Budget solid-state drives make up a big chunk of the storage market, but cheaper products mean smaller margins. As a result, drive makers are constantly on the prowl for ways to cut costs and eke out a little more profit. One of the most straightforward ways to trim the fat is simply to store a third bit per flash cell, making triple-level-cell, or TLC, flash.
Samsung first introduced the world to triple-level cell NAND flash in 2012 with its 840 Series solid-state drives. TLC’s debut was met with skepticism from enthusiasts, since many folks expected the increased data density would lead to poorer performance and endurance than single- and multi-level-cell-based drives. As it turned out, TLC flash offered solid performance for the mass market, and we proved that concerns about endurance were largely unfounded.
After Samsung’s widespread success with the new technology, other drive makers started producing TLC offerings of their own. This year, both OCZ and Crucial brought new low-cost SSDs to market. OCZ’s Trion 100 and Crucial’s BX200 SSDs occupy the lowest-end slots in these firms’ solid-state storage lineups. These two drives represent their makers’ first forays into TLC flash, and both companies happen to offer a 480GB model. Since these SSDs are so similar on so many levels, we couldn’t pass up the opportunity to perform a shootout-style review. Let’s get up close and personal with the contenders.
The Trion 100 slots in beneath our budget-favorite Arc 100 as OCZ’s cheapest solid-state drive. Like the Arc 100, the Trion 100 is built on Toshiba’s 128Gb A19 19-nm NAND. This time around, the A19 flash is TLC, as opposed to the Arc 100’s MLC. The Trion 100 is also powered by a Toshiba-branded TC58 controller instead of the Arc 100’s Barefoot 3. The NAND in the 480GB model is distributed into four packages on a single-sided board.
OCZ Trion 100 | ||||
Capacity | Max sequential (MB/s) | Max Random (IOps) | ||
Read | Write | Read | Write | |
120GB | 550 | 450 | 79k | 25k |
240GB | 550 | 520 | 90k | 43k |
480GB | 550 | 530 | 90k | 54k |
960GB | 550 | 530 | 90k | 64k |
OCZ also offers 120GB, 240GB, and 960GB versions of the Trion 100. The two lower-capacity versions have lower sequential and random write specs, likely because they lack enough NAND dies to fully saturate the controller’s channels.
Toshiba hasn’t released much technical detail about the controller in the Trion drives, but we know this chip includes proprietary error-correction technology and a pseudo-SLC caching system (like Samsung’s TurboWrite or SanDisk’s nCache).
In Crucial’s corner, we have the BX200 480GB. Unlike the Trion series, Crucial hasn’t introduced a new product line and naming convention for its TLC drive. Instead, the BX200 is being presented as a successor to last year’s well-received BX100.
Crucial BX200 | ||||
Capacity | Max sequential (MB/s) | Max Random (IOps) | ||
Read | Write | Read | Write | |
240GB | 540 | 490 | 66k | 78k |
480GB | 540 | 490 | 66k | 78k |
960GB | 540 | 490 | 66k | 78k |
Crucial’s BX200 lineup is rounded out with 240GB and 960GB variants. Crucial claims the same performance numbers across the three versions, unlike OCZ’s specifications for the Trion 100. We’d expect the 240GB version to fall a bit short of the others, but we don’t have one on hand to confirm that suspicion.
The now-discontinued BX100 was built with Micron’s 16-nm MLC flash and Silicon Motion’s SM2246EN controller, but the BX200 uses Micron’s new 16-nm 128Gb TLC flash and a newer Silicon Motion SM2256 controller. The BX200’s NAND is bundled into eight packages alongside the controller and memory. This newcomer has learned a couple new tricks that the outgoing BX100 doesn’t know: a pseudo-SLC caching scheme and a new TLC-targeted error-correction scheme that Silicon Motion calls “NANDXtend.”
We’ve already noted that both the Trion and the BX200 use TLC NAND, but we can get even more specific. Both drives use 128 Gbit TLC in a 32-die configuration. That’s a sufficient number of dies to saturate each controller’s I/O channels, so performance differences we uncover between the drives should come down to the controllers’ capabilites and the quality of the NAND itself.
Both drives come with a three-year warranty, but their endurance specs differ substantially. The Trion 100 claims to be good for 120TB of writes, while the BX200 offers a more cautious 72TB. The first point goes to OCZ, but there are a lot of points to be tallied yet. Let’s run the drives through our benchmark suite and see how they do.
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.
Both drives come out of the gate with solid sequential read numbers. They even beat some bigger, fancier drives at a queue depth of one, and the BX200 hangs with the pack at QD4.
