If you are interested in a deeper look at the SanDisk Extreme II architecture we're already published an Architecture Deep Dive earlier today. This review will touch on some of the facts in that article, but we're going to expand on the NAND flash a bit more here.
Nearly a year ago we sat down with two SSD's at Computex for an early look at IMFT 20nm flash and Toshiba 19nm Toggle Flash. Both drives used identical LSI SandForce SF-2281 controllers and matching firmware. By the time we walked away, it was clear that 20nm ONFi flash had lost a lot of performance compared to 25nm ONFi, and that 19nm Toggle wasn't much different than 24nm Toggle when it came to performance. At the time we thought IMFT would get it together, but now that we have real products in hand, it's becoming clear that 20nm ONFi flash isn't very good for smaller capacity SSD's, like the 120GB capacity size.
Even though today's review is on the SanDisk Extreme II, I can't help myself from drawing attention to the 19nm vs. 20nm differences. The SanDisk Extreme II uses the Marvell 88SS9187 controller, the same as the Plextor M5 Pro / Xtreme and Crucial M500. Since we have all of these drives in the 128GB class size, we can see how NAND flash affects performance. A large portion of our Architecture Deep Dive linked above talks about SanDisk's new ALB flash with nCache.
Marvell doesn't deliver programming / firmware with their products so if you want to use a Marvell SSD controller, you need to build your own firmware, like SanDisk did. The other option is to go to a third party like LiteOn or MemoRight to have it made for you. When you use a third-party firmware, you need to work with them on fixing issues that may pop up. Working with a third-party means delays, something SanDisk wants to avoid after the original Extreme SSD issues.
Specifications, Pricing and Availability
The SanDisk Extreme II starts shipping today and in three capacity sizes - 120GB, 240GB and 480GB. Many of us would love to see a 960GB version, but SanDisk chose to leave a massive capacity size model off the table, at least for now. At the heart of Extreme II is a Marvell 88SS9187 controller with custom SanDisk firmware. Paired with the 9187 is SanDisk 19nm MLC Toggle flash rated for 3K P/E cycles.
In a move designed to set a new standard, or to rebut the old ways, SanDisk used CrystalDiskMark to determine the marketing performance numbers. For years, ATTO and IOMeter were used to come up with marketing performance results, but over time, companies moved away from the default ATTO tests (QD4) and IOMeter can be manipulated with compressible or incompressible results. To stop any confusion, SanDisk chose to use CDM for their marketing materials.
With that explanation out of the way, let's look at the specs. The 120GB Extreme II delivers 550MB/s sequential read performance and 340MB/s sequential write performance. IOPS performance comes in at 91K read and 74K write.
Each capacity size ships in two flavors - a Notebook version with a 7mm to 9.5mm adapter and a full Desktop Upgrade Kit with a desktop adapter bracket, SATA cable and mounting screws. Extreme II also has a five year warranty.
Just days before going live with the reviews we received word about pricing. SanDisk's MSRP for the 120GB Extreme II is $129.99, 240GB is $229.99 and the 480GB is $439.99.
SanDisk Extreme II 120GB
SanDisk delivered three drives, one of each capacity size, but we didn't receive the retail package to show today. We don't expect much of a change from the Ultra Plus or original Extreme. This is a 7mm z-height product with a metal base and a plastic top cover.
The capacity size, model and serial number are all located on the back of the drive.
The 7mm z-height, 2.5" form factor uses the same mounting points on the bottom and sides as traditional notebook drives that are 9.5mm tall.
The SATA power and data connectors are also the same as standard notebook drives. The Extreme II will fit in your notebook and many ultrabooks without issue. Desktop users can purchase the Desktop Replacement Kit and receive a desktop adapter bracket that puts the drive in a standard 3.5" HDD bay.
Here we get our first real shot of the SanDisk Extreme II. The 120GB is the only drive with just four NAND flash chips. The higher density in each chips is part of the reason why we're seeing 120GB performance drop off so much with many of the SSD's we've tested lately. Without lanes to more chips, interleaving is reduced. We'll talk more about this in the Final Thoughts and how SanDisk worked around this issue.
