Sign in with
Sign up | Sign in
Ten 60 GB SandForce-Based Boot Drives, Rounded-Up
By ,
1. The Great 60 GB SandForce SSD Round-Up

Hardly a month goes by that we don't see another SSD launch into an already-crowded market. Most of the time, the introduction centers around a SandForce controller, and there's a good reason for that. The company's technology is fast. It's also designed to work alongside lower-quality flash memory, if need be, driving down the costs of SSDs at a time when price is the primary barrier keeping enthusiasts from adopting solid-state storage.

More than one year has passed since the controller company launched its line-up of second-generation logic, and its solutions are still at the top of their game. OCZ was the first vendor with a SandForce-based desktop-oriented drive way back then. And within the last 30 days, Intel even threw its hat into the ring with its SSD 520, built on top of the very same controller, but complemented with the company's own cherry-picked NAND. Truly, SandForce deserves kudos for remaining a relevant force, even in the face of newer controllers from competing sources. 

Although the vendors selling drives centering on SandForce's hardware often sit at the top of our benchmark charts, there are, admittedly, a lot of them. On one hand, that's a good thing. Competition between many similar products pushes prices down as low as they can possibly go. But as you see in the picture below, consisting of only 60 GB SSDs, the number of choices can quickly get overwhelming. 

Are all of those drives exactly the same? Decidedly not. Although each SSD employs the same controller and boasts the same capacity, there is a limited number of knobs and dials that vendors can tweak to affect pricing and performance. And that's the idea behind today's round-up. We want to explore how one SandForce-based drive differs from the next, and hopefully end up with a better idea of the 60 GB drive you'd want to boot from in your next PC.

On deck, we have:

  • Adata S511 60 GB
  • Corsair Force 3 60 GB
  • Corsair Force GT 60 GB
  • Kingston SSDNow V200+ 60 GB
  • Intel SSD 520 60 GB
  • OCZ Agility 3 60 GB
  • OCZ Vertex 3 60 GB
  • OWC Mercury Electra 6G 60 GB
  • Patriot Pyro SE 60 GB
  • RunCore Pro V 60 GB


As we dig into these 10 SSDs, know that there is a performance hierarchy distinguishing drives based on SandForce's controller hardware. Generally, two SF-22xx-based SSDs of a given capacity perform almost identically, provided that they both employ the same NAND interface. Vendors do use different NAND, though, and that's why there is variation between drives centering on the same controller hardware.

Second-gen SandForce SSDs are available with different flavors of memory, and this is their order of performance, from highest to lowest.

  1. SandForce controller with Toggle DDR NAND (Mushkin Chronos Deluxe, Patriot Wildfire, OCZ Vertex 3 Max IOPS)
  2. SandForce controller with Synchronous ONFi NAND (OCZ Vertex 3, Corsair Force GT, Kingston HyperX)
  3. SandForce controller with Asynchronous ONFi NAND (OCZ Agility 3, Corsair Force 3, Mushkin Chronos, Patriot Pyro)

Right off the bat, we're able to sort our round-up into two groups: one that employs synchronous memory for better performance, and the other that uses asynchronous flash in order to save money.

Synchronous
Asynchronous
Adata S511 60 GBCorsair Force 3 60 GB
Corsair Force GT 60 GBKingston SSDNow V200+ 60 GB
Intel SSD 520 60 GBOCZ Agility 3 60 GB
OCZ Vertex 3 60 GBOWC Mercury Electra 6G 60 GB
Patriot Pyro SE 60 GB
RunCore Pro V 60 GB


How significant is the difference between the various memory interfaces? From there, can we distinguish between the drives in one group or the other? We’ll let the benchmarks do the talking.

2. Test Setup And Firmware Notes

In a departure from what we've done in the past, we're adding out-of-box performance back into test suite today. So long as TRIM and garbage collection do their job, the performance of a desktop-oriented SSD should stay close to that level. We'll also cover the other end of the spectrum, though, conveying the performance during reads and writes to a dirty drive. Performance will always fall between those two extremes.

Firmware notes:

Because this story was already in progress when Micron released its 0309 firmware update, we wanted to note that the patch primarily resolves a BSOD error related to a power-on SMART counter. According to Crucial's change log, there should be no effect on performance.

