Date: June 22, 2011
Author(s): Robert Tanner
OCZ’s Vertex series of SSDs have earned the reputation as being some of the best, by offering excellent throughput and random 4KB IOPS performance. Vertex 3 takes things to an entirely new level, in some cases doubling the performance of its predecessor, and gives us an uncompromising balance across any performance metric you care to throw at it.
Solid-state drives have always held a bit of a fascination with me, at least ever since 2008 when Intel brought them to the attention of the consumer market. SSDs have progressed a long ways in the three years since, a feat I find particularly underscored by the SSD in today’s review.
The simple fact is, there isn’t another part of the computer industry where leaps of 50% let alone 100% or more are still occurring with every successive generation of hardware. The change from the SandForce-based SF-1200 to the SF-2281 controller that powers the OCZ Vertex 3 solid-state drive is one of those leaps.
Granted, the new SATA 6Gb/s interface featured on the Vertex 3 has a significant role in the new levels of performance, but it is only due to the SandForce controller that the Vertex 3 is able to set new levels of performance previously only seen from dual and even quad-controllers in RAID 0 before now. It should therefore go without saying that consumers wishing to get the full performance out of a Vertex 3 or Agility 3 SSD will need a system or laptop with an available SATA 6Gb/s port.
The Vertex is OCZ’s premier model of solid-state drive. Like the Agility 3, the Vertex 3 is powered by a SF-2281 controller and features a SATA 6Gb/s interface, but after that things begin to differ slightly. The Vertex 3 makes use of Intel’s 25nm MLC NAND and uses a custom PCB of OCZ’s own design.
The Vertex 3 series is offered in capacities of 120GB, 240GB, and 480GB, all of which feature slightly over 12% of the given capacity as spare area. This “over-provisioning” of the SSD allows it to better maintain performance when full and provides a scratch area to replace any NAND sectors as they wear out, but is why the capacity seen within the OS is lower than what is listed on the box. Curiously the Vertex 3 datasheet shows an un-mentioned 60GB model, so it is possible a more affordable 60GB model Vertex 3 may show up within the near future.
|OCZ Vertex 3 (6Gbps)|
Up to 550MB/s
Up to 550MB/s
Up to 530MB/s
Up to 500MB/s
Up to 520MB/s
Up to 450MB/s
4KB Random Read
4KB Random Write
As with past OCZ drives the Vertex 3 comes in a small package which includes the mounting screws and a solid metal 3.5″ bay adapter, which serves double-duty as a layer of armor plating during shipment. The plastic top drive housing leaves a little something to be desired, but that’s purely a cosmetic quibble.
For those curious, it is worth mentioning that OCZ also offers Agility 3 and Solid 3 solid-state drives. The Agility 3 uses less expensive NAND and will offer a few percentage points lower performance because of it, but otherwise remains identical to the Vertex series. The Solid 3 is the value offering, and although it still uses a SF-2281 controller it appears to halve the NAND chips on the PCB, which would explain why its listed performance drops so much. In any event the Solid 3 doesn’t appear to be available yet, but we would easily recommend the Agility 3 as the price/performance sweet spot to the Vertex 3.
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 Solid-State Drive Test System
Intel Core i7-2600 – 3.50GHz (Locked) Quad-Core
ASUS P8P67 Deluxe
4GB Kingston DDR3-1866
AMD Radeon HD 5770
Hitachi 7200RPM 2TB Hard Drive
Kingston V+ Series 128GB
OCZ Vertex 2 120GB
OCZ Vertex 3 240GB
Antec NeoHE 550W
Windows 7 Ultimate SP1 64-bit
Our Windows 7 Desktop for SSD Testing
When preparing our SSD testbed for benchmarking we follow these guidelines:
For our new Sandy Bridge storage testbed we have migrated to using test images for our drives. All drives are imaged with the cloned test image to ensure all drivers, programs, and settings remain identical for testing purposes. We feel disk cloning software and SSD controller technology has matured to the point where potential issues such as non-aligned sectors are no longer a potential issue.
For testing, we run all tests five times dropping the highest and lowest results, then take the average of the middle three. And who said that college statistics class wouldn’t prove useful? If any anomalous results are seen the test will be run again. Given the complexities of modern computers, and especially today’s operating systems and the software that runs on them, 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 always actively 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.
Futuremark’s PCMark benchmarking suite should need no introduction as it has been a staple of PC benchmarks for the better half of a decade. PCMark offers a range of tests to gauge every aspect of a computer’s performance and presents it in a neat simple final result. Thankfully it also breaks down the overall score with individual subsystem scores (such as Memory, Storage, etc) in addition to given individual test results.
