Date: March 2, 2011
Author(s): Robert Tanner
There couldn’t be a better time than the present to purchase an SSD, and on the same token, it’s also a strange time. Performance drives cost the same as budget drives, and a perfect example of this is Corsair’s Force F90. It’s priced-right, offers incredible performance, and makes perfect use of its SandForce SF-1200 controller.
Whenever I think about solid-state drives I am reminded of a famous quote I heard, “There’s a basic principle about consumer electronics: it gets more powerful all the time and it gets cheaper all the time.” – Trip Hawkins. Okay, maybe that quote was actually heard in a game of Civilization V, but still its validity certainly couldn’t be any truer.
When SandForce SF-1200 based solid-state drives launched in 2010 they started off at over $3 per gigabyte. The Crucial C300 launched closer to $4 per GB. Heck, when Intel launched its first 160GB SSD in 2008 it was priced at over $6 a gigabyte!
By comparison, for the start of 2011 prices are just below $2 per GB, although for consumers wishing to make every penny count, promotions can sharply drop this figure to around $1.50. We have already observed one major rebate promotion (expired) that pegged a SandForce drive at an eye-popping $1.10 per GB, and I expect that this is just the start of the trend. Consumers have very low NAND market prices to thank for such a rapid decrease in SSD costs, and with foundries starting up <30nm NAND production and new, even higher performance SSDs expected to launch this year, it is a safe bet that we will see high-performance drives reach $1 per GB prices this year.
With SSD prices compacted together, it is possible to buy SandForce drives for less than the cost of the lowest common denominator budget SSDs, so consumers should absolutely not settle for less than the current best performance SSDs when choosing what model to buy. This is probably why OCZ quietly canceled plans for its Vertex Plus, as there simply wasn’t room in the market for a new budget alternative. Flat out we can say either the SandForce SF-1200 or the Micron C300 drives would get our recommendation, although SandForce drives can be found for marginally less and perform just as admirably despite their SATA 3Gbps interface.
We are going to take a look at one such very affordable SSD. Its 90GB capacity allows it to undercut 120GB model prices while still offering a more than satisfactory amount of capacity for users that wish to install both programs alongside the OS onto the drive. The Corsair Force series solid-sate drives are, of course, SandForce-powered, making them an excellent choice for any potential laptop or desktop upgrade.
We would like to pause and applaud Corsair for not obfuscating which Force products use 25nm NAND flash, as users can determine if the drive they are looking at is using 25nm NAND by whether or not an “A” is present on the tail end of the model name, for example F80-A, or F115GB2-BRKT-A (the latter also denoting which drives include a 3.5″ bay adapter).
While leaving out much of the backstory and technical explanation, an unfortunate fact is that as NAND shrinks its lifespan also shrinks with it. While we are very willing to stake our own money (and data) on a 25nm flash Corsair SSD, we can understand consumers’ natural preference for 34nm NAND models where possible. As we will illustrate on our last page however, even with the lower write/erase cycles, the lifespan of a 25nm NAND drive should last well over five years in most situations. And according to the specification, NAND memory itself is only guaranteed to hold a charge for ten years regardless.
220 / 180
64, 128, 256, 512
|Performance 3 Series|
410 / 210
64, 128, 256
285 / 275
40, 60, 80, 90, 115, 120, 160, 180, 240
250 / 170
270 / 195
32, 64, 128, 256
As old stock can still be found online it is important to note that the original “P” Performance series have been replaced by the P3 brand, and both the Reactor and Extreme models have been discontinued for the obvious reasons. Given the price disparity between all these models we can easily recommend focusing on the Force series, which often are the lowest cost and highest performing option out of all these contenders.
Only the Nova series slots in at $10 less, but that $10 will buy you a much higher performing SSD making it a straight-forward choice. Although the 115GB Force model offers a slightly better price per GB ratio, the 90GB model we are reviewing is close at $189 as of this writing.