Although Extreme II doesn't use the pads on the back of the PCB for host power fail protection, I wouldn't be surprised to see this product tip up at some point in time with more enterprise features. The Cxxx letter / number designation means capacitor and Extreme II could use a large number of capacitors if SanDisk wanted to enable the feature. It's common to see consumer SSD's share the same PCB with enterprise parts.
As we mentioned in the introduction, the Extreme II uses the Marvell 88SS9187 controller. This is an 8-channel design, the same used Plextor's M5 Pro and M5 Xtreme as well as Crucial's M500.
Buffering data is a Samsung 1GB DDR3 DRAM chip.
The star of the show, as you will soon see.
Benchmarks - Test System Setup and ATTO Baseline Performance
Desktop Test System
Lenovo W530 - Mobile Workstation
We use two systems for SSD testing. The desktop runs a majority of the tests and the Lenovo W530 runs the notebook power tests as well as the real-world file transfer benchmark.
ATTO Baseline Performance
Version and / or Patch Used: 2.34
ATTO is used by many disk manufacturers to determine the read and write speeds that will be presented to customers.
Our ATTO test shows a maximum read speed of 553MB/s and a maximum write speed of right around 340MB/s. Even though we use a different method than SanDisk, our numbers are in line with their marketing numbers.
The 340 MB/s write number may seem low, but it's important to remember we are working with next-gen flash and not 24/25nm. To put Extreme II's write performance into perspective, the Crucial M500 with the same controller, different firmware and different flash only managed to product 140MB/s write speed in ATTO.
Benchmarks - Sequential Performance
HD Tune Pro
Version and / or Patch Used: 4.00
Developer Homepage: http://www.efdsoftware.com
Product Homepage: http://www.hdtune.com
HD Tune is a Hard Disk utility which has the following functions:
Benchmark: measures the performance
Info: shows detailed information
Health: checks the health status by using SMART
Error Scan: scans the surface for errors
HD Tune Pro gives us accurate read, write and access time results and for the last couple of years has gained popularity amongst reviewers. It is now considered a must have application for storage device testing.
This review isn't just going to be about write performance though, the Extreme II can bring it in sequential read performance, too. Here we see a list of top performers and popular SSD's on the market today. The SanDisk Extreme II wrestles the highest average speed away from the Phison controlled MyDigitalSSD BP4, another drive with new 16K page size Toggle NAND.
On this chart the SanDisk Extreme II turns in a 322MB/s sequential write performance result. This test uses compressible data so the LSI SandForce controlled drives get to use their compression technology. We'll look at incompressible sequential results later in this review with CrystalDiskMark.
HD Tach - Sequential Write Performance after Random Writes
Here we see the sequential read and write results after a few random read and write tests. The write performance only drops below 200MB/s on a few occasions.
Benchmarks - AIDA64 Random Access Time
AIDA64 Random Access Time
Version and / or Patch Used: 1.60
Developer Homepage: http://www.aida64.com
Product Homepage: http://www.aida64.com
AIDA64 offers several different benchmarks for testing and optimizing your system or network. The Random Access test is one of very few if not only that will measure hard drives random access times in hundredths of milliseconds as oppose to tens of milliseconds.
Drives with only one or two tests displayed in the write test mean that they have failed the test and their Maximum and possibly their Average Scores were very high after the cache fills. This usually happens only with controllers manufactured by Jmicron.
Read latency for Extreme II sneaks in lower than the Crucial M500 120GB and on average matches the SuperSSpeed S301 drive with SLC NAND.
When it comes to write latency, we have to preface the test because we are using the second run from the Samsung 840 Pro. In our normal testing order, the 840 Pro produces extremely high write latency, even more than the Crucial M500. The high write latency is what keeps us from recommending the 840 Pro 128GB to any power user who will write a large volume of data to the drive quickly.
