When we visited Adata at Computex just a few weeks ago, it was clear that the company’s focus was on M.2 and PCIe storage. A representative even went as far as to say that Adata may phase out SATA drives entirely, replacing them with PCIe x2 drives to sate the budget market.
But there’s no telling when that day may come. For the time being, Adata is still hard at work designing and releasing upgrades to its existing SATA offerings. We have one such upgrade on hand today: the Ultimate SU900. As outlined in the table below, the lineup includes 256GB, 512GB, 1TB, and 2TB versions.
|Adata Ultimate SU900|
|Capacity||Max sequential (MB/s)||Max random (IOps)|
Like the SU800 before it, the SU900 stores your bits in IMFT’s 3D NAND. This time around, though, Adata ponied up for MLC chips instead of TLC. This is the opposite of what we’ve seen most brands do of late—Samsung and Micron both used their respective non-planar technologies as an excuse to move product lines from MLC to TLC. Adata’s reversal is a welcome change of pace. The SU900’s MLC NAND should yield appreciable speed and endurance increases over the SU800.
Adata could only send us a 256GB sample of the SU900, so we won’t have an apples-to-apples comparison against the SU800 512GB we tested some months ago. We’ll forge ahead regardless. First, we’ll void the warranty and sneak a peek under the hood.
Aside from the upgrade to MLC, little distinguishes the SU900 from the SU800. Silicon Motion’s SM2258 is running the show again, and it still offers pseudo-SLC caching for burst performance when the DRAM cache is overwhelmed. The NAND dies are distributed among four packages, two on either side of the PCB. The chips themselves are IMFT’s 32-layer, 256Gb stuff. The 256GB drive doesn’t require enough of these dies for the controller to leverage significant interleaving over each of its four channels. We expect higher capacity SU900s to operate faster for this reason, even if Adata’s specs don’t suggest so.
Adata is bullish on the SU900’s endurance, backing the 256GB drive with a five-year warranty and a 200 terabytes-written specification. Newegg is currently hocking the drive for $110, but Amazon will send the 256GB unit your way for an even hundo. Those prices may seem dear, but that’s simply the state of the SSD market given the global NAND shortage. Competitors’ products are no cheaper. Let’s escape from our wallets’ troubles by diving straight into testing.
IOMeter — Sequential and random performance
IOMeter fuels much of our latest storage test suite, including our sequential and random I/O tests. These tests are run across the full capacity of the drive at two queue depths. The QD1 tests simulate a single thread, while the QD4 results emulate a more demanding desktop workload. For perspective, 87% of the requests in our old DriveBench 2.0 trace of real-world desktop activity have a queue depth of four or less. Clicking the buttons below the graphs switches between results charted at the different queue depths.
Our sequential tests use a relatively large 128KB block size.
The SU900’s sequential read speeds are as quick as any SATA drive’s can be, demonstrating a 25% improvement over Adata’s older (and planar) MLC drive, the XPG SX930. The SU900’s’ sequential writes, however, seem to suffer in the same way that the SU800’s did.
At that time, Adata’s explanation was that the drive’s pseudo-SLC cache hadn’t recovered in time to be of any use during IOMeter’s scripted write testing. The resulting disappointing numbers reflected the raw write speed of the NAND. On the upside, those raw numbers are better for the SU900’s 3D MLC than they were for the SU800’s 3D TLC.
The SU900 will have plenty of opportunities in our test suite to make up for its sequential write results, so let’s move on to random response times.
The drive’s read response times are right in the middle of the pack. Write response times are a bit lackluster, but we’ve seen worse before.
Thus far, the SU900 has displayed blazing fast reads, but it’s suffered a setback in its sequential write performance. Let’s see how it fares under more complex workloads.
Sustained and scaling I/O rates
Our sustained IOMeter test hammers drives with 4KB random writes for 30 minutes straight. It uses a queue depth of 32, a setting that should result in higher speeds that saturate each drive’s overprovisioned area more quickly. This lengthy—and heavy—workload isn’t indicative of typical PC use, but it provides a sense of how the drives react when they’re pushed to the brink.