As with other models, Corsair uses a plastic clamshell package to protect the drive during shipment. Behind the clamshell is a solid steel bracket that not only gives the drive armor plating protection during shipment, but serves double duty as the 2.5″ to 3.5″ bay adapter. Out of the myriad of adapter kit designs we have seen we prefer the simplicity and durability of this particular adapter the most. As for the SSD itself the housing is black brushed aluminum and is solid, but light enough to be mounted in a variety of situations if needed.
At Techgage, we strive to make sure our results are as accurate and real-world applicable as possible. We list most of the steps and processes involved in setting up and conducting our benchmarking process below, but in the interests of brevity we can’t mention every last detail. If there is any pertinent information that we’ve inadvertently omitted or you have any thoughts, suggestions, or critiques, then please feel free to email us or post directly in our forums. This site exists for readers like you and we value your input.
The table below lists the hardware used in our current storage-testing machine, which remains unchanged throughout all of our testing, with the obvious exception of the storage device. Each drive used for the sake of comparison is also listed here.
Techgage Hard Drive Test System
Intel Core 2 Quad Q6600 – 2.4GHz Quad-Core
4GB Corsair 800MHz CAS 4
Foxconn 8800 GTS 320MB
Corsair Force F90 90GB
Corsair Nova V128 128GB
Intel X25-M G1 80GB
Kingston SSDNow V Series 40GB
Kingston SSDNow V Series 128GB
Kingston SSDNow V+ Series 128GB
OCZ RevoDrive 120GB PCI Express
OCZ Summit 60GB
OCZ Vertex Turbo 120GB
OCZ Vertex 2 100GB
Seagate Barracuda 7200.10 320GB
Western Digital SiliconEdge Blue 256GB
PC Power & Cooling Quad Silencer 750W
Arctic Freezer 7 Pro
Windows 7 Ultimate 64-bit
Our Windows 7 Desktop for SSD Testing
When preparing our SSD testbed for testing we follow these guidelines:
All solid-state drives start in a factory fresh or HDDErase fresh state prior to testing. Windows 7 is manually installed and then SYSmark 2007 Preview is installed. Due to the nature of SYSmark, Windows 7 must be reinstalled upon completion of testing in order for many programs to function normally and benchmark consistently. As not all SSDs support TRIM technology, this process is important and ensures all of the drives are in a “dirtied” state before we collect the benchmark results.
For the time-being, cloned test images are not used as these can result in non-aligned partitions, which if it occurs will result in degraded SSD performance. Just as with Windows XP’s default sector offset causing degraded SSD performance, non-intelligent cloning software can have the same effect.
For testing, we ran all tests five times, dropping the highest and lowest results to finally average the middle three. And who said that college statistics class wouldn’t prove useful? If any anomalous results were seen, the test was run again. Given the complexities of modern computers, and especially today’s operating systems, we feel this provides the most accurate results possible.
Finally, we are seeking to constantly improve and expand upon our SSD testing methodology. We are activity seeking real-world workload scenarios that are bottlenecked by hard drives, so if you have any suggestions whatsoever or there is a program you would like to see included in our SSD content, then please drop by our forums and let us know! We are always looking to expand our SSD benchmarks and provide more useful and real-world results, and not just synthetic numbers.
There are few PC enthusiasts who are unfamiliar with the name “Futuremark”, as the Finland-based developer has been producing quality benchmarks to help us gauge our computer’s worth for years. Originally known as Madonion, Futuremark has expanded its focus to go beyond its bread and butter, graphics and gaming, and tackle other areas, such as full system performance. That’s where PCMark comes into play.
The company’s most recent addition to the PCMark family is Vantage. For most users, a full suite would be run, but because we’re focused on storage performance only, we instead run only the storage-specific tests. Fortunately, Futuremark makes this easy for us to do as it has split up the entire suite into seven separate sub-tests, one being the aptly named “HDD Suite”.
PCMark’s HDD Suite may look simple on the surface, but it’s actually quite exhaustive. While the benchmark does deliver a simple “overall” result, it actually tests I/O performance based on a variety of scenarios, from adding music to Windows Media Player, to loading applications in succession, to editing video, to running a malware scanner, and more. It even includes metrics to evaluate a simulated Windows Vista boot time, so Futuremark has done a fine job of combining many useful scenarios into a single button press.