But when it comes to sequential write performance? Oh, dear. These are the slowest-writing drives we’ve seen in a very long time. The BX200 falls especially far behind, losing to even the Intel X25-M G2, a SATA 3Gbps drive that was released in 2009. Yikes. Both drives fall far short of the high bar set by Samsung’s TLC in the 850 EVO series—even the 250GB EVO. To be fair, the 850 EVOs are packing 3D V-NAND, not planar NAND. The older, MLC-based Arc 100 and BX100 look very good by comparison.
Next, we’ll turn our attention to performance with 4KB random I/O. The tests below are based on the median of three consecutive three-minute runs. SSDs typically deliver consistent sequential and random read performance over that period, but random write speeds worsen as the drive’s overprovisioned area is consumed by incoming writes. We’ve reported average response times rather than raw throughput, which we think makes sense in the context of system responsiveness.
The sequential results are reversed here. The two drives post relatively poor random read response times, but very respectable random write results. On the read side, the BX200 gets beaten again by the six-year-old X25-M G2. When we switch to writes, though, the BX200 manages to beat the BX100. Not too shabby, since we’d expect the BX100’s MLC to give it a clear advantage over the BX200. The difference is likely due to the new SLC-esque caching ability that the BX100’s controller lacks.
As we noted, the preceding tests are based on the median of three consecutive three-minute runs. SSDs typically deliver consistent sequential and random-read performance over that period, but random-write speeds worsen as the drive’s overprovisioned area is consumed by incoming writes. We explore that decline in the next set of tests.
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, which 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.
Like all SSDs, these drives exhibit higher random write rates before their overprovisioning is consumed, but their peak rates during that period are an obvious cut below the rest.
To show the data in a slightly different light, we’ve graphed the peak random-write rate and the average, steady-state speed over the last minute of the test.
Our two drives both peak and stabilize at lower IOps than the other drives in our data set, save for the ancient X25-M. There’s no getting around the fact that these two drives’ write speeds are mediocre at best. The Trion 100 consistently comes out ahead of the BX200, but it’s the difference between distressingly low and, well, a bit more distressingly low. Meanwhile, the Arc 100 boasts steady-state random write speeds an order of magnitude faster than the Trion 100 and BX200. This marks the third rig we’ve tested the Arc 100 on, and it never fails to remind us why we gave it a TR recommended award the first time.
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 max 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.
These aren’t NVMe drives, so it’s no surprise they don’t scale appreciably into the higher queue depths. What’s unexpected is that these drives hardly seem to scale at all. Performance at a queue depth of eight is just barely higher than at a depth of four. The Trion 100 and BX200 look about as bad here as they do in our sustained write tests, never breaking 5000 IOps. As a point of comparison, the 850 EVO and BX100 manage to climb into the 7500 IOps territory. As for the Arc 100? well, just take a look at the next set of graphs.
Below, we use the same data to plot the Trion 100 and BX200 against the Arc 100 and BX100. Click to toggle between read, write, and total IOps.
It’s a slaughter. The Arc 100 blows away the two TLC drives. These results also highlight an uncomfortable truth for Crucial. The BX200 can’t put up numbers even half as good as its predecessor.
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 | |
Media | 459 | 21.4MB | 9.58GB | 0.8% |
Work | 84,652 | 48.0KB | 3.87GB | 59% |
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 results of our media set tests mirror what we saw in our IOMeter synthetics. The Trion 100 and BX200 more or less manage to keep up with the pack in read speeds, but they fall far behind in our write scenarios. The good news is that neither drive got beaten out by the decrepit X25-M this time around. It’s a small victory, but we’ll take what we can get at this point.
function trset8() { document.getElementById('work-1t').src='/r.x/2015_11_22_Trion_100_BX200/robo-read-1t-work.png'; document.getElementById('work-8t').src='/r.x/2015_11_22_Trion_100_BX200/robo-read-8t-work.png'; } function trset9() { document.getElementById('work-1t').src='/r.x/2015_11_22_Trion_100_BX200/robo-write-1t-work.png';4 document.getElementById('work-8t').src='/r.x/2015_11_22_Trion_100_BX200/robo-write-8t-work.png'; } function trset10() { document.getElementById('work-1t').src='/r.x/2015_11_22_Trion_100_BX200/robo-copy-1t-work.png'; document.getElementById('work-8t').src='/r.x/2015_11_22_Trion_100_BX200/robo-copy-8t-work.png'; }
In the work set, the BX200 finally starts showing signs of life. Perhaps the smaller file sizes here are more suited to leveraging this drive’s SLC-like caching mechanism. Whatever the reason, we’re glad to see that the BX200 isn’t stuck at the bottom of the barrel.
We were holding out hope that these drives’ writes would fare better in the “real-world” I/O scenarios that constitute RoboBench, but that didn’t turn out to be the case. They stayed at the tail end of the pack for the most part. But their read speeds are respectable, so it’s not all bad news.
