Personal computers are systematically moving to the use of solid-state drives with NVMe interface. Such SSDs have ceased to be exotic solutions for individual enthusiasts chasing maximum performance, and have taken a place in the mind of the average buyer. Even in the domestic market, which is characterized by serious inertia, NVMe SSDs now occupy about a quarter of the total sales of consumer drives. And if we talk about models with a capacity of terabytes and above, then in this segment the share of NVMe SSDs is even higher — every third buyer prefers such drives.
However, this does not come as much of a surprise. Flash memory drives have generally fallen in price significantly this year, losing more than 50% in price, and a large number of models have begun to appear among NVMe devices that are trying to get closer in price to SATA drives. And sometimes they succeed. In some cases, the overpayment for a more modern interface is less than 10-20% — despite the fact that NVMe solutions are more compact, more convenient when assembling the system, and, most importantly, obviously more productive.
Of course, one of the main initiators of the price convergence of SATA and NVMe SSDs was Kingston, which flooded the market with very affordable NVMe drives of the A1000 series and systematically moved the price of this offer to the price of mid-range SATA SSDs throughout the year. However, it cannot be said that the mass acceptance of NVMe drives happened only through the efforts of Kingston. Inexpensive SSDs with a progressive interface have appeared from other manufacturers of the second or third echelons. Moreover, if the Kingston A1000 is largely a compromise solution that uses a truncated PCI Express 3.0 x2 bus for data transfer, then other developers began to offer drives with a similar price, but with more impressive characteristics. Here, as examples, we can cite the previously reviewed Transcend SSD 110S drive with HMB technology or the hero of today’s review — ADATA XPG SX6000 Pro.
We started testing this novelty from ADATA due to the fact that it is presented by the manufacturer as the best option among inexpensive consumer NVMe solutions. ADATA has a passion for experimenting with SSD platforms, and the XPG SX6000 Pro is a great continuation of that tradition. It uses an extremely rare Realtek-developed bufferless controller, which nevertheless promises performance that is quite high for its class. We have not yet seen Realtek controllers in the case, so we cannot give at least some preliminary assessment of such a solution. This means that you will have to get acquainted with the XPG SX6000 Pro carefully and in detail. That is how this review came about, in which we will try to answer the question of whether the XPG SX6000 Pro can be considered as a solid system SSD for mid-range computers.
So, ADATA XPG SX6000 Pro is a drive based on the Realtek RTS5763DL controller. And this is important, because no other second-tier SSD manufacturers based on this chip existed at the time of writing. Therefore, the drive we are dealing with today is a unique model. At the same time, XPG SX6000 Pro is not the first experience of cooperation between ADATA and Realtek. Previously, ADATA had an XPG SX6000 model in its range that used a Realtek RTS5760 controller, so we can say that the appearance of an updated “professional” drive with a fresher base chip is not a big surprise.
But still, it seems a little strange that when switching to another controller, ADATA limited itself to just adding the word Pro to the name. And in general, the new drive is fundamentally different from its predecessor. Moreover, the XPG SX6000 Pro was originally shown at trade shows under the name SX7100, and that would have been a more fair name for it. Judge for yourself: in the new model, data is exchanged with the system via four rather than two PCI Express 3.0 lanes, and the declared performance indicators have approximately doubled. In addition, compatibility with the NVMe Express 1.3 protocol has appeared in the XPG SX6000 Pro series, the model range has expanded towards increasing capacities, and the drives themselves have become more reliable due to the addition of error correction technologies for LDPC encoding to the firmware. Is all this not enough to increase the numerical index in the product name?