With that over with, onto the results!
The overall PCMark score is fairly spectacular, especially given this is not the first SandForce SF-1200 based drive in our graphs. The only cause for such an improvement would be several iterations of firmware updates, as the other SandForce drives in our graphs had used the initial release firmware. It is nice to see such an appreciable jump in performance as the firmware matures.
Quixotically, once we break down the single total score, the Corsair 90GB Force doesn’t take a similar lead in the HDD Suite score, illustrating one problem with relying on a single overarching number. Instead the Corsair 90GB Force managed its impressive overall score thanks to a huge performance jump in the Communications subtests , along with an additional increase in the Gaming and Productivity subtests (neither are listed above).
All in all, Futuremark’s PCMark Vantage suite provides a good bellwether for system responsiveness, and in traditional SSD fashion the Corsair Force series drives have no trouble delivering on exactly that. If you are new to SSDs in general we recommending paying attention to the lone mechanical hard drive in our tests, as performance jumps like the above are typical when upgrading to a good quality solid-state drive.
Originally developed by Intel, and since given to the open-source community, Iometer (pronounced “eyeawmeter”, like thermometer) is one of the best storage-testing applications available, for a couple of reasons. The first, and primary, is that it’s completely customizable, and if you have a specific workload you need to hit a drive with, you can easily accomplish it here. Also, the program delivers results in IOPS (input/output operations per second), a common metric used in enterprise and server environments.
The level of customization cannot be understated. Aside from choosing the obvious figures, like chunk sizes, you can choose the percentage of the time that each respective chunk size will be used in a given test. You can also alter the percentages for read and write, and also how often either the reads or writes will be random (as opposed to sequential). I’m just touching the surface here, but what’s most important is that we’re able to deliver a consistent test on all of our drives, which increases the accuracy in our results.
Because of the level of control Iometer offers, we’ve created profiles for three of the most popular workloads out there: Database, File Server and Workstation. Database uses chunk sizes of 8KB, with 67% read, along with 100% random coverage. File Server is the more robust of the group, as it features chunk sizes ranging from 512B to 64KB, in varying levels of access, but again with 100% random coverage. Lastly, Workstation focuses on 8KB chunks with 80% read and 80% random coverage.
Because these profiles aren’t easily found on the Web, with the same being said about the exact structure of each, we’re hosting the software here for those who want to benchmark their own drives with the exact same profiles we use. That ZIP archive (~3.5MB) includes the application and the three profiles in an .icf file.
Iometer is one application that can replicate the most brutal conditions for solid-state drives, although for a traditional hard disk such scenarios would be simply dreadful. These scenarios in particular are why vast arrays of 15,000RPM SCSI drives are utilized in high and especially highly random I/O applications.
When having to seek to random data a solid-state drive has an inherent advantage. It can look up a read request in as little as 0.1 of a millisecond, as opposed to 7ms for a 10,000RPM drive or 12-14ms for most consumer level 7,200RPM hard drives. Compared to any kind of mechanical hard disk, these near instantaneous access times are what give SSDs most of their mojo, and why a difference in system responsiveness between one a system with an SSD and one without can by physically noticed just through normal use.
Iometer testing showcases which of these SSD controllers are the most capable at servicing numerous small reads and/or write requests concurrently. Think of the SSD controller as the drive’s own CPU, as that is basically what it does; even hard disk drives have a controller. What this test focuses on is testing how much it takes to max out and saturate the controller in read/write requests.
As these Iometer tests are designed to max out the SSD we are left with some extremely large variances in results between our contenders, but for those looking for a quick idea to put them in a context: anything around the old Vertex Turbo drive or higher is more than sufficient. Anything higher, while impressive, only translates into higher read/write bandwidth. Which is the same thing as saying while the Bugatti Veyron can exceed 250mph, anything above 80mph doesn’t mean a whole lot unless you have a dedicated racing strip to use it on.
The same is true with SSDs, we are reaching a point where higher IOPS is great to have, but unless users have a dedicated workload that actually requires it, anything extra is superfluous for typical enthusiast desktop use. All that being said, it is still nice to have and as expected, the SandForce SF-1200 controller in the Corsair Force SSD has power aplenty under its hood.