Test Hardware
Processor
Intel Core i5-2400 (Sandy Bridge), 32 nm, 3.1 GHz, LGA 1155, 6 MB Shared L3, Turbo Boost Enabled
Motherboard
Gigabyte GA-Z68X-UD7-B3
Memory
Kingston Hyper-X 8 GB (2 x 4 GB) DDR3-1333 @ DDR3-1333, 1.5 V
System Drive
OCZ Vertex 3 240 GB SATA 6Gb/s
Tested DrivesAdata S511 60 GB SATA 6Gb/s, Firmware: 3.3.2

Adata S511 60 GB SATA 6Gb/s, Firmware: 3.3.2

Corsair Force GT 3 60 GB SATA 6Gb/s Firmware: 1.3.3

Corsair Force 3 60 GB SATA 6Gb/s Firmware: 1.3.3

Crucial m4 64 GB SATA 6Gb/s, Firmware: 0009

Kingston SSDNow V200+ 60 GB, Firmware: 3.3.2

Intel SSD 520 60 GB SATA 6Gb/s, Firmware: -

OCZ Agility 3 60 GB SATA 6Gb/s, Firmware: 2.15

OCZ Vertex 3 60 GB SATA 6Gb/s, Firmware: 2.15

OWC Mercury Electra 6G SATA 6Gb/s, Firmware: 3.3.2

Patriot Pyro SE SATA 6Gb/s, Firmware: 3.3.2

RunCore Pro V 60 GB SATA 6Gb/s, Firmware: 3.3.2

Samsung 830 64 GB SATA 6Gb/s, Firmware CXM0
Graphics
Palit GeForce GTX 460 1 GB
Power Supply
Seasonic 760 W, 80 PLUS Gold
System Software and Drivers
Operating SystemWindows 7 x64 Ultimate
DirectX
DirectX 11
DriverGraphics: Nvidia 270.61
RST: 10.6.0.1002
Virtu: 1.1.101
Benchmarks
Iometer 1.1.0
# Workers = 1, 4 KB Random: LBA=16 GB, varying QDs, 128 KB Sequential
PCMark 7
Storage Suite
HD Tune Pro
v5.0, LBA= Full Span
3. 4 KB Random Performance

Random Read Performance (background info)

Examples include antivirus scans and typing in Word

From what we've seen in testing, it can be more difficult to generate consistent data using SandForce-based SSDs than many competing drives. In some benchmarks, we note up to 50 MB/s swings at low queue depths during successive runs. Yet, if we take the median results, all of the 60 GB models collected for this story perform fairly similarly, regardless of their NAND interface. The only exception is Intel’s SSD 520, which achieves a 50 MB/s lead over the competition at queue depths between two and four.

Meanwhile, the 64 GB Samsung 830 falls behind the SandForce-based pack. Crucial's m4 performs respectably, nearly matching the SSDs being rounded-up today. We see one anomaly at a queue depth of four, where the 64 GB m4 falls behind by roughly 90 MB/s.

Random Write Performance, Compressible

Examples include email, file compression, and Web browsing

In general, all 60 GB SandForce-based SSDs perform within a fairly tight range when we measure performance using compressible random writes. The only clear leaders are the Vertex 3 and SSD 520, but they only differentiate themselves at queue depths between four and eight.

The 64 GB Crucial m4 and Samsung 830 fall further behind in this test. We'd expect that, though, considering neither controller benefits from compression in the same way as SandForce's.

4. 128 KB Sequential Performance

Sequential Read Performance

Examples include file copying, transcoding, game level loading, some gameplay, watching video, and editing video

The Vertex 3's advantage over the SSD 520 that we noticed previously goes away when we switch to sequential read testing. Again, regardless of NAND interface, we see most of these drives fall within a very tight range.

Interestingly, Crucial’s 64 GB m4 keeps up with the pack at every queue depth. Meanwhile, Samsung’s 64 GB 830 shows some weakness at a queue depth of two after leading significantly with only one outstanding command. When you consider that most sequential operations occur at a queue depth of one, Samsung actually is the favorite in this test with a sequential read speed of 450 MB/s.

Sequential Write Performance

Examples include Application Installation, Document Backup

When it comes to writing compressible data sequentially, all of the SandForce-based drives throw up the exact same benchmark results, regardless of NAND interface or price.

5. Incompressible Sequential Write Performance: SandForce's Weakness

SandForce’s DuraWrite technology handles compressible data very efficiently, but there are real world cases where you you move incompressible information, whether you know it or not.

Random Write Performance, Incompressible

Examples include file compression and encryption

Writing incompressible data in a random manner is one of the worst things you can do to a SandForce-based SSD. The type of data, first of all, completely circumvents the controller's architectural advantage. Second, random access is inherently slower than sequential.

Consequently, Crucial's 64 GB m4 turns in the most impressive numbers at a queue depth of two or greater. At a queue depth of one, the SandForce-based drives featuring synchronous NAND enjoy a ~10 MB/s advantage over the competition. Meanwhile, the SandForce-based drives armed with asynchronous flash are roughly on par with the 64 GB Crucial m4 and Samsung 830.

Incompressible Sequential Write Performance

Examples include copying/creating multimedia, archive manipulation, encryption, some gameplay, and video recording

Even writing incompressible data sequentially isn't SandForce-friendly. Theoretically, this benchmark should yield a good idea of how various NAND interfaces behave. And indeed, you see drives clump together based on the type of memory they use (the differences almost assuredly aren't attributable to firmware optimizations). Within the clusters, there's only about 4 MB/s separating the models, and of course those numbers are indistinguishable in the real world.