Boot times
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’ve often seen that even the most lethargic SSDs can be useful as system drives. They tend to boot the OS and load applications almost as quickly as the fastest drives money can by. Let’s hope that’s true for the Trion and the BX200, as they’re running out of ways to redeem themselves.
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.
So far, so good. Neither drive is going to break any records, but at least they aren’t significantly slower to boot Windows than the rest of the crowd. Both drives best the notoriously slow-booting Intel 750 Series PCIe SSD, and that’s enough to satisfy us here.
Load times
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 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.
As usual, differences in the application load times of the various drives are minimal. We’re talking about a few tenths of a second separating the slowest drives from the fastest. No point declaring winners and losers here. It’s a relief to see our two budget drives keep up with the rest of the pack. Let’s take a look at in-game performance next.
Like the application load times, the game load times are just fine. You won’t notice any difference between a BX200 and an NVMe 950 Pro when it comes to firing up your favorite games.
These boot and load results give us the good news we’ve been hoping for. Despite an underwhelming showing in IOMeter and RoboBench, the Trion 100 and BX200 redeem themselves a little bit in our OS and application load tests.
That’s it for performance testing. Read on for a breakdown of our hardware and test methods.
Test notes and methods
Here’s are the essential details for all the drives we tested:
Interface | Flash controller | NAND | |
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 |
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 |
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 XP941 256GB | PCIe Gen2 x4 | Samsung S4LN053X01 | 19-nm Samsung MLC |
Samsung SM951 512GB | PCIe Gen3 x4 | Samsung S4LN058A01X01 | 16-nm Samsung MLC |
Samsung 850 Pro 500GB | SATA 6Gbps | Samsung MEX | 32-layer Samsung 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 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 |
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 and 950 Pro requires 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, if you missed its turn on the runway on the first page:
Processor | Intel Core i5-4690K 3.5GHz |
Motherboard | Asus Z97-Pro |
Firmware | 2601 |
Platform hub | Intel Z97 |
Platform drivers | Chipset: 10.0.0.13 RST: 13.2.4.1000 |
Memory size | 16GB (2 DIMMs) |
Memory type | Adata XPG V3 DDR3 at 1600 MT/s |
Memory timings | 11-11-11-28-1T |
Audio | Realtek ALC1150 with 6.0.1.7344 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 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 |
Power supply | Corsair AX650 650W |
Operating system | Windows 8.1 Pro x64 |
Thanks to Asus for providing the systems’ motherboards, Intel for the CPUs, Adata for the memory, and 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.
Conclusions
We did set this up as a head-to-head matchup, so we’re going to have to declare a winner. Though neither drive blew us away, the victory goes to the OCZ Trion 100. Both drives struggled to match the speeds of much older hardware (like Intel’s X25-M) in some of our tests, but the Trion 100 put up a better fight. To be fair, these drives aren’t intended to blow your socks off like the high-end drives we typically test. They’re positioned more for buyers who are just getting around to making the switch from mechanical storage or for folks who want a higher-capacity SSD without taking out a second mortgage. These drives would make great replacements for systems still booting off hard disk drives. Their boot and load-time performance is up there with the best, and they handle read-heavy workloads with aplomb.
I believe one problem with the BX200—aside from its lackluster speeds—is simply its name. OCZ played it smart by creating a new and distinct product line for its TLC drive and leaving the Arc 100 in its lineup. The Trion name didn’t create any expectations for me. Contrast that with Crucial’s approach. The company opted to bill the BX200 as both the successor and replacement for the BX100. The natural expectation is for the newer drive to be an evolution and improvement of the one that came before. Alas, the BX200 is anything but. Its MLC-based predecessor is the far better drive.
Newegg currently sells the 480GB Trion 100 for $160, and the 480GB BX200 for $150. The Trion 100 is certainly worth the $10 premium. But if I had $160 to buy an SSD, I’d put $10 back in my pocket and buy Samsung’s 500GB 850 EVO for $150. Right now, the Samsung drive even comes with a free copy of the most recent Assassin’s Creed release to sweeten the deal.
The 850 EVO’s current price demonstrates an issue instrinsic to the budget segment: the ever-changing landscape of SSD deals almost always makes some higher-end drives available for lower prices than budget drives at their full prices. To really get a good value out of “value-oriented” drives, you have to wait for a steep discount to go into effect for your drive of choice.
To sum up, although neither of these drives lead the pack, they have a place in the market. Their speeds may seem sluggish compared to our stable of hot performers, but they still beat the pants off spinning platters in most cases. Once these drives have been around long enough to be included in the regular SSD sales frenzy, their discounted per-gigabyte cost could make them solid options for low-end PC builds.