As the name might hint, AS SSD is a nifty little program written exclusively for SSDs. It can be run on mechanical hard drives, but be warned that what should take minutes will take over an hour to benchmark! This handy little tool provides several read/write tests at important file sizes, but also includes a benchmark to simulate the transfer of three types of large files.
We selected this program for its precision, ability to generate large file sizes on the fly, and that it is written to bypass Windows 7’s automatic caching system. The tool does not bypass any onboard cache.
AS SSD is an application that was written specifically for solid-state drives as an easy tool for any computer user to download and run for some quick, comparable results. We utilize it for its appropriate focus on small file operations and because it bypasses Windows 7’s automatic caching system in the process.
It is interesting to note that SandForce drives don’t fare particularly well in this program, and we finally have an answer as to why. It appears to be because of the type of data being written isn’t compressible. SandForce drives are unique in how they function; they attain their stellar performance through data compression. However if the data being written is already highly compressed (such as an MP3 or JPEG ) then it will offer lower performance. This is why the OCZ Revo, which is unarguably fast enough to exceed the SATA 3Gbps interface and had to be put on a PCIe card, had such trouble with the write portions of the tests. As such we can say these would be the worst-case scenario in terms of possible performance.
However, as most users don’t plan to utilize their SSD as an expensive file storage drive we don’t think these results should be any sort of deterrent, especially given both the read and write performance is still in another universe compared to mechanical drives. The worst-case scenario would be few and far between if utilized as an OS and Program Files drive.
Synthetic benchmarks have typically been favored for performance testing, but the results they provide can be fairly abstract, and the methods they use to assign their scores can be dubious at times. By contrast, real-world application benchmarks provide performance metrics that apply directly to real-world usage, and we endeavor to apply both in our performance comparisons.
SYSmark 2007 Preview from BAPCO is a special case, because its synthetic scores are derived from tests in real-world applications. However, we still believe that synthetic benchmarking scores are best used to directly compare the performance of one piece of hardware to another, and not for developing an impression of real-world performance expectations. SYSmark is more useful than most synthetic benchmarking programs in our opinion, because its tests emulate tasks that people actually perform, in actual software programs that they are likely to use.
The benchmark is hands-free, using scripts to execute all of the real-world scenarios identically, such as video editing in Sony Vegas and image manipulation in Adobe Photoshop. At the conclusion of the suite of tests, five scores are delivered: an E-learning score, a Video Creation score, a Productivity score, and a 3D Performance score, as well as an aggregated ‘Overall’ score. These scores can still be fairly abstract, and are most useful for direct comparisons between test systems.
A quick note on methodology: SYSmark 2007 requires a clean install of Windows 7 64-bit to run optimally. Before any testing is conducted, the hard drive is first wiped clean, and then a fresh Windows installation is conducted, then lastly, the necessary hardware drivers are installed.
SYSmark provides an exhaustive battery of real-world usage tests by replaying the same recorded commands given by one individual as they used a library of content creation and productivity applications. According to BAPCo, differences of three points or less is not meaningful for the final score. Although we have no idea why the program is still marketed as a “Preview” given the programs utilized are older versions (Photoshop CS2 anyone?), they still make for an effective test today.
It is important to note that a SSD is just one part of a computer system, and just because it can make disk accesses faster, the OS more responsive, and keep the system running smoothly that it doesn’t automatically make all programs execute faster. We can see here that most SSDs are capable of attaining the best scores, to see a larger difference we would need to first upgrade our Q6600 test platform. SSDs make the most sense in fast, modern systems and especially laptops where slow 5,400RPM hard drives are still often utilized today (*cough* Apple).
HD Tune has long been one of our favorite storage benchmarks, thanks in part to its ease-of-use, and its ability to deliver consistent results (which is obviously important). Since we are using HD Tune on storage devices that also house our OS, we’re unable to test the write performance, so here, we stick to both Read and Access Times.
The Corsair Force 90GB puts up more good results in the HD Tune program, although the larger capacity Vertex 2 gains a slight edge. It is good to see that despite a one-fourth reduction in capacity, that drive performance isn’t appreciably changed.