We’re reporting IOps rather than response times for these tests. Click the buttons below the graph to switch between SSDs.
The SU900’s peak speeds are extremely high, but it can’t hold on to them for very long. It oscillates rapidly before quickly collapsing to a low steady-state rate. Let’s see what the actual numbers are.
The SU900’s peak is the highest we’ve seen from a 250GB-class drive, whether SATA or PCIe. It’s important to bear in mind that the drive can’t sustain that rate for any substantial time period. Its steady-state speeds, by contrast, are quite pedestrian. Not markedly worse than other 250GB-class drives, but not better either.
Our final IOMeter test examines performance scaling across a broad range of queue depths. We ramp all the way up to a queue depth of 128. Don’t expect AHCI-based drives to scale past 32, though—that’s the maximum depth of their native command queues.
For this test, we use a database access pattern comprising 66% reads and 33% writes, all of which are random. The test runs after 30 minutes of continuous random writes that put the drives in a simulated used state. Click the buttons below the graph to switch between the different drives. And note that the P3700 plot uses a much larger scale.
The SU900’s scaling curve is barely a curve. The drive doesn’t exhibit any willingness to ramp up performance as queue depth increases. At least its speeds didn’t regress. Just for fun, let’s compare it to some other drives.
The SU800 512GB, SP550 480GB, and MX300 750GB aren’t particularly inclined to scale with queue depth, but their results look downright curvaceous next to flatlands of the SU900.
The SU900 didn’t hit us with any nasty surprises during our sustained and scaling testing. Everything falls in line with expected performance for a 250GB-class SATA drive. Next up, real-world tests.
TR RoboBench — Real-world transfers
RoboBench trades synthetic tests with random data for real-world transfers with a range of file types. Developed by our in-house coder, Bruno “morphine” Ferreira, this benchmark relies on the multi-threaded robocopy command build into Windows. We copy files to and from a wicked-fast RAM disk to measure read and write performance. We also cut the RAM disk out of the loop for a copy test that transfers the files to a different location on the SSD.
Robocopy uses eight threads by default, and we’ve also run it with a single thread. Our results are split between two file sets, whose vital statistics are detailed below. The compressibility percentage is based on the size of the file set after it’s been crunched by 7-Zip.
|Number of files||Average file size||Total size||Compressibility|
The media set is made up of large movie files, high-bitrate MP3s, and 18-megapixel RAW and JPG images. There are only a few hundred files in total, and the data set isn’t amenable to compression. The work set comprises loads of TR files, including documents, spreadsheets, and web-optimized images. It also includes a stack of programming-related files associated with our old Mozilla compiling test and the Visual Studio test on the next page. The average file size is measured in kilobytes rather than megabytes, and the files are mostly compressible.
RoboBench’s write and copy tests run after the drives have been put into a simulated used state with 30 minutes of 4KB random writes. The pre-conditioning process is scripted, as is the rest of the test, ensuring that drives have the same amount of time to recover.
Let’s take a look at the media set first. The buttons switch between read, write, and copy results.
The SU900 puts on a strong showing in our media set tests. Across read, write, and copy, it lands among the upper echelon of SATA drives. The heavy sequential writes demanded by our media write test pose no problem for the drive, proving that its pseudo-SLC and DRAM caches function just fine in real-world conditions. The SU900 matches or surpasses the SU800’s speeds across the board, unfazed by its capacity handicap.
Next up, the work set.
The work set tests tell the same story. The SU900’s real-world random performance is up to snuff, landing the drive in the middle of the pack and just behind its predecessor’s.
The SU900 breezed through all of our RoboBench tests with ease. The drive’s real-world speeds are far better than the last Adata MLC drive we tested, the planar-MLC-powered SX930 240GB. On the next page, we’ll subject the drive to our boot and load tests.
Until now, all of our tests have been conducted with the SSDs connected as secondary storage. This next batch uses them as system drives.
