OCZ ARC 100 in the test

Intro

OCZ was a pioneer in the consumer SSD industry. Unfortunately, the reputation of its SSDs suffered in the past due to the relatively high failure rates of some models. After the bankruptcy and the rebuilding under Toshiba, OCZ re-enters the consumer market with the ARC 100. A sophisticated Barefoot 3 controller and reliable Toshiba Flash should convince and win over customers. In today's test we clarify what is hidden behind the ARC 100.

Originally known as a memory and power supply manufacturer, OCZ was a pioneer in the consumer SSD industry and especially in the home user sector. After the first consumer SSD models from various manufacturers appeared on the market, some of which suddenly put in seconds of thought while writing and brought the system to a standstill, OCZ launched models that were price-performance-related with popular models from Intel such as the X25-M series could compete. Together with Barefoot-1 controllers from Indilinx and controllers from SandForce, OCZ released a relatively large number of different models, some of which also became very popular.

However, the reputation suffered due to the relatively high failure rates of some consumer models, and in the forums there are sometimes even reservations about the durability against SandForce controllers, although the same controller in models from other manufacturers did not attract negative attention in this regard. OCZ was finally taken over by Toshiba after bankruptcy, which eliminated the actual cause of the high failure rates and losses at OCZ. In the following section we will go into more detail about the causes and the reorganization, because these are relevant if you want to consider the reliability of the new OCZ (OCZ Storage Solutions) with the old OCZ (OCZ Technology Group).

The ARC 100 now represents OCZ's return to the entry segment and aims to impress with a good price-performance ratio. And although the model was placed in the entry-level segment, it should already convince with a higher long-term performance. Our test will clarify how the model performs.

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:

• 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

OCZ's eventful past

As already mentioned at the beginning, OCZ initially brought a large number of SSD models onto the market and thus gained some popularity. In March 2011, OCZ bought the controller manufacturer Indilinx and thus acquired the ability and patents to develop controllers in-house. For NAND Flash, however, they were still dependent on purchases from other manufacturers. At this time, the market demand for SSDs rose sharply, while the production of NAND flash could not yet meet this due to a lack of sufficient capacities.

Got OCZ media reports was having trouble getting enough NAND and was unable to meet sales targets. This helped put the company in the red. And there were more dark clouds on the horizon: The number of SSD returns, i.e. defective OCZ SSDs at customers, rose and rose. It is possible that OCZ made too many compromises in terms of the quality of the purchased NAND Flash, be it in order to achieve sales targets or to enter the market at a low price. Others speculated on problems with the SandForce controllers used, which were not so noticeable in competing products.

Response figures are rarely disclosed, but when retailers from the SSD market do leaked, the then OCZ Technology Group performed relatively poorly in comparison with the competition. The French website hardware.fr (or, until 2012, its English-language website BeHardware.com) publishes the return figures for various hardware components from a European etailer every six months. The figures are therefore not representative of the world, but they do provide a guide. The Octane and Petrol models were particularly noticeable with rates of up to 40%, but Agility and Vertex SSDs also stood out with rates of over 5%.

Defective returns cause dissatisfied customers and costs for warranty and guarantee expenses. If the failure rates are significantly higher than expected, these costs can also fall on the company's feet. After the company had already had several quarters of losses, OCZ encountered further supply and delivery problems in October 2013 and filed for bankruptcy in November. The power supply division was then sold to Firepower Technology (USA) in February 2014 and memory production was discontinued beforehand. The technology group Toshiba was interested in the SSD sector - unexpectedly for many.

The takeover by Toshiba

In December 2013 it was announced that Toshiba would like to buy the relevant parts of the company. If you looked at the consumer market, this decision was incomprehensible to some, because you still had the high error rate of consumer products in mind. OCZ had started some time ago to build products for corporate solutions. Since these played in a different, significantly higher price range, there were no such failure problems that were caused by cheap NAND.

Toshiba is also a Flash manufacturer itself. As a Toshiba subsidiary, OCZ Storage Solutions now has direct access to the flash memory of an experienced and well-established memory manufacturer who produces both eMLC flash for business solutions and MLC flash for consumer SSDs. The problem of the "old" OCZ of getting reliable flash memory in larger quantities no longer exists. And by purchasing OCZ, Toshiba ultimately has a company in the group that can access controllers, firmware development and flash memory within the group. With the appropriate financial support, Toshiba / OCZ will now try to compete against other in-house developers such as Samsung and Intel.