Again we show the access times as these are the hallmark of solid-state drives, which can’t be underscored enough. Three milliseconds to read a 1MB file sounds like a long time by SSD standards, but in actuality it is the quickest time here. The lone mechanical drive requires ten times as long to access the same size file by comparison. Bandwidth isn’t just a flat value, it is determined by the type as well as the size of the file operation that is utilized for the comparison.
HD Tach is a program similar to HD Tune, and although it hasn’t been updated in a few years, it’s still decent for testing SSDs. It offers a different method for calculating burst rates, as well as offering access time measurements below 0.1ms, which is unfortunately the limit for HD Tune. With a massive new program rewrite in the works, we look forward to seeing what the upcoming new version can do.
HD Tach is unfortunately showing its age as it is the only program that isn’t updated to natively support Windows 7 or Vista. A complete rewrite of the program was announced, but until then we will toss numbers in from this one to contrast with HD Tune. The 90GB Force didn’t fare well in this program, but again neither did drives like the Revo so we won’t read into this one.
With this we reach the first of our real-world tests where there are no unusual testing or scoring algorithms to leave us scratching our heads! First we start with a simple file transfer from one part of the same drive to another. Please note that straight file transfers to a different destination would post even higher results.
For this test we took a 7.16GB Dual-layer DVD image and timed how long was required to transfer it to a different 10GB partition created on the same drive. Keep in mind with a hard disk, this requires the actuator arm to seek back and forth between the source and destination sectors of the disk platter, while any SSD can instantly read and write to multiple flash chips at once. Still, this is not a test of the raw write speeds of a SSD as the SSD controller cannot read & write to the same parts of the NAND flash simultaneously.
Keep in mind our dual-layer DVD image is already highly compressed, but the SandForce controller is still able to work with it to achieve a satisfactory time of just over two minutes, which is a far cry better than the over six minutes required by the mechanical hard drive. It is worth nothing that large sequential file writes are the best case scenario for a mechanical drive.
For this test we utilize 500 large RAW files, and import them into Lightroom. We time how long it takes the program to import the files, cache them, and build the image library.
This test didn’t work out as originally envisioned as it isn’t sufficiently disk intensive, but we kept it in our reviews to illustrate that not all programs can see a benefit to an SSD. Here all the results are split into two main groups, which indicates to us that for most drives the CPU remains the bottleneck. Keep in mind Lightroom only utilizes two cores when loading images and generating a library, so it is not fully utilizing our quad-core CPU. For the second grouping the drive itself appears to be the bottleneck.
This test is perhaps the most important in our battery of benchmarks as it gives us truly real-world results. It is designed to simulate three kinds of multi-tasking scenarios in order to see how well the storage drive can cope with concurrent workloads. The better a drive performs here, the quicker and more responsive it should feel in everyday tasks. It shouldn’t need to be said that this is where SSDs shine and where traditional HDD bottlenecks are most directly experienced.
In addition to stressing the controller with a demanding, large queue to sort through (NCQ support helps here), this test will give any weak controller a complete panic attack by overwhelming it with simultaneous random read/write operations to juggle with large sequential writes, which is the Achilles’ heel of many cheaper SSDs.
Queue depth and IOPs optimizations have long been a strength of Intel’s own SSDs, however, there is enough differing types of workloads here that regardless of drive, every SSD should see some part of the workload playing to its own unique strengths and weaknesses in some fashion. A good-quality SSD should allow the system to remain responsive as the tasks are carried out in the background at all times. (Please for your own safety don’t try this at home on your HDD!)
Our Medium test consists of the following:
Our Heavy test consists of all-the-above in addition to a full Anti-Virus scan running concurrently in the background with the start of the test. The AV scan uses a static, unchanging 5.1GB test folder that contains 19,748 files and 2,414 sub-folders created from the Program Files directory.
Granted, even with a Core i7 processor, no computer user would be performing all of these tasks concurrently unless they wish to see their computer go unresponsive for up to 30 minutes at a time, but with an SSD, this is almost child’s play! If you think we are exaggerating then just look at what a fairly typical SATA HDD is able to offer, which is a representative sample for any other desktop HDD.
