Corsair Neutron XT in the test

If you want to see SSD controllers apart from the Marvell and Samsung mainstream, Corsair is always a good candidate. Some time ago we encountered the LAMD LM87800 controller in Corsair's Neutron GTX, which performed very well. With the Neutron XT, the GTX got a successor based on the Phison S10 controller.

Intro

The former pure storage specialist has now clearly shifted its focus. The 08/15 as well as the enthusiast gamers have become the target group of Corsair today, which one would like to serve in addition to memory modules with housings, mice or keyboards, power supplies and of course also with SSDs for every price range. The fact remains, however, that Corsair always wants to see its name in connection with quality and, as a result, the products can be a bit more expensive.

Since Corsair buys the controller and NAND flash for SSDs on the market, there is of course a greater flexibility in the technology used. A connection to a specific controller manufacturer can really not be found with Corsair: LAMD controller in the Neutron GTX, Silicon Motion in the Force LX and now Phison S10 in the Neutron XT. The manufacturer tries to find the right model for each series and the corresponding segment in order to create the best possible construct.

The Phison insert is not that new for Corsair, because the smaller brother Phison S8 was already used in the Force LS. Phison you can practically see yourself as an old hand on the market, because the Taiwanese company has been developing and producing controllers for flash-based storage media for 15 years.

With the Neutron XT, Corsair is placing a fresh model in the performance segment for home users and is now relying on the combination with Phison S10 and Toshiba NAND. We test what the new performance model can do.

SSD bookmarks:

• Structure
• Empty or free?
• Wear leveling
• The root of all evil: Read-Modify-Writes garbage disposal
• Spare area and overprovisioning
• TRIM does not delete!
• Speed
• SLC, MLC, eMLC, TLC
• SSD caching - how-to

Recent SSD Reviews:

• AMD / OCZ Radeon R7 SSD with 240 GB
• OCZ ARC 100 with 256 GB
• Crucial MX100 with 256 GB
• Corsair Force LX with 256 GB
• Crucial M550 with 256 GB and 1 TB
• SanDisk Extreme II and Ultra Plus
• Samsung 840 Pro and EVO
• SSD reboot

The test candidate

Key data and technology

As already mentioned, the heart of the XT is the Phison S10 controller. It is a quad-core CPU, whereby the tasks should be strictly distributed: One core takes care of the requests from the host system, and three cores take care of computationally intensive internal tasks such as garbage collection and analysis Execution of wear leveling. The following table briefly compares the properties of the XT with those of the previous high-end GTX model.

The NAND flash is connected to the controller with up to eight channels and consists of Toshiba MLC-NAND of the A19nm class. The 240 and 480 GB versions accommodate 64 Gbit dies, while the 960 GB version uses 128 Gbit dies.

SmartFlush and GuaranteedFlush

Corsair mentions the SmartFlush and GuaranteedFlush technologies in the data to protect against data corruption in the event of a power failure. The aim of SmartFlush is to minimize the time that data is in the cache. However, this only reduces the likelihood that data will still be in the cache in the event of a power failure, or that there is less data in it.

GuaranteedFlush, on the other hand, relies on the flush cache command (E7h) of the ATA specification, which is supported by the Phison S10. This command can be used to trigger the clearing of the cache on the NAND or hard disk in a controlled manner. In this way, critical data such as the mapping table can be kept consistent.

As with most consumer drives, there are no capacitors on the SSD that can supply the drive with power for a short period of time. At this point, for the sake of comparability, it must be pointed out again that drives like the Crucial MX100 were initially assumed to have corresponding capacitors. But this is not what we are looking for here have been received.

In summary, it can be said that the Corsair XT, like the competition, does not have any "emergency power capacitors", but can still ensure the consistency of the mapping table and existing data. Data to be written is passed on from the cache to the NAND flash as quickly as possible in order to minimize the loss of data that has not yet been written in the event of a power failure.