With the latest 2011 release of PCMark 7 we should hopefully see quite a few changes to how SSDs are handled, and the resulting scores computed, as previously, results were biased towards sequential read and write performance. With its Windows 7 focus PCMark 7 offers a variety of storage system tests, such as simulating a Windows Defender scan and using Windows Media Center to using other built-in programs for video and music file manipulation. But for those that just want a nice overarching score, it has those too.
Interestingly, it appears the new PCMark 7 suite does change how SSD scores are computed, as its predecessor Vantage produced some notoriously unexplainable results for us in the past. Pure sequential writes no longer are given as much precedence which is as it should be.
OCZ Vertex 3 does not disappoint and perches neatly atop of the charts – every single one of them. It’s a promising, although not unexpected showing for the SF-2281 powered Vertex 3!
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.
Not much can be said here. Small random file writes of 512B to 4KB are stressful for even an SSD designed to handle them, and Iometer is a particularly brutal test for consumer SSDs in general. Even so the Vertex 3 completely dominates the V2 here. Performance like this is literally unheard of for these types of small file access patterns in these tests. Even trying to compare the Vertex 3’s results to a typical hard drive is just futile here.
As the name hints, AS SSD is a nifty little program written exclusively for solid-state drives. It can be run on a mechanical hard drive, but be warned what takes a few minutes will require the better part of an hour to complete! This handy tool measures sequential reads and writes in addition to the important 4KB random reads and writes, then ranks the results with a final score for quick comparisons with other SSDs.
In addition to the main test there is a secondary benchmark that simulates the type of data transferred for ISO, Program, and Game files. With version 1.6 a compression benchmark was also added although not utilized here. 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.
The SATA 6Gb/s interface allows the Vertex 3 room to stretch its legs. When it comes to read requests SSDs have been bottlenecked by the SATA 3Gb/s interface for some time. The interface isn’t responsible for the 4KB and 4K-Thrd scores, however; these are purely due to the controller.
4K-Thrd is similar to the 4K test but spawns multiple requests; basically this tests how good the SSD is at handling multiple file actions at once. Some older SSDs and even current budget SSDs such as the Toshiba controller in the V+ drive are only capable of processing a single data request at a time (known as a queue depth, or QD 1). Obviously the Vertex 2’s SF-1200 could multitask, but its successor takes it up another level entirely. The Vertex 3 simply cleans house with this SSD benchmarking program.
Since we included a program designed to benchmark SSDs, we will include HD Tune as it benchmarks both hard disks and SSDs. Because the test drive houses the OS itself, HD Tune will not perform any write tests; we will have to be content with both the Read and Access times. HD Tune 4.6 added a new quick benchmark that we will include for users that wish to make a quick comparison with their own drives.
Average throughput is important, although minimum is perhaps even more so as it will give a worst case scenario. This is partly a holdover from HDDs, because as a mechanical hard drive is filled up the drive head moves towards the center of the platter. So for physics-related reasons, it means a HDD will lose performance as it fills with data as can be seen in the first chart.
It is hard not to notice the Vertex 3’s dominance in the same chart, not even the Vertex 2 is able to come close. The second chart handles random file reads, and again the Vertex 3 takes no prisoners, it’s downright ruthless.
Lastly, access times are also looking good. It is worth noting access times are partly what give SSDs their mojo, as in the computer realm the difference between 14 milliseconds and .14 of one millisecond is literally an eternity!
Finally, we reach the first of our real-world tests where there are no unusual testing or scoring algorithms to leave us scratching our heads, just simple tests to see how an SSD changes actual system performance.
For the File Transfer test we took a 4.5GB archive and timed how much time was required to transfer the file to another destination on the same drive. Keep in mind that 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 concurrently read and write to separate flash chips at once.
The OCZ Vertex 3 easily leaves the other SSDs in the dust, including the Kingston V+ which itself has a high sequential write speed compared to most other SSDs. Usually it is hard to create a balanced SSD that can deliver good sequential write speeds without losing random write performance, and vice-versa. There are quite a few examples of SSDs that fit into one or the other category still on the market. I say usually, because the SF-2281 controller inside the Vertex 3 proves it can, in fact, be done.
With Adobe Lightroom, importing image files with “Copy” simply acts like a file transfer, exactly like our previous test. Rather than simply time how long it takes to create a duplicate set of 500 RAW files we elected to choose the “Copy as DNG” import option. This will convert the NEF files (Nikon’s equivalent to RAW) into the Digital Negative standard while importing them to its image library.