The ShieldPlus guarantee

This composition might not look bad on paper, but more is needed to regain customer trust. That was also clear to OCZ, which is why they were ShieldPlus guarantee introduced. This applies to the ARC 100 and Vertex 460A for three years each, and four years for the AMD Radeon R7 SSD, which OCZ builds with and for AMD.

In practice, the procedure works like this: If you have a problem with the SSD, contact support. The serial number of the model is sufficient as legitimation, no proof of purchase is required. If the support determines that there is obviously a defect, a new SSD will be sent directly to the customer, together with a return slip for the free return of the old SSD.

OCZ has a page on the topic on its website Quality Management dedicated. In addition to the usual commitments and graphics, there is a representation relatively far down on the page that deals with the error rates from the past.

OCZ is obviously convinced of the reliability of the new products and secures the buyer accordingly with the guarantee. The advance exchange by OCZ is much more convenient for the customer in the case of a guarantee, because it is simply much faster than sending in a defective device and only getting a replacement after weeks.

The test candidate

Key data and technology

The ARC 100 is available in the variants 120, 240 and 480 GB. The Barefoot 3 family used has now been on the market for over two years; the slightly lower clocked M100 version of the controller is used in the ARC 10. One can expect that it is appropriately mature and that the firmware programmers are already well acquainted with it.

The maximum read rates, which at 475 to 490 MB / s, do not fully exhaust the SATA interface, are noticeable. This is a bit unusual now, as you have got used to the fact that almost all SSDs make the leap over 500 MB / s. Since permanent sequential reading is a special case in practice, the practical tests have to show whether this has any negative effects in real use.

Equipment

The placement in the entry-level segment takes its toll on the equipment list. 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. When you pick up the drive, you notice that, at 110 grams, it feels heavier compared to competing models (e.g. Samsung 850 Evo with 66 grams). Both of these factors don't make it an ideal notebook drive. Low weight and low idle consumption are essential for mobile platforms.

Lifespan

Now let's take a look at durability. OCZ promises the customer an average write volume of 20 GB per day over three years "with typical end-user loads". The manufacturer explicitly mentions workstations as an area of ​​application. According to Adam Riese, the bottom line is a guaranteed write volume of 21 terabytes. The Kitguru.net website has received five ARC-100 drives and is currently putting them to an endurance test. As of this writing, all five drives already have the 200 terabyte mark exceeded and thus already achieved ten times as much as promised by the manufacturer.

Impressions

In this price segment, additions are not common. There are no physical additions in the packaging.

In the 256 GB version, the NAND flash memory was divided into 16 packages, 8 each on the front and rear. As a result, there are more dies in the individual packages in the 512 GB variant.

Software equipment

With the OCZ Toolbox, the customer gets a tool for firmware updates and checking the SSD properties. A YouTube video illustrates this process. The tool can be used for Microsoft Windows 7 and 8 (.1), Linux and Mac downloaded .

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:

• OCZ ARC 100 240 GB (Amazon offers)

Comparison models:

• 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)
• 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.

Applications

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 let go of as many 4K writes as possible to the SSD using an 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 that is never deleted, but only overwritten, any influence from TRIM in these two repetitions is excluded. The increase in performance through TRIM itself is then measured in a fourth run. This takes place after a quick format, whereby the drive is "trimmed". 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.

With the steady-state measurements, the information in IOPS has the additional advantage that you can directly compare the maximum IOPS information usually advertised by the manufacturers with the real results.

Measurement results

Sequential reading

As already announced in the data sheets, the sequential reading performance is up to 20% behind the competition. In daily use, however, this is seldom noticeable, if at all, since most read accesses do not consist of reading gigabytes sequentially. Exceptions are the copying of very large files - assuming a data carrier to which the copy can be written quickly enough - as well as the field of video processing, provided the CPUs and the processing program can read more than 450 MB / s.

Sequential writing

The sequential write performance under Iometer, on the other hand, looks much better. Although positioned as an entry-level device, its sequential write performance is ahead of most other entry-level SSDs. The SanDisk Ultra Plus is faster, but it collapses more strongly after load.