We’ll start with boot times measured two ways. The bare test depicts the time between hitting the power button and reaching the Windows desktop, while the loaded test adds the time needed to load four applications—Avidemux, LibreOffice, GIMP, and Visual Studio Express—automatically from the startup folder. Our old boot tests focused on the time required to load the OS, but these new ones cover the entire process, including drive initialization.
The SU900 lands right next to its bigger, elder brother. Both drives boot quickly and without fuss.
Next, we’ll tackle load times with two sets of tests. The first group focuses on the time required to load larger files in a collection of desktop applications. We open a 790MB 4K video in Avidemux, a 30MB spreadsheet in LibreOffice, and a 523MB image file in the GIMP. In the Visual Studio Express test, we open a 159MB project containing source code for the LLVM toolchain. Thanks to Rui Figueira for providing the project code.
No anomalies here. The SU900 loads our productivity applications with typical SSD speeds.
Game load times also reveal nothing unusual. The SU900 is a fine home for your Steam sale purchases, but the 256GB version might fill up rather quickly.
The SU900 turned in a faultless performance as a primary storage drive. With that, we’re out of tests. Flip to the next page for a summary of our test methods, or skip ahead to the conclusion.
Test notes and methods
Here are the essential details for all the drives we tested:
|Adata Premier SP550 480GB||SATA 6Gbps||Silicon Motion SM2256||16-nm SK Hynix TLC|
|Adata Ultimate SU800 512GB||SATA 6Gbps||Silicon Motion SM2258||32-layer Micron 3D TLC|
|Adata Ultimate SU900 256GB||SATA 6Gbps||Silicon Motion SM2258||Micron 3D MLC|
|Adata XPG SX930 240GB||SATA 6Gbps||JMicron JMF670H||16-nm Micron MLC|
|Corsair MP500 240GB||PCIe Gen3 x4||Phison 5007-E7||15-nm Toshiba MLC|
|Crucial BX100 500GB||SATA 6Gbps||Silicon Motion SM2246EN||16-nm Micron MLC|
|Crucial BX200 480GB||SATA 6Gbps||Silicon Motion SM2256||16-nm Micron TLC|
|Crucial MX200 500GB||SATA 6Gbps||Marvell 88SS9189||16-nm Micron MLC|
|Crucial MX300 750GB||SATA 6Gbps||Marvell 88SS1074||32-layer Micron 3D TLC|
|Intel X25-M G2 160GB||SATA 3Gbps||Intel PC29AS21BA0||34-nm Intel MLC|
|Intel 335 Series 240GB||SATA 6Gbps||SandForce SF-2281||20-nm Intel MLC|
|Intel 730 Series 480GB||SATA 6Gbps||Intel PC29AS21CA0||20-nm Intel MLC|
|Intel 750 Series 1.2TB||PCIe Gen3 x4||Intel CH29AE41AB0||20-nm Intel MLC|
|Intel DC P3700 800GB||PCIe Gen3 x4||Intel CH29AE41AB0||20-nm Intel MLC|
|Mushkin Reactor 1TB||SATA 6Gbps||Silicon Motion SM2246EN||16-nm Micron MLC|
|OCZ Arc 100 240GB||SATA 6Gbps||Indilinx Barefoot 3 M10||A19-nm Toshiba MLC|
|OCZ Trion 100 480GB||SATA 6Gbps||Toshiba TC58||A19-nm Toshiba TLC|
|OCZ Trion 150 480GB||SATA 6Gbps||Toshiba TC58||15-nm Toshiba TLC|
|OCZ Vector 180 240GB||SATA 6Gbps||Indilinx Barefoot 3 M10||A19-nm Toshiba MLC|
|OCZ Vector 180 960GB||SATA 6Gbps||Indilinx Barefoot 3 M10||A19-nm Toshiba MLC|