This test was not particularly effective as Adobe Lightroom 3.4 only utilizes two threads, meaning that even with the power of a Core i7 that has eight threads available, the CPU was still the main bottleneck. When Adobe deems fit to update Lightroom to take advantage of more threads we will see a real need for faster storage.
Even so, upgrading to any SSD will save a minute and a half off the import time over a mechanical drive. The Vertex 3 shows there was a little more to gain and manages to stretch out a short lead by dropping the import time by over two minutes.
These tests are perhaps the most important in our battery of benchmarks as they give us a wide range of real-world results. They range from very light to downright grueling, and will showcase which drives can shine under the most demanding scenarios they might encounter in your personal system. Few computer users run their tasks in a vacuum; often several programs are in use concurrently while others are running in the background.
To excel in these tasks the SSD controller and firmware will need to be well-balanced. It will need to have excellent random read, small random write capability, and still have enough sequential writes to get the job done quickly. It is admittedly hard for an SSD controller to be optimized for all three things at once, and typically some SSDs are only optimized for sequential writes at the expense of everything else. Still, fast access times will give any SSD an inherent advantage over a mechanical hard drive.
First up is our light batch test. This test is a simple batch file placed into the startup folder, which Windows 7 will automatically execute at startup. This is perhaps the most directly relevant test to our readers, as almost everyone has to endure boot times and then the additional time it takes for their usual or favorite programs to load before they can start using their system.
The batch file will open four websites in Firefox, load five 5MB or greater images in Photoshop CS5, and open a document in Word, Excel, and PowerPoint each, which adds an additional 15MB. As a final measure, a few small system monitoring applications are started, a 3MB PDF file and zip archive are both opened for viewing, and while everything proceeds to load, an old, favorite FLAC music file (56MB) is loaded into Winamp for playback. If it sounds like the light batch file needs to go on a diet, then the results should surprise!
Our Medium batch test is similar although timers are built in to space out the user commands. Time begins counting from the launch of the batch file and ends when all tasks have completed. The medium test consists of the following:
To keep things simple, the heavy batch test is identical to the medium test in all respects save for one key difference. Computer users should be familiar with the slowdown or even molasses-like feel that occurs from an anti-virus scan running in the background. The heavy test will capitalize on this by running an anti-virus scan from Microsoft Security Essentials on a static, unchanging 5.1GB test folder that contains 19,748 files and 2,414 sub-folders copied from the Program Files directory. Because it is otherwise identical, results from the medium & heavy batch tests are directly comparable.
Granted, even with a Core i7 processor, no computer user using a hard drive 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. For a good quality SSD, the above isn’t even enough to make the system crawl or go unresponsive. Playing a game with an anti-virus scan in the background without losing FPS is very possible. So if this sort of system abuse, or “multitasking” sounds vaguely like your daily routine when you sit down at the PC, then the results may be of interest.
So, a few walls of text and a single graph later, what do we have? Well from the press of the power button to the very last image loading in Photoshop CS5, the entire system boot and Light batch test completes in an average of 39.4 seconds with the OCZ Vertex 3. Since most computers can’t even load Windows 7 in that time-frame, that should say something. A traditional hard drive requires almost three minutes completing the same test the Vertex 3 can finish in less than forty seconds, system boot time included.
The medium batch test is where the real stress begins for these drives. As we continue to re-add drives to our new test system, the full spread of SSD results will again become apparent, but it is pretty clear the Vertex 3 brings more muscle to the table than even its own predecessor. OCZ’s Vertex 3 completed the entire workload in slightly under an impressive three and-a-half minutes. The nearest SSD, the V2, required an additional 108 seconds or 1:52 to complete. To keep things in perspective a large capacity, the modern mechanical drive mustered up nine minutes for the same task.
Things get really interesting with the heavy batch test. Results are directly comparable to the medium test here, as the only change was the addition of the anti-virus scan. Raw sequential performance won’t win this benchmark. Those controllers with high queue depths to handle the multitasking load will shine here, and will quickly separate the men from the boys.
Although it was clear the Vertex 3 was going to win, it was disconcerting to see an anti-virus scan only add an average of 20 seconds to the task time. To try and put this into perspective, a full scan of a 5GB folder containing 19,748 program files only required an additional 20 seconds to perform concurrently while the medium batch test was running. That same task more than doubled the time required for the Toshiba-based SSD to complete the entire test, and an additional five minutes to the hard drive time which ended up taking over 15 minutes to finish.
