec Filemate, Super Talent MasterDrive SX and OCZ Agility EX. The group of leaders is headed by the Intel X25-M which is faster than 200 MBps on large files!
We’ve got three leaders at copying. These are the OCZ Agility EX, Intel X25-M and Super Talent UltraDrive GX with MLC memory. We can note that the Samsung controller in the Super Talent MasterDrive SX seems to like to copy large files. It is trying to catch up with the leaders under that load, but is closer to the losers under the others.
PCMark 2005 has the same tests as the 2004 version (not only in names, but also in results as we have seen a lot of times in our previous reviews), so we only discuss one test from PCMark 2004 which is not available in the 2005 version. It is called File Copying and measures the speed of copying some set of files. The other PCMark 2004 results can be learned from the table below. The PCMark 2005 tests are:
The final result is the average of ten runs of each test. Click the following link to view the complete results table for PCMark 2004.
We’ve got new leaders in the copying test from PCMark 2004 which are different from those of FC-Test: the Intel is now accompanied by the Kingston V+ series drive and the Super Talent MasterDrive SX which behaved so oddly in FC-Test.
The Intel and the Kingston V+ series model are better than the others at loading Windows XP and its applications. We won’t repeat again that the losers are the pair of SSDs based on the JMicron controller. They are always very, very slow.
The General Usage test produces a new pair of leaders: the Wintec and the OCZ Agility. Does it mean that this test shows the advantage of SLC memory? No. The Super Talent UltraDrive GX performs without any luster.
Every SSD copes with this version of multithreaded read load easily. Their speeds are close to their theoretical maximums.
The Toshiba-based Kingston V+ series drive suddenly leaves all its opponents far behind in this test. Its result is impressive indeed. Interestingly, second place goes to the Super Talent MasterDrive SX based on the Samsung chip which beats all of the Indilinx-based models.
The Kingston V+ series SSD also has the highest overall score. This is an expected result as this HDD was far from brilliant in the previous tests.
Once again we want to note that today’s multichannel SSDs leave HDDs far behind in sheer performance. An HDD can only score about 10,000 points in this test at best. Of course, it doesn’t mean that your computer is going to be up to 3 times faster, but you can expect all operations involving large amounts of data to be read from or written to the disk subsystem to take less time if you replace your HDD with an SSD.
To make this part of our test session complete, we are going to run the latest version of PCMark called Vantage. Compared with the previous versions, the benchmark has become more up-to-date and advanced in its selection of subtests as well as overall Windows Vista orientation. Each subtest is run ten times and the results of the ten runs are averaged.
Here is a brief description of each subtest:
Basing on these subtests, the drive’s overall performance rating is calculated.
The multithreaded load looks different here. The speeds are far from the theoretical ones. The Intel is ahead together with the Kingston V+ series model.
This series of tests produces similar results: the Kingston V+ drive is on top, followed by the Intel and the MLC-memory SSDs when there is a lot of reading to be done. If there is more writing to be done, the SLC-memory models are right behind the leader.
The Intel SSD only goes ahead in the Applications Loading test, pushing the swift newcomer to second place.
The resulting scores are impressive. The Kingston V+ series is first and this is an outstanding result for an SSD whose controller lacks caching and is a derivative from the notoriously low-performing JMicron 602B. Once again we have to note that SLC memory does not ensure any significant performance benefits. Modern controllers lift MLC memory up to the same level. It is the speed of the memory chips and how successfully they combine in the specific amount with the given controller that becomes important. So, the controller still remains the decisive factor for the performance of each particular solid state drive.
Next goes our homemade test of defragmentation speed. We created a very defragmented file system on a 32GB partition of a disk by loading it with music, video, games and applications. Then we saved a per-sector copy of the disk and now copy it to the disk we want to test. The tested disk is connected to the mainboard’s SATA controller whose operation mode (AHCI/Standard SATA) is controlled from the mainboard’s BIOS. Next we run a script that evokes the console version of the Perfect Disk 8.0 defragmenter and marks the time of the beginning and end of the defragmentation process. AHCI is turned on. You can refer to this article for details about this test.
Strictly speaking, this test makes no practical sense for solid state drives because there is nothing to defragment on them. Every memory cell is equivalent to any other, so defragmentation won’t have any effect. However, this test will allow us to compare how much time SSDs spend reading and writing the same amount of small data blocks.
This test never stops surprising us. Sometimes its results are just as expected, but at other times, they make no sense. So what do we have here? The MLC-memory models are in the lead. Yes, it is the supposedly slower type of flash memory. The Intel is on top and the next three places go to the MLC-based models including the swift Kingston V+ series. The JMicron-based drives did not make it into the leading group, though. They split up again, by the way: the 128GB model is almost two times as slow as its series mate.
The OCZ Agility is an unpleasant surprise. Yes, the SLC-memory models with the Indilinx controller are not really good in this test, but the Agility is about the slowest of all! We don’t even have the slightest inkling as to the reason for such a poor performance.