The formal specifications of the XPG SX6000 Pro are as follows:
|Series||XPG SX6000 Pro|
|Form Factor||M.2 2280|
|Interface||PCI Express 3.0 x4 — NVMe 1.3|
|Memory chips: type, interface, manufacturing process, manufacturer||Micron 64-layer 256-Gbps TLC 3D NAND|
|Buffer: type, volume||No|
|Max. sustained sequential read speed, MB/s||2100||2100||2100|
|Max. sustained sequential write speed, MB/s||1200||1500||1500|
|Max. random read speed (blocks of 4 KB), IOPS||190 000||250 000||250 000|
|Max. random write speed (blocks of 4 KB), IOPS||180 000||240 000||240 000|
|Power consumption: idle / read-write, W||N/A|
|MTBF (mean time between failures), million h||2.0|
|Recording resource, TB||150||300||600|
|Overall dimensions: L × H × D, mm||80×22×3.58|
|Warranty period, years||five|
The Realtek RTS5763DL controller used at the heart of the XPG SX6000 Pro is a stripped-down version of the RTS5763 chip, which is a full-fledged high-performance controller with a DRAM buffer and eight channels. However, the older RTS5763 chip has not yet reached the final products, and the four-channel and bufferless RTS5763DL, which, however, is also not bad in its niche, has to take the rap for two. After all, it follows from the specifications that the speeds of the XPG SX6000 Pro should be higher than that of the Kingston A1000 drive based on the Phison E11 controller and than that of the Transcend SSD S110 based on the Silicon Motion SM2263XT chip, and its resource should be at least as good.
I must say that now there is a clear trend of transferring low-cost SSDs to QLC-memory. It also affected NVMe drives — for example, Intel SSD 660p and Crucial P1 are based on four-bit memory. But ADATA XPG SX6000 Pro, despite the close price, uses traditional TLC memory, which should also have a positive effect on both reliability and performance. Moreover, the ADATA drive uses not just any TLC memory, but a successful 64-layer TLC 3D NAND manufactured by Micron, which is formed from small-capacity crystals and allows you to get memory arrays with very decent performance.
Nevertheless, the XPG SX6000 Pro does not do without SLC caching technology, which in this case is fully dynamic. You can illustrate its work with the help of a graph of the speed of continuous sequential data writing to a 512 GB drive (measurements were made on a free SSD).
In the accelerated SLC mode, the drive manages to write about 174 GB of data, while the write speed is approximately 1.5 GB / s, that is, it corresponds to the value declared in the specification. But then the free space in the flash memory array runs out, and the recording goes into slow TLC mode. Moreover, simultaneously with the recording, the drive controller is forced to free up space and compact the data in the cells written in the SLC mode. As a result, performance drops by an order of magnitude — up to about 170 MB / s. But during normal home use, such situations will rarely be encountered — they can only arise when transferring volumes of information to SSD that can take up more than a third of the free space.
As you can see, SLC caching quite well compensates for the shortcomings of TLC memory, and there is no problem here. In practice, the lack of a dedicated DRAM buffer in ADATA XPG SX6000 Pro, which is needed to store a copy of the address translation table with quick access, can cause much more trouble in practice. But the Realtek RTS5763DL controller is ready to provide some compensation — HMB (Host Memory Buffer) technology. The bottom line is that instead of allocated dynamic memory inside the SSD, the controller uses the main RAM of the computer to store the working copy of the address translation table. Direct access to it from peripheral devices is one of the features of the PCI Express bus, so this is not a problem for NVMe drives. Everything works out of the box and does not require any additional configuration.
However, specific implementations of HMB technology may vary. Different drives can request different amounts of memory from the operating system for their needs and manage it differently. In order to evaluate the performance of the HMB in the ADATA XPG SX6000 Pro, we tested the performance of the SSD in random small block reads, which require multiple lookups in the address translation table. The dependence of the random read speed on the amount of data within which requests are made makes it possible to reveal which part of the translation table is cached in the PC’s RAM and under what conditions the read performance of ADATA XPG SX6000 Pro does not suffer from the absence of a dedicated DRAM buffer in the drive design.
When testing a bufferless drive Transcend SSD 110S, we were convinced that HMB technology can quite effectively cache the most demanded part of the address translation table. However, in the ADATA XPG SX6000 Pro considered today, the computer’s RAM is not used so intensively. High speed of arbitrary operations is provided only within a small, 4-gigabyte area, which may not be enough under real load. In other words, the HMB technology in the XPG SX6000 Pro only partially compensates for the lack of a DRAM buffer in this drive. With serious small-block loads that involve accessing large amounts of data, the performance of this SSD can be seriously reduced.