In the AS SSD benchmark with its very short sequential write spurts, the EVO competitor with its TurboWrite mechanism can still outperform, but otherwise the distribution remains more or less the same.

Random reading

But even with random 4K read accesses, the ACR 100 cannot distinguish itself. It occupies the penultimate place for both the Iometer and the AS-SSD. We will see later in the web server and application benchmarks whether this disadvantage also has an effect in practice.

Random writing

Here, however, our test subject under Iometer turns up: In the entry-level segment, only the Crucial MX100 has to admit defeat. But what is much more conspicuous and remarkable: The ARC 100 barely breaks in the load scenario and is even in the range of the performance model Corsair Neutron GTX. This was even equipped with twice as much Flash and thus a larger spare area.

The ARC 100 also plays a leading role in the short write bursts of AS-SSD and only has to admit defeat to the M550 performance model from Crucial.

Web server, file server, workstation

The web server benchmark allows data of various block sizes to be read continuously and shows that the lower synthetic reading performance does not have as negative an effect in practice as the synthetic benchmarks suggest. In the entry-level segment, the ARC 100 only has to admit defeat to the Samsung EVO.

In the two write-oriented tests, the ARC 100 wipes the floor with the competition. Only the 1 TB version of the Crucial M550 is ahead of it in the file server test. That this is mainly due to the larger capacity can be seen in the 256 GB version of the M550, because it is only half as fast as the ARC 100. In the workstation test, the OCZ SSD is even ahead of the 1-TB-M550. OCZ's promise of long-lasting write performance under load can confidently be regarded as fulfilled.

HT4U OpenOffice copy test

In the copy test, the test person runs along in the middle field without any abnormalities.

PCMark7 trace benchmarks

With the slight reading weakness and the strong writing behavior, the ARC 100 cannot really set itself apart from the competition in the more reading-oriented practical tests.

Continuous load curves

Under a continuous load of random 4K write processes with a queue depth of 32, the OCZ ARC 100 starts with values ​​around 86.000 IOPS at a very good level. As with any SSD, the performance drops if there are no more free blocks and the garbage collection has to actively ensure free blocks even while writing. The ARC 100 falls to an average of only 18.000 IOPS, which puts it well ahead of the competition.

Furthermore, it can be stated that the controller clears some of the blocks via garbage collection during idle times. He showed that this cannot be taken for granted Test the Samsung 840 Pro, which therefore also did relatively poorly for a performance model in the load scenario in the sequential Iometer write test.

input

The idle rates are relatively high at around 1 watt. Since additional energy-saving modes such as DevSleep are not supported, the energy consumption cannot be reduced any further.

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

OCZ wants to regain trust with its new products. With the Toshiba NAND Flash that is now available and the sophisticated controller, the chances are very good that this will succeed. The long-term test by Kitguru.net mentioned in the section "Lifespan" also suggests this. In the absence of a crystal ball, binding statements can never be made in this area anyway.

However, it is difficult to believe that the controller, in conjunction with the large number of die packages on the SSD, cannot parallelize enough read processes to exhaust the SATA interface. Either this is an artificial limitation to differentiate the beginners from the high-performance models in the OCZ portfolio, or it is a trade-off in favor of the higher continuous writing performance. Whatever the cause, the ARC 100 is a well-rounded product with two small notches: not so good for mobility, and the poorer reading performance. On the other hand, it can boast with its significantly better write performance, whereby it outperforms the entry-level models of the competition in terms of sequential write rates and even outclasses the performance models in the steady state and server tests.

Let's compare the current prices with the other models from the entry-level segment:

With the ARC 100, OCZ directly challenges Crucial's price fighter MX100. The strengths of the ARC 100 lie in its very stable writing performance and the service (in the event of a defect, replacement in advance by quoting the serial number, even without an invoice). The MX100, on the other hand, has better reading performance and does not do without eDrive compatibility and additional energy-saving modes. Here everyone has to decide according to their own needs. With future products, OCZ should not lose sight of the fact that these features are also desired by customers in the entry-level segment. All in all, the ARC 100 is an impressive performance in terms of price-performance ratio.

[ri], February 2, 2015