Crucial’s 64 m4 makes for an interesting comparison point because it employs synchronous NAND as well. Its performance advantage suggests that Marvell's controller is simply more adept at writing sequential data once the benefits of compression are factored out SandForce's results. Samsung’s 64 GB 830 blitzes to the top of the chart with a sequential write speed of ~170 MB/s. That’s more than twice as fast as the SandForce-based SSDs with asynchronous flash and 90% faster than those using synchronous NAND.

6. PCMark 7 And Power Consumption

PCMark 7

Samsung’s 64 GB 830 jumps to the top of our PCMark 7 chart, followed by Crucial's 64 GB m4.

The SandForce-based drives don't do as well since this test consists of both compressible and incompressible information. The resulting workload isn't a best-case reflection of what its controller hardware can do. We do, however, see a very clear break between the drives with synchronous memory and the less expensive models equipped with asynchronous flash. On average, the faster SandForce-based SSDs deliver about 8.5% more performance than the more value-oriented models.

Power Consumption

Because SSDs are so fast, they sit idle most of the time. In our almost 30-minute virus scan, the SSD is only busy for 281 seconds. As a result, idle power consumption is the most important figure to consider in a desktop environment.

Samsung's 64 GB 830 does exceptionally well here. Even though it employs a beefy triple-core ARM-based controller, it consumes slightly less power than Crucial's m4 and its dual-core ARM-based Marvell controller.

When it comes to the SandForce-based drives, our results are less consistent than we might have expected. Though, with all of these sub-1 W measurements, we're not certain any of these SSDs can claim an advantage.

Samsung continues to excel, even under more of a load. Its 64 GB 830 uses less power when it's active than Intel's SSD 520 at idle. Crucial's m4 takes a second-place finish, which is interesting in light of its less impressive finish in the idle chart. 

There's again some variation between the SandForce-based drives, with no discernible trend in the finishing order. But we're still hesitant to flag a victor, given a fairly narrow range.

7. Endurance Testing

SSD vendors selling SandForce-based drives are incredibly enthusiastic about differentiating their offerings. There are three aspects of solid-state storage that affect performance: the controller, the NAND, and the firmware. We all know that these drives center on the same firmware. We've seen that the flash does have some affect on performance, though two drives with the same configuration are pretty much comparable. So, they try to sell us on custom firmware with home-brewed optimizations not offered by other vendors.

Can we create of list of what those tweaks entail? Unfortunately not. No SSD vendor has ever gotten specific with us about what its "golden" or "purely in-house" firmware includes that other vendors don't have.

What we do know is that the basic core of SandForce’s compression technology cannot be altered. We tested for this in Intel SSD 520 Review: Taking Back The High-End With SandForce by measuring endurance by writing highly compressible data. What we found were close to identical values for write amplification. When write amplification is similar, then we know that two drives (in this case, the oldest and newest SandForce-based SSDs) are benefiting from the same level of compression.

128 KB Compressible Sequential Write
1 Hour, QD=1
Intel SSD 520
60 GB
OCZ Vertex 3
60 GB
Host Writes
1258 GB
1301 GB
NAND Writes176 GB
182 GB
Write Amplification
0.13990
.139892


In the time between publishing our SSD 520 review and now, we've seen similar results from all of the 60 GB SF-22xx-based SSDs in our lab, suggesting that every vendor using SandForce's technology enjoys the same degree of compression, which most influentially affects the performance of these drives.

8. Exploring The Performance Of A Full SandForce-Based SSD

One of the reasons we brought back fresh out-of-box testing in Iometer was to free up time to explore other aspects of performance. For this round-up, we specifically want to investigate the read performance of a full drive. After all, a 60 GB drive isn't very large, so it's very likely that you'll be dealing with this exact scenario on and off.

In theory, read speeds shouldn't be significantly impacted by the drive's free space. But there is a noticeable drop-off on the lower-capacity SF-22xx-based SSDs. Compression is one reason for this. Compressed data has to be decompressed, which requires sufficient free space. SandForce's garbage collection mechanism is a second explanation. Because most of its clean-up happens in the foreground, decompression requires a rotation of scratch space so that the NAND wears out evenly. This is known as wear leveling.

Both factors penalize the 60 GB drives we're rounding up more than the larger models as a result of limited NAND bandwidth, a side-effect of leaning on fewer NAND dies. We're using HD Tune to illustrate.

The image above shows how read performance varies. After using Acronis True Image Home to clone our system drive, we set HD Tune to read the entire LBA space sequentially. As you can see, reading back information written to our 60 GB Agility 3 falls between 150-200 MB/s. Once we move on to free space, performance accelerates to 450 MB/s. That's a big difference.