In addition, this approach, which involves placing a part of the address translation table in the PC’s RAM, requires special support from the operating system. At the moment, it can only be provided by Windows 10 version 1709 and later, or Linux, starting with release 4.14. In all other cases (for example, in Windows 7), ADATA XPG SX6000 Pro will work «in compatibility mode» — as an extremely slow bufferless drive.
⇡ # Appearance and internal structure
For testing, we used the average representative of the ADATA XPG SX6000 Pro lineup — with a capacity of 512 GB. On the one hand, this version no longer has artificially limited performance due to insufficient parallelism of the flash memory array, and on the other hand, it costs very little, only about $7.00.
The ADATA XPG SX6000 Pro is a board in the M.2 2280 form factor, made on a black PCB. Conveniently, all the chips on such a board are located on one side: thanks to this, the drive can be installed in the “low-profile” M.2 slots found in some thin laptops.
Another positive feature of the XPG SX6000 Pro is that the manufacturer, despite the low cost of this model, paid some attention to the issue of heat dissipation. In the box with the SSD, you can find an aluminum heat-distributing plate with an adhesive layer, which the user, if desired, can stick on the surface of the chips. True, its small thickness and smooth profile are unlikely to provide high heat dissipation efficiency, but even this option is still better than nothing.
The element base used in the ADATA XPG SX6000 Pro includes five microcircuits. First of all, this is the Realtek RTS5763DL controller itself. Please note that it has a rather small size, which is explained by its internal simplicity: a four-channel architecture and the absence of an integrated DDR4/DDR3 SDRAM controller.
The remaining four chips are flash memory. They on the XPG SX6000 Pro have their own ADATA marking, which means that the manufacturer purchases TLC 3D NAND in the form of semi-finished wafers and independently cuts them into individual crystals, tests and packs them. This approach allows you to reduce the cost of the drive, but you need to keep your eyes open. Despite the fact that Micron is the manufacturer of the flash memory in the XPG SX6000 Pro, ADATA itself determines what quality crystals can be used in this drive. In other words, the drive in question is formally based on a 64-layer second-generation TLC 3D NAND from one of the leading manufacturers, but in fact, such memory may differ in reliability from the similar original TLC 3D NAND used in higher-end SSDs.
The 64-layer Micron TLC 3D NAND dies have a capacity of 256Gb, so the Realtek RTS5763DL controller in the 512GB ADATA XPG SX6000 Pro works with a flash array with 4x interleaving on its 4 channels. It is this factor that makes the drive under consideration a sufficiently productive solution for sequential operations.
The traditional marking sticker is found on the back of the ADATA XPG SX6000 Pro. However, it is not particularly informative: it only reports the name, capacity and serial number of the SSD. There is no information about the firmware version or production date.
With ADATA’s service software, the situation is far from being the best. A proprietary utility for the company’s drives exists, but it develops extremely sluggishly, and its capabilities and interface leave much to be desired.
However, the existing SSD Toolbox utility is compatible with the XPG SX6000 Pro and still provides basic functionality.
Thus, ADATA SSD Toolbox not only provides complete diagnostic information about the device, but also allows you to check the flash memory of the drive, send a TRIM command package to it, or automatically configure the operating system settings (disabling Superfetch, Prefetch and defragmentation).
Also through ADATA SSD Toolbox, you can update the firmware and perform the Secure Erase procedure.
In addition, after registering the purchased XPG SX6000 Pro on the manufacturer’s website, you can get a key to the popular Acronis True Image HD 2013/2015 data cloning program.
⇡ # Testing methodology
Testing is carried out in the Microsoft Windows 10 Enterprise x64 Build 16299 operating system, which correctly recognizes and maintains modern solid state drives. This means that in the process of passing the tests, as in normal everyday use of the SSD, the TRIM command is supported and actively involved. Performance measurement is performed with drives in a «used» state, which is achieved by pre-filling them with data. Before each test, the drives are cleaned and maintained using the TRIM command. Between individual tests, a 15-minute pause is maintained, allotted for the correct development of garbage collection technology. All tests use randomized incompressible data.