Equipment

The SSD encrypts data with 256-bit AES, but unfortunately does not support the TCG Opal specifications, which are necessary for the use of Microsoft's eDrive standard, for example. Unfortunately, there is also no support for the additional energy-saving modes such as DevSleep and HIPM + DIPM. The drive is therefore less suitable for mobile corporate use or use in small mobile hardware, where very low idle consumption is important.

Lifespan

Corsair guarantees a write performance of 124 terabytes for the life of the flash. Since the value is the same for the three drive types, it can be assumed that the value is set quite low. On a 960 GB drive, which is four times as large, the cells would only be loaded a quarter as often as if the 124 terabytes were written to the small 240 GB version. The bottom line is that you get a guaranteed write volume of almost 70 gigabytes per day if you take the five-year guarantee period as a basis.

Impressions

 

As with many competitors, the SSD comes with a 2m installation spacer in bright red.

The Phison S45 controller rotated by 10 ° is clearly visible. In the 480 GB version, the NAND flash memory was divided into eight packages, each with eight dies with 64 Gbit each. In the 980 Gbyte variant, however, dies with 128 Gbit are installed in order to achieve the necessary capacity with the same number of packages.

Software equipment

The SSD does not come with any extra software, but Corsair still provides the SSD toolbox. This can be used to carry out functions such as overprovisioning, secure erase and the cloning of drives and to read drive and SMART information. A description of the software can be found here .

If you want, you can use other means to ensure that the operating environment is optimally matched to the SSD drives. Important parameters are:

• Does the SATA port run in AHCI mode?
• Does the operating system support TRIM?
• Has any automatic defragmentation of the operating system been deactivated?

Test environment

Hardware

Test station:

• CPU: Intel Core i3 3220 - 2 x 3,3 GHz (Turbo: off) [Amazon offers]
• Motherboard: ASUS P8H77M (H77 chipset) [Amazon offers]
• Memory: 8 GB (4 x 2 GB) Team Xtreem - SPD operation: DDR3-1333 9-9-9-24-1T at 1,5 volts [Amazon offers]
• Power adapter: NZXT 650 Watt HALE82 Series [Amazon offers]
• Boot drive: OCZ Vertex-2-SSD as boot drive [Amazon offers]

The test candidate:

• Corsair Neutron XT 480GB (Amazon offers)

Comparison models:

• AMD / OCZ Radeon R7 240 GB (HT4U test / Amazon offers)
• Corsair Force LX 256GB (HT4U test / Amazon offers)
• Corsair GTX 480GB (HT4U test / Amazon offers)
• Crucial M550 256GB (HT4U test / Amazon offers)
• Crucial M550 1TB (HT4U test / Amazon offers)
• OCZ ARC 100 2406 GB (HT4U test / Amazon offers)
• Samsung 840 120GB (HT4U test / Amazon offers)
• Samsung 840 EVO 250GB (HT4U test / Amazon offers)
• Samsung 840 Pro 256GB (HT4U test / Amazon offers)
• SanDisk Extreme 240GB (HT4U test / Amazon offers)
• SanDisk Extreme II 240GB (HT4U test / Amazon offers)
• SanDisk UltraPlus 256GB (HT4U test / Amazon offers)

Software

Our benchmark course

Our benchmark course aims to answer the following questions:

• How fast is the SSD reading and writing large files sequentially and reading and writing small files at random?
• How do fragmented blocks (not to be confused with file fragmentation!) And the resulting read-modify writes affect performance after a heavy write load?
• How fast is the SSD in a continuous load scenario (steady state)?
• Can TRIM restore full performance?
• How effective is garbage collection?
• How fast is the SSD when certain mixes of large and small blocks occur?

Synthetic benchmarks

The use of synthetic benchmarks cannot be avoided, since only with these the technical limits of the SSDs become visible. They show the maximum achievable.

With these benchmarks we determine the performance in the following states:

In this way it can be seen whether and to what extent the performance of the SSD is falling and whether TRIM can restore the original performance.