Now we are going to show you one more interesting test in which we use WinRAR version 3.8 to compress and then uncompress a 1.13GB folder with 8118 files in 671 subfolders. The files are documents and images in various formats. These operations are done on the tested drive. This test depends heavily on CPU performance, but the storage device affects its speed, too.
There is a group of fast SSDs when creating the archive. It includes both models of the Super Talent UltraDrive GX series, the Intel X25-M and the Kingston V+. The Wintec and the 128GB Kingston V series take the most time to perform this task.
We’ve got a new leader when unpacking the archive. It is the Super Talent UltraDrive GX, with MLC memory chips. The rest of the podium is occupied by the Kingston V+ drive and the Super Talent MasterDrive SX. The pair of JMicron-based drives is still very poor, the 128GB being slower of the two.
You can refer to our article called Hard Disk Drive Power Consumption Measurements: X-bit’s Methodology in Depth for details on this test. We will just list the specific modes we measure the power consumption in:
Let’s check out each mode one by one.
Interestingly, the best drive in this test, the Super Talent MasterDrive SX, needs only one third of the startup current of the worst one, which is the Super Talent UltraDrive GX with SLC memory. Overall, the MLC-memory models need less power but you should also note that the controllers differ in their power requirements, too. Anyway, the SSDs are comparable to modern 2.5-inch hard disk drives in this respect. Most of 2.5-inch HDDs need a startup current of about 1 ampere, too, although newer models try not to exceed this mark.
When idle, the SSDs split up into three groups varying in power consumption. The Super Talent MasterDrive SX on Samsung controller and the new Kingston V+ series on Toshiba controller need less than 0.3 watts here and win this test.
Next goes the group of Indilinx-based SSDs. The MLC-memory model consumes somewhat less power, but the difference is really negligible.
And finally, the Intel X25-M and the two JMicron-based products have the highest power consumption in idle mode. This must be due to voracious controllers they have inside. Well, they can only be called voracious in comparison with other flash memory storage but not with hard disk drives. The least economical SSD in this test needs about the same amount of power as the best of 5400RPM 2.5-inch HDDs!
It’s easy to see the general trend: every SSD’s power consumption at random writing is directly proportional to its consumption at reading. The ratio is roughly the same for every model excepting the Super Talent MasterDrive SX which is economical at reading but needs more power at writing and the Intel X25-M which demands quite a lot of power at reading but is economical at writing. It is only Kingston’s products that are generally more voracious than the others in this test, the new V+ series model requiring the most power of all.
Again we’d like to compare these SSDs with 2.5-inch hard disk drives. The flash-memory storage devices are more economical at reading, being up to four times as economical as the best of power-efficient HDDs. The difference is only twofold at writing, though, and there is even no difference at all if you take the least economical drives from both categories.
It is hard to name the most economical product at sequential operations, too. Perhaps the Super Talent MasterDrive SX and the Super Talent UltraDrive GX with MLC memory are just slightly better than the others. It is easy to see which drives consume the most: these are the two models from Kingston based on the JMicron controller, and the Intel X25-M. We really wonder why the Kingston V series drives need so much power for writing. What do they do with it if their writing performance is so low? Interestingly, the Kingston V+ series need less power than the other SSDs from the same brand but delivers higher performance.
We’ve made a nice discovery in this roundup, namely, the Kingston V+ series with the Toshiba T6UG1XBG controller. The developers of this controller have created a small miracle. The SSD based on it managed to overtake and outperform such renowned opponents as the Intel X25-M and a number of Indilinx-based products under loads typical of home computers. It must be noted that this controller is much worse under server loads, especially if there are any write requests to be processed. Thus, the Kingston V+ series is a very good SSD for home users.
The Super Talent MasterDrive SX with Samsung controller showed all the typical traits of the latter including a satisfactory performance under server loads, average performance in desktop applications, high sequential speeds and low power consumption. We guess this SSD is going to be just fine for users who care about power saving. It will also be a good choice if you want a high speed of sequential reading/writing.
Some time ago there were rather few SSDs with the Indilinx Barefoot controller. But now there are a lot of such products, both with SLC (Super Talent UltraDrive GX FTM and ОСZ Agility EX) and MLC chips (Wintec Filemate and Super Talent UltraDrive GX FTD). It must be noted that this controller, with a large cache buffer it can make good use of, nearly eliminates any difference between SLC and MLC memory in terms of performance (the difference in service life still exists, of course). Yes, the SLC-memory models are somewhat better at processing a large number of random-address write requests but the difference is not as huge as we saw with the first-generation SSD controllers. The Indilinx-based SSDs seem to be all-purpose products capable of handling any type of load.
It is the SSDs based on the JMicron 602B controller that are no good at all. They were represented in this review by Kingston’s first-generation V series models which looked depressingly slow against their opponents.