The partition within which the speed of operations is tested has a size of 32 GB, and the duration of each test is forty seconds. Such parameters, in particular, will allow you to get more relevant results for those SSDs that use various SLC caching technologies.
Applications and tests used:
- Iometer 1.1.0
- Measuring the speed of sequential reading and writing data in blocks of 128 KB (the most typical block size for sequential operations in desktop tasks). Testing is carried out at different request queue depths, which makes it possible to evaluate both realistic and peak performance parameters.
- Measuring the speed and latency of random reads and writes in 4 KB blocks (this block size is used in the vast majority of real operations). The test is carried out twice — without a request queue and with a request queue with a depth of 4 commands (typical for desktop applications that actively work with a forked file system). The data blocks are aligned with the flash memory pages of the drives.
- Establishing the dependence of random read and write speeds when the drive is working with 4-kilobyte blocks on the depth of the request queue (in the range from one to 32 commands). The data blocks are aligned with the flash memory pages of the drives.
- Establishing the dependence of random read and write speeds when the drive is working with blocks of different sizes. Blocks from 512 bytes to 256 KB are used. The depth of the request queue during the test is 4 commands. The data blocks are aligned with the flash memory pages of the drives.
- Measuring performance under a mixed multi-threaded load and establishing its dependence on the ratio between read and write operations. The test is performed twice: for sequential read and write operations in 128 KB blocks, performed in two independent threads, and for random operations with 4 KB blocks, which are performed in four independent threads. In both cases, the ratio between reads and writes varies in 20 percent increments.
- Investigation of SSD performance degradation when processing a continuous stream of random write operations. Blocks of 4 KB and a queue depth of 32 commands are used. The data blocks are aligned with the flash memory pages of the drives. The duration of the test is two hours, instantaneous speed measurements are taken every second. At the end of the test, the ability of the drive to restore its performance to its original values is additionally checked due to the operation of the garbage collection technology and after the TRIM command has been processed.
- Crystal Disk Mark 6.0.2
- Synthetic benchmark that provides typical SSD performance measured on a 1 GB area of the disk «on top» of the file system. From the entire set of parameters that can be evaluated using this utility, we pay attention to the speed of sequential read and write, as well as the performance of random reads and writes in 4-kilobyte blocks without a request queue and with a queue of 32 instructions deep.
- PCMark 8 Storage Benchmark 2.0
- A test based on emulating real disk load, which is typical for various popular applications. On the tested drive, a single partition is created in the NTFS file system for the entire available volume, and the Secondary Storage 2.0 test is carried out in PCMark 8. As test results, both the final performance and the speed of execution of individual test traces generated by various applications are taken into account.
- Real file load tests
- Measuring the speed of copying directories with files of different types. For copying, a standard Windows tool is used — the Robocopy utility, as a test set, a working directory is used, including office documents, photographs and illustrations, pdf-files and multimedia content with a total volume of 8 GB.
- Measuring the speed of archiving files. The test is carried out with the same working directory as the copying, and the 7-zip archiver version 9.22 beta is chosen as a tool for compressing files. The Deflate method is used to reduce the impact of processor performance.
- Research of archive unfolding speed. The test is carried out with an archive obtained by measuring the archiving speed.
- Evaluation of the speed of launching a game application. Measures the performance of the disk subsystem when executing a script captured when launching Far Cry 4 and loading a custom save level into it. To minimize the impact of processor and memory performance, all delays that occur due to their fault were removed from the test scenario.
- Evaluation of the startup speed of applications that form a typical working user environment. The performance of the disk subsystem is measured when executing a script captured when running an application package that consists of the Google Chrome browser, Microsoft Word text editor, Adobe Photoshop graphics editor, and Adobe Premiere Pro video editor with working files. To minimize the impact of processor and memory performance, all delays that occur due to their fault were removed from the test scenario.