It doesn't matter whether you copy a few hundred MP3 or video files or simulate this work with Iometer, the effort is the same for the SSD. Differences resulting from the file system of the operating system then affect all SSDs equally, so that the ratios of the performance differences remain the same.

Trace benchmarks

Real life, on the other hand, can be simulated using trace benchmarks such as PCMark or Iometer profiles, which simulate use cases. With these tests, practical accesses are carried out in a reproducible manner.

For practical results, we carry out these tests after the SSD has already been written with load profiles several times and is occupied with active data except for a remaining 10 GB. This gives you the performance values ​​of an SSD that has already been used and is currently mostly full.

Treatments

We test less per application itself. There are two main reasons for this: First, the CPU limit falsifies the performance gap between the SSDs. For example, when the SSD has to wait for the CPU to process certain data before the SSD can continue working when the application starts. Due to the CPU limit, the SSDs move closer together than would be the case with faster CPUs later. Second, many applications can only be measured with a stopwatch, which is too imprecise for us, especially since the results are sometimes only tenths of a second apart. However, we carry out our long-serving OpenOffice copy test because it is easy to reproduce. We have only increased the amount of data there by a factor of 12. It is now 3,06 GB of data in over 48.000 files of various sizes that will be duplicated on the test drive.

Continuous load measurements

As described in the section “Load behavior”, SSDs collapse under a continuous random write load if the garbage collection cannot provide free blocks quickly enough. Such a load behavior occurs only rarely in normal home use. For one or the other reader, however, it might be interesting whether an SSD is also suitable for somewhat tougher use. For example, as a data carrier for a virtualizer, where a lot of small accesses can occur in parallel, or as a disk for a database test environment.

For this test, we unleash as many 4k writes as possible on the SSD via Iometer and create a graph that shows the performance over time. We repeat this test after a 30-minute or 12-hour break to see whether the garbage collection was able to provide enough free blocks for high performance during this time. Since Iometer works with a large test file, which is never deleted but only overwritten, influences by TRIM in these two repeat runs are excluded. The increase in performance through TRIM itself is then measured in a fourth run. This takes place after a quick format, which "trims" the drive. The test file is then created again.

We would like to point out that this goes well beyond the normal requirements for SSDs for home use. If an SSD does not do so well here, it is therefore not counted negatively. But we want to find out which SSDs stand out positively from the crowd. In addition, this test makes it easier to see to what extent the garbage collection is working.

MByte / s or IOPS?

Usually we give the measurement results in megabytes per second. In the profile tests, however, we give the results in IOPS (Input / Output Operations per Second = input and output commands per second). An input or output command can mean reading or writing a block. This does not affect the comparability. If a data carrier manages 128 IO per second in a write test with 1.000 KB blocks, then mathematically this results in 1.000 * 128 KB = 128 MB per second. When an operating system writes MP3 files or videos, it does so in blocks, too, and the block sizes ultimately depend on the size of the files and the formatting of the file system. With many small files, this may limit the number of IOPS and with large files the maximum write rate of the SSD. Therefore, it makes sense to use the specification of IOPS wherever a high number of read and write operations take place and / or different block sizes are involved.

In the case of continuous load measurements, the indication in IOPS has the additional advantage that the maximum IOPS information usually advertised by the manufacturers can be compared directly with the real results.

Measurement results

Sequential reading

The drive tops the field in both benchmarks. But since the SATA interface is the bottleneck here, the drive cannot logically separate itself and the differences are minimal.

Sequential writing

Sequential writing is also very quick. On the other hand, like the Samsung models, the drive drops heavily after load. The reason for this is the same as with the Samsung models and is visible and discussed in the chapter "Continuous load curves".

Random reading

When reading randomly, the picture is split. If you only measure over a short period of time like with AS-SSD, the Corsair XT is ahead. But if you measure consistently over a longer period of time as in Iometer, the drive is only in the middle.