The Vertex 3’s predecessor finally shows its true strength here as well, but even the Vertex 2 required an extra two minutes here. The Vertex 3 was simply able to finish in 2/3rds the time of the V2. These tests throw everything except a photo of the kitchen sink at these drives, and stress every possible aspect of the SSD controller. Queue depth for multitasking, random reads and lots of small random writes, and even some sequential reads and writes, but the Vertex 3 doesn’t seem to have any weaknesses regardless of workload, even despite the huge jump in sequential read and write performance this drive brings to the table.
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 starts, from the motherboard to the BIOS/EFI configuration, so these times should not be used as an expectation of how fast the SSD will boot in your respective system. Thanks to motherboards replacing the BIOS with UEFI boot times have dropped significantly in many cases.
It appears that after a certain point the storage drive is not the bottleneck. Users that overclock their Core i7-2500K or 2600K processors can attain even lower results, but changing the OS will have the largest impact at this point.
The Vertex 3 manages a blazing quick 31.5 second boot time from start to finish, but does even faster with warm boots (a restart with the machine powered on previously) requiring around twenty seconds.
Last but certainly not least of our benchmarks are the game level-load times. SSDs are great at decreasing load intervals, and having an SSD can appreciably improve game immersion by minimizing load delays. It may not seem like much, but after a few levels, having the load times decrease by even a third compared to a hard drive adds up fast.
For our new regimen we chose Portal 2 and Civilization V. Portal 2 is already a very well optimized game and isn’t particularly demanding, and Civilization V is anything but either of those. For Portal 2 we chose to load the larger sp_a3_03 chapter, while with Civ V we loaded a save game file from late in a large game.
Out of our entire benchmark suite this is probably the worst showing for the Vertex 3, but even so it still wins both game level load tests by a fraction of a second. The important part to take away from this is if you are considering upgrading from a hard drive you will see some very real gains in performance, but the same can’t be said for most SSDs.
The Vertex 3 is 38.2% faster than the hard drive when loading the chapter in Portal 2. After just a couple chapter loads that 38% starts to make a tremendous difference and will go a long ways toward keeping the game experience smooth and immersive.
I will just say OCZ has simply done it again. The OCZ Vertex 3 lives up to its heritage and delivers another leap forward in performance. Sequential write speeds alone can be more than double that of its predecessor, and 4KB random write IOPS have improved similarly too. Thanks in part to the Vertex 3’s SATA 6Gb/s interface, the second generation SandForce controller has no trouble dominating the competition.
We would do a disservice to not point out that competing drives such as the Crucial m4 and Intel 320 Series were not compared here. But the simple truth is even if they were, the Vertex 3 offers significantly higher sequential and random write throughput than the Intel 320 Series. On the flipside against the m4 SSD, the Vertex 3 offers significantly higher random IOPS performance. And before anyone mentions I left out the Intel 510 Series, it is worth pointing out that they utilize a similar Marvell controller to that found in the m4 SSDs and as such include the same performance issues as the m4 series.
There simply isn’t a competitor on the market right now that can take on the this drive in terms of performance; the Vertex 3 is simply the drive to have, assuming one has a SATA 6Gb/s port to pair it with.
That leaves just price. At slightly over $2 per GB, the price is still too high for widespread adoption, but even then the V3 SSD is priced the same as the closest thing the Vertex 3 has to any competition, including the Intel 320 Series. The Intel 320 Series is a good drive with a reputation for endurance, but from a performance standpoint it is outgunned unless you are stuck with using a SATA 3Gb/s interface. If a SATA 6Gb/s port is available then there is little reason to choose anything other than the Vertex 3.
I’d like to note that the original Vertex SSD received our Editor’s Choice award and as such it was the drive to have for my personal PC. After a year of rigorous service I have had no complaints and was forever sold on solid-state drives since. I can sit down at a PC and quickly tell simply by use if it has an SSD or not; the difference in responsiveness alone is tangible. As OCZ’s Vertex 3 will receive our Editor’s Choice award, it is only fitting that I will be upgrading the SSD in my personal rig to one in the near-future.
The question “what makes a good solid-state drive” is a particularly complex one to answer, and not every single facet was directly addressed in this review. On early SSDs stuttering used to be a problem, but better controllers optimized for random IOPs and queue depths >1 emerged to fix that. TRIM support became widespread in order to increase performance and improve NAND longevity. Random 4KB write IOPS is not everything in of itself, but it is a very large portion of the final recipe. The Vertex 3 simply takes all this and wraps it up into one single package. It offers absurdly fast sequential reads and writes, yet still manages even more absurd amounts of random write throughput at any file size.
The OCZ Vertex 3 is, simply put, the current pinnacle in SSD performance.
OCZ’s Vertex 3 SSD
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.