⇡ # Test stand
With the release of Coffee Lake Refresh processors, we decided to once again update the test system, which is used to measure the performance of NVMe SSD models. Still, such drives are primarily bought by enthusiasts moving to new platforms, and therefore it is logical to use the latest platform in test tests.
As a result, a computer with an ASRock Z390 Taichi motherboard, a Core i7-9700K processor with an integrated Intel UHD Graphics 630 graphics core and 8 GB DDR4-2666 SDRAM is used as a test platform. Drives with an M.2 interface during testing are installed in the corresponding slot of the motherboard connected to the chipset. Drives in the form of PCI Express cards are installed in a PCI Express 3.0 x4 slot, which also works through the chipset.
The volume and speed of data transfer in benchmarks are indicated in binary units (1 KB = 1024 bytes).
A separate explanation should be made regarding the closing of the Meltdown and Specter processor vulnerabilities. Existing patches significantly reduce the performance of SSDs, so measurements are taken with deactivated OC patches dedicated to closing these vulnerabilities.
⇡#List of test participants
The ADATA XPG SX6000 Pro is an inexpensive NVMe drive. Its direct competitors are not solutions of the Samsung 970 EVO, Intel 760p and WD Black NVMe class, but much cheaper models. Therefore, along with regular test participants on the charts, you will find performance indicators of Kingston A1000 and Transcend SSD 110S playing in the same price category.
As a result, the list of tested models looks like this:
NVMe driver versions used:
- Intel Client NVMe Driver 220.127.116.117;
- Microsoft Windows NVMe Driver 10.0.16299.371;
- Samsung NVM Express Driver 18.104.22.1682.
⇡#Sequential read and write performance
Linear operations do not put a lot of load on either the SSD controller or the address translation table. And here bufferless drives can show their best qualities. However, this cannot be said from the results: the reduced number of channels in the flash memory array leaves its mark. As a result, the read speed of the ADATA XPG SX6000 Pro is on a par with the Transcend SSD 110S, that is, about one and a half times slower than the Samsung 970 EVO class drives.
But on the other hand, the effective implementation of dynamic SLC caching allows the XPG SX6000 Pro to show good performance during linear recording. The result of this accumulator clearly indicates that dynamic algorithms are certainly better than static ones.
In general, the ADATA XPG SX6000 Pro based on the Realtek controller, as well as the Transcend SSD 110S with the SMI controller, look like good options for budget NVMe drives. At least this is how they can be characterized against the background of the Kingston A1000, which costs about the same, but in terms of linear speeds, it is not far from the SATA SSD.
⇡#Random read performance
As was shown when reviewing the ADATA XPG SX6000 Pro passport characteristics, the controller of this drive tries to get by with caching only a small part of the address translation table. As a result of high speed performance, random reading does not work. In fact, according to this parameter, this is one of the slowest NVMe SSDs, outperforming only the outsider Kingston A1000. However, this may be quite enough to find a suitable niche for the XPG SX6000 Pro. After all, it is at least as good as a SATA SSD.
⇡#Random write performance
Random write speed can’t be called ADATA XPG SX6000 Pro’s forte either. With an increase in the depth of the request queue, its performance during such operations practically does not scale, which is generally typical for bufferless SSDs. However, in fairness, it should be noted that in the absence of a request queue, the XPG SX6000 Pro is quite good. It not only shows the best result among entry-level NVMe drives, but even reaches the performance of “full-fledged” models with a DRAM buffer and an eight-channel flash memory array.
⇡#Performance under mixed load
But for the ADATA XPG SX6000 Pro, mixed operations are not a particular problem. This drive performs better than other inexpensive NVMe SSDs with multidirectional requests, especially when it comes to sequential accesses. However, one should keep in mind that scenarios with a noticeable predominance of random read operations still remain a rather weak point of the XPG SX6000 Pro. All this means that ADATA’s budget offer will not be the best fit for the role of a universal system SSD.