Random writing

Here, too, the Corsair XT cannot stand out from the field and suffers from poor performance when writing after load. The previous GTX model is faster in both Iometer and AS-SSD.

Web server, file server, workstation

The tests with the load profiles show once again that the drive is noticeably behind the top field in the write-oriented load tests with random patterns. The lag is not so great in the read-intensive web server profile.

HT4U OpenOffice copy test

PCMark7 trace benchmarks

Continuous load curves

A few things can be seen in this diagram for the Neutron XT: First, the high initial power can be maintained over a relatively long period of time. When fresh or completely trimmed, the initial level is maintained for almost five minutes. At around 55.000 IOPS, this is not particularly high compared to the competition, which of course means that more time passes before the SSD saturates. The longer high level is certainly also favored by the fact that 512 GB gross are available on the SSD and the difference to the 480 GB user space is used for spare area and parity data. A long, high initial level indicates that a lot of this memory is used as a spare area, i.e. relatively little parity data is kept. The SSD will not be able to compensate for the failure of a complete die.

Furthermore, the Neutron XT does not perform any preventive garbage collection when it is idle, because the green graph "after 12h" idle starts at a low level. Only the TRIM command ensures higher performance again. This is disadvantageous for all users who write to existing files with smaller blocks.

The course of the power curves is strongly dynamic, because even at the high initial level there are strong periodic drops. On the other hand, under constant load there are periodically short phases with high performance when the garbage collection has ensured free blocks under load. These peaks ensure the higher average value under continuous load compared to many competitors. But no matter from which direction you look at it, the previous Neutron GTX is more stable and faster under load.

input

The idle consumption is very low. Since the SSD does not support any other energy-saving modes such as DevSleep and the like, some energy is wasted on newer systems when idling. In a PC, the 0,4 W potential savings are certainly not important to most. The energy consumption during reading and writing is in the middle.

For technological reasons, paperwork is the most power-hungry, as the cells then have to be charged and discharged and the controller has the most computing effort. Consumption of 5 watts sounds like a lot for SSDs, but you shouldn't lose sight of the fact that in most cases these are idle, and when something needs to be done, it's mostly read access. You should therefore always see these results in relation to your own use of the SSD.

Summary

Corsair is promoting the XT as the most powerful solution in its portfolio for the most demanding requirements in the areas of games, graphics and video. If you look at the areas of sequential processing that are important for graphics and video, then the statement is correct, because here the XT is at the front of the field in our measurements, both for reading and writing. Reading files is just as important for games, but random access also plays a role here. Interestingly, the XT is noticeably ahead under the AS-SSD, but can only land in the middle field under constant load under the Iometer.

If you also consider other areas such as workstation or semi-professional server use, the XT cannot hold its own against other performance SSDs such as the AMD Radeon R7 from OCZ, the M550 from Crucial or the SanDisk Extreme II. Even the predecessor Corsair Neutron GTX performs noticeably better here. The XT suffers from the same handicap as Samsung's 840 series: In the absence of proactive garbage collection during idle times, only TRIM and the performance-consuming garbage collection under load ensure order.

If we compare the current prices (April 2015) of the 480/512 GB models from the performance segment, the following picture emerges:

* - The somewhat older Corsair GTX was only listed on Geizhals by a lesser-known dealer. This price therefore comes from idealo, where several listings were entered. The prices for the discontinued model are certainly no longer representative.

Corsair's Neutron XT is the manufacturer's new performance flagship. It slams larger data into the Flash a little faster than the competition, but cannot quite keep up with the current pricing. This is mainly due to the competitive price of the Crucial M550. The Micron daughter can apparently afford that with her mother behind her. The competition is currently not (yet) willing to accept such a price point. The new Neutron XT is ideal for those who do not leave the video / graphics / games area and is a really good choice there. But the Corsair surcharge has to be met.

[ri], April 13, 2015