Last but not least, the Light test changes things slightly. This test is a batch file dropped into the Startup folder designed to load several programs as soon as Windows 7 reaches the desktop. This light test will open four websites in Firefox, load five images in Photoshop CS4, start playing our favorite 8 minute (56MB FLAC) music file in Winamp, and open a single large Word, Excel, and PowerPoint document, in addition to a single PDF file. For this specific test in particular, we start measuring from the moment the power button is pressed to the moment the last program and all files have been fully loaded ready for use.
In the time required for a typical computer using a hard disk drive to boot to desktop, it is possible for an SSD to have reached desktop and fully load a multitude of programs and files. In fact, the Corsair Force 90GB can load all the above programs and files in 76 seconds, which is still HALF the time required for a mechanical hard drive just to load Windows. That begs the question: maybe it isn’t the OS that is so slow, but the storage device most commonly used?
Keep in mind boot times are also dependant on the motherboard and system hardware, a newly released Sandy Bridge motherboard with UEFI that replaces the BIOS will drop boot times significantly below those seen in our older testbed. (The newer SB processor will certainly contribute, too).
Although not at the top, the 90GB Force drive puts in respectable performance in our medium batch test. One thing not shown in our graph is the remarkable consistency SandForce drives in particular seem able to deliver. Despite seven gigabytes of data being copied across partitions, a few more being generated from archive extractions, and hundreds to possibly thousands of file reads and writes occurring nearly at once, during all five runs results always fell within the same 14 second range. Now that is consistency! Those SSDs with good controllers in particular have this effect for generating very consistent results which is why they see widespread adoption in all sorts of benchmarking platforms, but from my personal experience SandForce drives are currently one of the best at it.
As mentioned already our heavy batch test uses the same scenario as the medium test, but adds a concurrent anti-virus scan of a static, 5.1GB test folder that consists of nothing but Program Files. If I was using a mechanical hard drive this is usually the part of the test where I could start it up, then watch last night’s evening news and come back in time to catch it finishing. (Okay, admittedly only if I skipped through the drug and oil commercials). Whereas a 320GB mechanical drive required an average of 23 and a half minutes, the 90GB Corsair Force on average was able to achieve the same result in a much more reliable seven minutes and eight seconds.
At less than a third of the time to complete the task it isn’t any wonder SSDs have become so attractive to end users and businesses alike. For consumers that don’t have the liberty of having their boss foot the bill they will need to decide if they have sufficiently demanding applications with disk IO (besides games, mind you) that need a fast solid-state drive, or if they simply want one bad enough. However it is done, SSDs are well worth the upgrade.
We already mentioned the SSD consistency, for the heavy test the slowest and quickest results deviated only by 19 seconds between all five runs. Now go run a full anti-virus scan on a computer that uses a hard drive and try to get anything done at the same time, let alone reliably, it just isn’t going to happen. With the Corsair Force the system remained responsive and had plenty of performance left to handle opening programs or files, including stutter-free playback of music even with everything else running in the background. With an SSD program file installations, anti-virus scans, file searches, and program updates don’t have to interrupt whatever the user is doing.
This is a test that any Windows 7 user can perform on their own system without needing to download anything. For those curious, Vista should allow the same, but we can’t guarantee the results will be directly comparable due to changes made in the WinSAT program. To run the program, hit the Windows Key + R at the same time, and type cmd into the run box. In the command prompt window type (or right-click and paste) the following without quotes: “cd c:windowssystem32” and hit enter. Once there input, again without quotes: “winsat disk -drive c -ran -write -count 10” and enter again.
This command runs a small portion of the Windows Experience Index’s drive assessment, specifically it uses small random writes and calculates how fast in MB/s the drive can sustain it. As we mentioned before with the HD Tune results, regardless of what drive is in question, its actual performance depends on what file size is being referred to. The smaller the file size, generally the lower the performance for a hard disk drive. So again, the small random file writes will be brutal.