⇡#Performance in CrystalDiskMark
In this section, we decided to compare the ADATA XPG SX6000 Pro with the Kingston A1000 to show how the drive considered in this review is better than the most popular budget NVMe SSD. However, it should be taken into account that the high performance of the ADATA XPG SX6000 Pro is largely due to the fact that the CrystalDiskMark benchmark uses a test file of only 1 GB in size. Since the ADATA XPG SX6000 Pro uses only a few megabytes of memory for the address translation table, it is only able to deliver good speeds when working with relatively small amounts of data. The Kingston A1000 does not have such a drawback, since it has a dedicated full-fledged DRAM buffer in its design.
⇡#Performance in PCMark 8 Storage Benchmark 2.0
From the above diagram, we can conclude that the ADATA XPG SX6000 Pro is a typical inexpensive NVMe SSD. It can’t compete with higher-end drives like the Samsung 970 EVO, but in terms of price-performance ratio it still looks pretty good. You just need to understand that although the XPG SX6000 Pro has a 3.9 GB / s NVMe interface running through four PCI Express 3.0 lanes, this SSD is a compromise model that is only one and a half times faster than traditional SATA SSDs. At least that’s how PCMark 8 ranks this drive.
By the way, if we rely on the result of this benchmark, which evaluates the performance of the disk subsystem in real applications, we can conclude that the ADATA XPG SX6000 Pro is not the most attractive inexpensive NVMe SSD model. Bufferless drives based on the SMI SM2263XT controller can boast a better result. Examples of such models are Transcend SSD 110S, as well as all kinds of KingSpec drives from AliExpress or, for example, the HP EX900, which is widely represented on the American market.
The integral result of PCMark 8 should be supplemented with the performance indicators given out by the drives during the passage of individual test tracks, which simulate various variants of a real load. The fact is that with a diverse load, flash drives can behave in some special way.
⇡ # Performance under real load
File operations inside the drive are not the best load case for the ADATA XPG SX6000 Pro. Although it still does not fall into the number of outsiders, its performance is quite acceptable for an inexpensive NVMe SSD. You also need to keep in mind that the ADATA drive works better with sequential loads, so it will perform better when working with large files than with a large number of small files.
The ADATA XPG SX6000 Pro demonstrates not too impressive performance when measuring the speed of downloading applications and programs from it. However, the results shown by this drive are still 30-60% better than the speeds that can be obtained in similar scenarios from a SATA SSD. Therefore, in terms of performance, an inexpensive NVMe SSD like the XPG SX6000 Pro will be at least as good as any of the drives with a SATA interface. And in order for such an offer to find its buyer, this is quite enough.
⇡ # Features of TRIM processing
For drives without a DRAM buffer, it is not possible to check for performance degradation during long-term random operations. They handle small-block writes extremely poorly, and their performance is so low that it would take tens of hours to completely fill a half-terabyte SSD. So the issue of performance drop after exhaustion of free space in the flash memory array will have to be taken out of the brackets.
However, it’s not a big loss. This test has a synthetic nature, and from the point of view of practical work with an SSD, it means little, because during the usual copying of files or when installing software, the operations are not random, but sequential.
But we still need to control how well garbage collection works offline for the drive even after issuing the TRIM command. To investigate this issue, after filling the SSD with data and then deleting it with TRIM turned off, we wait 15 minutes, during which the SSD can try to autonomously recover due to garbage collection, and measure the performance. Then the TRIM command is forcibly sent to the drive — and the speed is measured again, which makes it possible to verify that the SSD is able to fully restore its passport speed using TRIM.
In this test, the ADATA XPG SX6000 Pro performs quite ordinary. After issuing the TRIM command, it fully restores its original performance, as it should be. Standalone garbage collection without TRIM does not work for this drive, which, however, is quite typical for drives with an NVMe interface.
Another important point related to TRIM concerns how much load on the controller is the processing of this command. The fact is that this is happening not so already and imperceptibly for the user. When the operating system informs the drive that some sectors are being retired by the file system, the SSD controller must consolidate these sectors and clean up the freed flash memory pages for future operations. Such a regrouping requires rewriting and clearing areas of memory, and this not only takes a noticeable amount of time, but also seriously loads the controller with work.