But what happens when we get more specific and narrow our focus to incompressible data? After adding about 6 GB of MP3s and H.264-encoded video, we can clearly discern performance at its worst on a SandForce-based drive. Read performance drops to about 150 MB/s. That's why you see performance bounce up and down when we only have Windows and a few apps installed: some of that data is compressible, and some isn't.

That's not the end of the story, though. We repeated this same tests on all of the drives in our round-up and found that the results only apply to 60 GB SF-22xx-based SSDs accessing asynchronous NAND. Let's take Intel's SSD 520 for a spin to compare the performance of a drive with higher-performing synchronous flash.

We already established (using early Iometer testing) that reading back free space is similar regardless of a drive's NAND interface. Yet, comparing the SSD 520 to OCZ's Agility 3, we see read performance in the space occupied by Windows and our application data falls between 250-300 MB/s. In a worst-case scenario, incompressible data is read back at 250 MB/s. Sure, that's low compared to the 450 MB/s we see otherwise. But it's also 100 MB/s faster than the 60 GB Agility 3 and its asynchronous NAND-based competition.

Now, to put it all in context. Check out the results from our 64 GB m4 in the chart below. Crucial mostly employs background garbage collection, and it doesn't lean on compression for its commendable performance. As a result, read performance is largely unaffected by fill state.

9. Performance Is Defined By Flash

SandForce's impact on the SSD landscape is undeniably significant. In focusing its efforts on the most influential component of solid-state storage performance and reliability, the company makes involvement in this growing segment more accessible to a number of vendors able to combine the controller and their own sourced NAND into a fairly reference-like package.

The firmware is, for the most part, completely furnished. And while some vendors claim exclusive optimizations, it's pretty clear that the performance impact of those tweaks is minimal. As a result, though, we're enjoying a lot more competition than if SandForce had never come along. Prices are consequently driven down, and enthusiasts win.

With all ten of these SSDs leveraging the same controller hardware, flash memory becomes the key element in defining a given drive's performance. That's not to say all of these drives sport similar reliability. Taking shortcuts by using cheaper power components, for instance, can negatively affect one brand's models more so than a competing vendor leveraging higher-quality parts. But when it comes to the performance data, synchronous NAND, for example, means OCZ's Vertex 3 behaves an awful lot like Corsair's Force GT. 

And although vendors are free to make their own firmware optimizations where they make sense, any tweaks that are being made are outweighed by the firmware elements these drives share in common. According to SandForce, it doesn’t matter if you own a PC or Mac. The relationship between hardware and firmware (and how it affects features like TRIM and garbage collection) functions identically from one drive to the next.

Lacking any other way to really quantify the differences between second-gen SandForce drives at 60 GB, we're left to consider variables like NAND quality. Right now, Intel is the only vendor making noise about the fact that it skims the very best flash die from IMFT for use in its SSD 520, and apparently that's enough to warrant the highest price per gigabyte in our round-up. Can you really put a price on the integrity of your data, though? Intel backs its claim with the only five-year warranty, so perhaps it's onto something. Using reliability as a differentiator, the company shows us that maybe we shouldn't always be looking for the cheapest SSD, but rather the most trustworthy one.


NAND Interface
Market Price
Price per GB
Warranty
Adata S511 60 GB
Synchronous$110
$1.83
3 years
Corsair Force 3 60 GB
Asynchronous$95
$1.58
3 years
Corsair Force GT 60 GB
Synchronous$107
$1.78
3 years
Intel SSD 520 60 GBSynchronous$135
$2.25
5 years
Kingston SSDNow 200+ 60 GB
Asynchronous$110
$1.83
3 years
OCZ Agility 3 60 GB
Asynchronous$90
$1.50
3 years
OCZ Vertex 3 60 GB
Synchronous$98
$1.63
3 years
OWC Mercury Electra 6G 60 GBAsynchronous$120
$2.00
3 years
Patriot Pyro SE 60 GB
Synchronous$113
$1.88
3 years
RunCore Pro V 60 GB
Synchronous$123$2.05
3 years


Should you necessarily fear a SandForce-based SSD that leverages cheaper NAND, then? Not at all, actually. Drives based on the company's controller are some of the most affordable performance-oriented solutions specifically because SandForce designed its controller to utilize lower-quality NAND dies without compromising reliability, even in the face of less endurance.

So, what's our lesson at the end of the day? When it comes to picking a 60 GB SandForce-based boot drive, NAND type is the biggest determinant of performance (although these drives are all significantly faster than anything with magnetic disks). And if you're worried about reliability, that difficult-to-quantify X factor, a vendor is only as good as its reputation. Some brands do better than others when it comes to supporting their products, so perhaps an exploration of rebate fulfillment, phone support, and RMA processing is in order next?