|Patriot Hellfire 480GB||PCIe Gen3 x4||Phison 5007-E7||15-nm Toshiba MLC|
|Plextor M6e 256GB||PCIe Gen2 x2||Marvell 88SS9183||19-nm Toshiba MLC|
|Samsung 850 EV0 250GB||SATA 6Gbps||Samsung MGX||32-layer Samsung TLC|
|Samsung 850 EV0 1TB||SATA 6Gbps||Samsung MEX||32-layer Samsung TLC|
|Samsung 850 Pro 500GB||SATA 6Gbps||Samsung MEX||32-layer Samsung MLC|
|Samsung 950 Pro 512GB||PCIe Gen3 x4||Samsung UBX||32-layer Samsung MLC|
|Samsung 960 EVO 250GB||PCIe Gen3 x4||Samsung Polaris||32-layer Samsung TLC|
|Samsung 960 EVO 1TB||PCIe Gen3 x4||Samsung Polaris||48-layer Samsung TLC|
|Samsung 960 Pro 2TB||PCIe Gen3 x4||Samsung Polaris||48-layer Samsung MLC|
|Samsung SM951 512GB||PCIe Gen3 x4||Samsung S4LN058A01X01||16-nm Samsung MLC|
|Samsung XP941 256GB||PCIe Gen2 x4||Samsung S4LN053X01||19-nm Samsung MLC|
|Toshiba OCZ RD400 512GB||PCIe Gen3 x4||Toshiba TC58||15-nm Toshiba MLC|
|Toshiba OCZ VX500 512GB||SATA 6Gbps||Toshiba TC358790XBG||15-nm Toshiba MLC|
|Transcend SSD370 256GB||SATA 6Gbps||Transcend TS6500||Micron or SanDisk MLC|
|Transcend SSD370 1TB||SATA 6Gbps||Transcend TS6500||Micron or SanDisk MLC|
All the SATA SSDs were connected to the motherboard’s Z97 chipset. The M6e was connected to the Z97 via the motherboard’s M.2 slot, which is how we’d expect most folks to run that drive. Since the XP941, 950 Pro, RD400, and 960 Pro require more lanes, they were connected to the CPU via a PCIe adapter card. The 750 Series and DC P3700 were hooked up to the CPU via the same full-sized PCIe slot.
We used the following system for testing:
|Processor||Intel Core i5-4690K 3.5GHz|
|Platform hub||Intel Z97|
|Platform drivers||Chipset: 10.0.0.13
|Memory size||16GB (2 DIMMs)|
|Memory type||Adata XPG V3 DDR3 at 1600 MT/s|
|Audio||Realtek ALC1150 with 126.96.36.19944 drivers|
|System drive||Corsair Force LS 240GB with S8FM07.9 firmware|
|Storage||Crucial BX100 500GB with MU01 firmware
Crucial BX200 480GB with MU01.4 firmware
Crucial MX200 500GB with MU01 firmware
Intel 335 Series 240GB with 335u firmware
Intel 730 Series 480GB with L2010400 firmware
Intel 750 Series 1.2GB with 8EV10171 firmware
Intel DC P3700 800GB with 8DV10043 firmware
Intel X25-M G2 160GB with 8820 firmware
Plextor M6e 256GB with 1.04 firmware
OCZ Trion 100 480GB with 11.2 firmware
OCZ Trion 150 480GB with 12.2 firmware
OCZ Vector 180 240GB with 1.0 firmware
OCZ Vector 180 960GB with 1.0 firmware
Samsung 850 EVO 250GB with EMT01B6Q firmware
Samsung 850 EVO 1TB with EMT01B6Q firmware
Samsung 850 Pro 500GB with EMXM01B6Q firmware
Samsung 950 Pro 512GB with 1B0QBXX7 firmware
Samsung XP941 256GB with UXM6501Q firmware
Transcend SSD370 256GB with O0918B firmware
Transcend SSD370 1TB with O0919A firmware
|Power supply||Corsair AX650 650W|
|Case||Fractal Design Define R5|
|Operating system||Windows 8.1 Pro x64|
Thanks to Asus for providing the systems’ motherboards, to Intel for the CPUs, to Adata for the memory, to Fractal Design for the cases, and to Corsair for the system drives and PSUs. And thanks to the drive makers for supplying the rest of the SSDs.