As a result, after deleting large amounts of data from a disk, SSD owners often experience temporary slowdowns or even “freezes” of the drive. In practice, this can cause serious discomfort, because no one expects that an SSD, the main advantage of which is an instant response to external influences, will freeze for a few seconds. Therefore, we added an additional study to the methodology, which allows us to track how imperceptibly for the user this or that SSD serves TRIM commands. The test method is very simple: immediately after deleting a large file — 32 GB in size — we check how the drive copes with random data read operations, controlling both the read speed and the waiting time that elapses from the moment of each data request to the response of the drive.
After deleting a 32 GB file, the drive needs about three seconds to clean up. During this period, he almost completely ceases to respond to external influences. Response time grows by two orders of magnitude, and performance drops to zero even when reading. This may serve as another confirmation of the thesis that such an SSD will not be very good as a working disk. The destiny of such solutions is to be a repository for games and programs, when working with which scenarios such as WORM (Write Once, Read Many) prevail.
⇡ # Checking the temperature regime
The problem of overheating is acute for high-speed NVMe drives in the M.2 form factor. However, the ADATA XPG SX6000 Pro is a relatively simple product with performance that is noticeably lower than what the flagships offer. Does this mean that such an SSD does not need any special cooling?
For a practical test, we monitored the temperature regime of the test drive when the load on it consisted of sequential operations with a request queue depth of 4 commands. The measurements were carried out on an open stand, and the SSD was not blown with additional air.
When reading, there are no problems with heating. In idle, the Realtek RTS5763DL controller has a temperature of about 45 degrees, and in the case of intensive read operations, the temperature rises by only 3-4 degrees.
But when recording, the situation with the temperature regime is seriously aggravated. In about a minute and a half of continuous operations, the controller warms up to 100 degrees, after which throttling is turned on. But do not think that ADATA XPG SX6000 Pro is prone to overheating. In fact, no, because in order to bring this drive into a high-temperature state, it is required to write more than 140 TB of data to it without interruption, which, of course, does not happen in real conditions.
In other words, in normal user scenarios with a predominance of read operations, the temperature regime of the ADATA XPG SX6000 Pro does not cause any concern.
The reviewed ADATA XPG SX6000 Pro is one of the products about which it is difficult to say something unambiguous. This drive is easy to present in both positive and negative light — depending on how you approach it.
On the one hand, the XPG SX6000 Pro is not at all like the “exemplary” NVMe SSDs like the WD Black NVMe or Samsung 970 EVO. It is noticeably slower than them, despite the fact that it uses exactly the same progressive NVM Express 1.3 interface and is included in the system via four PCI Express 3.0 lanes. But on the other hand, the reviewed ADATA offer is sold at a much more affordable price, and it is more likely to be compared with SATA drives, which, of course, are inferior to it in terms of speed.
In other words, ADATA XPG SX6000 Pro is a kind of intermediate option for those who want to get something faster than a SATA SSD, but are not ready to spend money on a normal flagship NVMe drive. And in this role, the XPG SX6000 Pro performs quite convincingly. It can be considered an alternative to solutions based on the SMI SM2263XT controller, the most popular option among which is the Transcend SSD 110S. ADATA’s offering uses a different Realtek-designed controller, but in practice it performs at a similar level of performance, albeit with slightly different accents in the performance profile.
This is due to the fact that the XPG SX6000 Pro is a bufferless drive that, although it supports HMB technology, uses a very small amount of RAM. And this means that intensive small-block reading is the most unfavorable load case for this model. Therefore, when relying on the ADATA XPG SX6000 Pro, you need to take into account that it will not show itself too well if you deploy an operating system and work applications on it. But if you have large files on it, for example, the same games, then the proposed level of performance should not disappoint. And it is precisely as an inexpensive gaming storage medium that the XPG SX6000 Pro has the right to life.