Windows 7 appears to rather enjoy the newer firmware on the Corsair Force drive, as it is able to surpass the Vertex 2 100GB SSD in this test by an opportune 5MB/s in WEI’s small file write test.
For the boot test, we perform a cold boot, with the stopwatch starting the moment the power button is pressed until the last systray icon has finished loading. A large number of factors can change how fast a computer boots, from motherboard to just the BIOS configuration, so these times should not be used as an expectation of how fast the SSD will boot in your respective system. With some newer motherboards condensing the time taken in the boot process, boot times could reach significantly lower than these.
Ignoring the Revo’s showboating, it is fairly clear the test system is limited to 55 seconds under most conditions. With ASUS replacing the old 80’s era BIOS with a new UEFI BIOS that finally brings the BIOS into the modern age, a P67, Z67, or mostly any new UEFI motherboard should allow for drastically lower boot times.
With the right settings we’ve heard 20 seconds is very possible, for example… my own personal gaming system is around 30 by comparison. Again this just illustrates that when considering any sort of computer upgrade, make sure the computer itself can actually fully utilize it before doing so. Back on topic however, it is clear the 90GB drive has no issue keeping up with the best boot times here, making it a good drive option for laptops in particular.
Last, but certainly not least, are the game level-loading times. SSDs tend to improve application load times significantly over their mechanical brethren, and games are no exception.
Crysis is still infamous for how well it could stress the entire PC, and although Crysis Warhead was a significant improvement and much better optimized than its predecessor, it still makes for one of the better gaming benchmarks to use. For this test, we timed how long it took to load the first level, Ambush. We also figured we would use the newer Left 4 Dead 2 game, for its slightly longer-than-average load times. Here, we timed how long it took to load the final chapter in the Hard Rain campaign.
While the difference between the best SSD and the typical HDD almost exactly halves the game loading time, that 15 seconds really doesn’t seem like much in L4D2, or 30 seconds in Crysis Warhead. At least until considering that each level load is one of many, then halving the load time really starts to add up. Some games involve quite a few load points or build them directly into the level (Half Life 2 being a good example), and smoother, more fluid transitions will greatly preserve game the immersion. In which case, for some gamers the answer becomes a definite yes.
Finally, we reach the most important of our benchmarks! Okay, maybe those were really the batch test scenarios, but admittedly games are more interesting! Choosing almost any SSD results in a near halving of load times over a mechanical drive, but don’t forget that is just a single level. Keep in mind just how many loading screens are seen during the typical single-player game; gamers can rather quickly come to appreciate having even a few seconds saved off load times rather quickly, and all the more so for those immersive titles.
Most SSDs are quick enough that there isn’t any distinction between them in our tests, and as hinted at before, this is because games, like most programs, are written to be especially frugal with hard drive accesses. Those that aren’t quickly earn a bad rep as poorly coded or simply unoptimized software due to the poor responsiveness they exhibit. Even so, we find it curious that the Force 90GB tied our best (and only) SSD to break away from the pack in the Crysis test. We can only attribute that to a quirk with Crysis itself, or perhaps some minor optimization in the drive’s newer firmware as compared to the Vertex 2.
Amongst consumers there is a fair amount of negativity being perpetuated regarding 25nm NAND. While it is true the NAND cells have a lower erase cycle endurance as opposed to 34nm NAND, it needs to be mentioned that these drives are over-provisioned with additional NAND capacity specifically to offset this problem. Mechanical hard drives have done this for decades; if a sector became unusable they would simply allocate a sector from the hidden spare area portion of the drive.
This leaves only the question of longevity… how long is long, exactly? The longevity of solid-state drives has long been one of the largest sticking points with consumers. (That pun is getting long in the tooth). It is one thing to pay a price premium for a product that would significantly outlast its cheap competitor, but with SSDs it was oft regarded that they wouldn’t last even as long as a hard drive. Let us put this one to rest. It has been indicated that 25nm NAND will offer somewhere between 3,000-5,000 write-erase cycles. 34nm NAND specifications indicated a minimum of 5,000 cycles, but until whitepapers are published let’s assume 3,000 is the new number.