We used the following versions of our test applications:
- IOMeter 1.1.0 x64
- TR RoboBench 0.2a
- Avidemux 2.6.8 x64
- LibreOffice 4.3.2
- GIMP 2.8.14
- Visual Studio Express 2013
- Batman: Arkham Origins
- Tomb Raider
- Middle Earth: Shadow of Mordor
Some further notes on our test methods:
- To ensure consistent and repeatable results, the SSDs were secure-erased before every component of our test suite. For the IOMeter database, RoboBench write, and RoboBench copy tests, the drives were put in a simulated used state that better exposes long-term performance characteristics. Those tests are all scripted, ensuring an even playing field that gives the drives the same amount of time to recover from the initial used state.
- We run virtually all our tests three times and report the median of the results. Our sustained IOMeter test is run a second time to verify the results of the first test and additional times only if necessary. The sustained test runs for 30 minutes continuously, so it already samples performance over a long period.
- Steps have been taken to ensure the CPU’s power-saving features don’t taint any of our results. All of the CPU’s low-power states have been disabled, effectively pegging the frequency at 3.5GHz. Transitioning between power states can affect the performance of storage benchmarks, especially when dealing with short burst transfers.
The test systems’ Windows desktop was set at 1920×1080 at 60Hz. Most of the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.
The Adata SU900 put up solid numbers across the entire test suite, with the exception of our IOMeter sequential write test. That sore point will hurt the drive in our overall rankings, but its steadfast performance elsewhere should offset the effect to some degree. We distill the overall performance rating using an older SATA SSD as a baseline. To compare each drive, we then take the geometric mean of a basket of results from our test suite. Only drives which have been through the entire current test suite on our current rig are represented.
Despite its smaller size, the SU900 256GB grabs a noticeable lead over the 3D TLC-equipped SU800 512GB. It also edges out the planar MLC XPG SX930, even though that drive handled IOMeter without incident. The SU900 is certainly a better drive than the XPG SX930, as our real-world performance results attest. If Adata or Silicon Motion could work out the kinks with the SM2258, write cache flushing, and IOMeter, the SU900 would land in a more covetable place in our rankings and enjoy a more comfortable lead over the XPG SX930.
But even as it stands, the SU900 is in a fine spot for a 250GB-class drive. Now let’s see what its price proposition looks like within the broader landscape. In the plots below, the most compelling position is toward the upper left corner, where the price per gigabyte is low and performance is high. Use the buttons to switch between views of all drives, only SATA drives, or only PCIe drives.
Our four 250GB-ish SATA drives have carved out their own little turf on our plot. Within the context of those four drives, the SU900 256GB looks compelling. The XPG SX930 is a bit pricier for less performance, and the Transcend SSD370 is unconscionably expensive at the moment. Samsung’s 850 EVO provides a significant jump in overall performance for a higher price, but it’s not a straightforward upgrade. The SU900’s real-world results largely beat the 850 EVO’s, but the 850 EVO fared much better in our IOMeter synthetic tests. Additionally, the SU900’s MLC might bring some buyers more peace of mind over the 850 EVO’s less-durable TLC.
Supply conditions are forcing all these drives’ prices higher than we’d like, but that’s not the SU900’s fault. It carries a $100 price tag at Amazon. $0.39 per gigabyte is hard to complain about in today’s market for an MLC drive, not to mention that two-bit-per-cell flash is increasingly an endangered species in consumer SSDs.
Adata’s 3D MLC drive has almost uniformly improved upon its 3D TLC and planar MLC forebears. With its strong real-world performance and reasonable cost, the SU900 is a strong contender for a TR Recommended award. If it wasn’t for this drive’s shortcomings in our IOMeter sequential write test, it may have just gone home with the prize. Nonetheless, we are pleased with the drive’s speed, price, and warranty, and we wouldn’t hesitate to put an SU900 in our own machines.