The average person will write somewhere between 2-5GB a day as a rough estimate. Most often it is less, but let us almost double that to 9GB a day. Assuming we buy a 90GB Corsair Force drive, it would take 10 days to fill the drive with data. Once filled, that means we have used 1 erase cycle to clear the drive and start anew. Keep in mind intelligent wear leveling will ensure all writes get evenly distributed across a drive over its lifespan. Ten days multiplied by 3,000 write-erase cycles equates into roughly 82 years before we would exhaust every last cell’s write-erase cycle. NAND flash will actually lose its charge after 10 years of age by comparison.
Admittedly things aren’t that simple thanks to how NAND memory is structured. To write something it occurs at the 4KB or 8KB size, what we call a page (instead of a sector on a HDD). But in order for us to erase the data in a single page, the drive must erase an entire block of pages that one page resides on. SSD’s can write to individual pages, but only erase at the block level. All SSDs do this regularly during operation all without notice by the user. This effect is known as write amplification. For the best drives this number should be as close to 1 as possible, but let’s assume 10 as a worst case scenario.
9GB / 90GB = 10 days. 10 * 3,000 cycles is 30,000 days, or 82 years
(9GB*10) = 90GB written a day. 90/90GB = 1 cycle a day. 1*3,000 cycles is 8.2 years.
Or take our first number and divide by 10, although we imagine realizing the controller itself could write 90GB of data a day (at the NAND/controller level) and still have the drive easily outlast its warranty period should help the idea sink in a bit. Oh, and we should mention the 34nm model 90GB Force drive has 96GB of NAND on board, we didn’t factor that extra 6GB of spare area into our above calculations.
Good quality controllers are required, (not all controllers are created equal), but otherwise solid-state drives should enjoy a lifespan well above that three year warranty period, and potentially well above that of most mechanical hard drives in the process. If the SSD did break, it at least shouldn’t be from the NAND wearing out. Of course, given the sharp decline at which fabrication size decreases write/erase cycles, we suspect the next process shrink from 25nm the next smaller node may require either new technological advances in NAND fabrication or some ingenious solutions. But until then I personally no longer have any qualm with using a 25nm NAND drive. And for those that still do, Corsair has made a clear distinction on which Force drives feature which NAND, something we like to applaud the company for.
There really isn’t any question about it; if considering the Corsair Force series solid-state drives, then they almost recommend themselves. For at least a little while longer SandForce controllers are the most powerful SSD controllers around, and Corsair is a company that stands behind its products.
With prices continuing to decline, the time hasn’t been better to find some amazing deals on a solid-state drive. Currently, street level prices on Corsair Force drives fall around $1.60-80 a GB, but we have already seen one store promotion reach considerably lower.
This leads us to the unspoken white elephant in the room. We would be negligent if we didn’t mention that SandForce will be launching a new, SF-2200 controller and that these SSDs will be even faster. If looking for only the absolute best and you do have a SATA 6Gbps port to use, then we recommend holding out for one if a few more months wait isn’t an issue. In addition, it was announced these will be carry a price premium over current SF-1200 drives, likely even more so for the initial months in all likelihood.
That said, if the motherboard or laptop in question only supports SATA 3Gbps then there is surprisingly little reason to contemplate a future SF-2200 drive, because without SATA 6Gbps the performance would be nearly identical to current SandForce drives just with a higher price premium. We can’t say it enough, if planning to use an SSD with a SATA 3Gbps port then you should simply stick to a SF-1200 drive such as the Corsair Force series.
Corsair’s 90GB Force series sold-state drive features one of the best SSD controllers currently available, delivering high-end performance at a more manageable price point. Whether for a desktop or a notebook upgrade the jump in system responsiveness will be tangible, and we would go as far to say that after making the switch to a solid-state drive we suspect you won’t ever wish to wait a normal hard drive based system again.
Have a comment you wish to make on this article? Recommendations? Criticism? Feel free to head over to our related thread and put your words to our virtual paper! There is no requirement to register in order to respond to these threads, but it sure doesn’t hurt!
Copyright © 2005-2020 Techgage Networks Inc. - All Rights Reserved.