That said, the SanDisk Extreme II 120GB manages to keep write latency in check with its tiered storage system.
Benchmarks - Anvil Storage Utilities
Anvil Storage Utilities
Version and / or Patch Used: RC6
So what is Anvil Storage Utilities? First of all, it's a storage benchmark for SSD's and HDD's where you can check and monitor your performance. The Standard Storage Benchmark performs a series of tests, you can run a full test or just the read or the write test or you can run a single test, i.e. 4K DQ16.
Anvil Storage Utilities is not officially available yet but we've been playing with the beta for several months now. The author, Anvil on several international forums has been updating the software steadily and is adding new features every couple of months.
The software is used several different ways and to show different aspects for each drive. We've chosen to use this software to show the performance of a drive with two different data sets. The first is with compressible data and the second data set is incompressible data. Several users have requested this data in our SSD reviews.
0-Fill Compressible Data
When comparing incompressible to compressible performance, we didn't notice a large difference moving between file types.
Read IOPS through Queue Depth Scale
Companies like to talk about their high queue depth IOPS performance, but in the real-world, most of us rarely hit a queue depth higher than 4. QD8 would be like a biblical event under normal desktop use. It's that reason we came up with showing low queue depth IOPS performance.
The Extreme II trails the 840 Pro in 4K reads ever so slightly at low queue depths and is faster than the OCZ Vector. At this point I'm comfortable in saying we have a true contender for the performance crown in this capacity size.
Scaling Write IOPS through Queue Scale
When writing 4K data the Extreme II is right with the 840 Pro in QD1 and QD2. Vector has a nice lead in these tests, but we know OCZ plans to replace Vector with Vector 150 in Q3, and if OCZ continues down the same path as Vertex 450, then the Vector 150 will be 20nm.
Benchmarks - CrystalDiskMark
Version and / or Patch Used: 3.0 Technical Preview
Developer Homepage: http://crystalmark.info
Product Homepage: http://crystalmark.info/software/CrystalDiskMark/index-e.html
Download here: http://crystaldew.info/category/software/crystaldiskmark
CrystalDiskMark is a disk benchmark software that allows us to benchmark 4K and 4K queue depths with accuracy.
* Sequential reads/writes
* Random 4KB/512KB reads/writes
* Text copy
* Change dialog design
* internationalization (i18n)
Note: Crystal Disk Mark 3.0 Technical Preview was used for these tests since it offers the ability to measure native command queuing at 4 and 32.
Looking at 4K performance through the throughput glasses, we see that the Extreme II is ever so close to 40MB/s at QD1. It really doesn't get much closer to 40 than 38.67MB/s. At a 1QD, the Extreme II is faster than the Vector and just behind the 840 Pro. The same can be said about QD4. Extreme II does outperform both when reading sequential data.
Vector's 4K QD1 write speed is faster since it's still using 25nm flash and 840 Pro is slightly faster. In this test we finally get to see sequential write performance with incompressible data from the LSI SandForce drives. The SuperSSpeed S301 uses SLC NAND, so it doesn't run into the compressible vs. incompressible issues, but it costs significantly more than any other drive on the chart.
Extreme II takes the SandForce drives to town when using incompressible data. It's also important to remember that the Team SandForce products, including the original Extreme, now cost significantly less than the Extreme II. I found the original Extreme 240GB just days ago for $159.99, just $30 more than the Extreme II 120GB MSRP.
Benchmarks - PCMark Vantage Hard Disk Tests
PCMark Vantage - Hard Disk Tests
Version and / or Patch Used: 1.0.0
Developer Homepage: http://www.futuremark.com
Product Homepage: http://www.futuremark.com/benchmarks/pcmark-vantage/
PCMarkï¿½ Vantage is the first objective hardware performance benchmark for PCs running 32 and 64 bit versions of Microsoftï¿½ Windows Vistaï¿½. PCMark Vantage is perfectly suited for benchmarking any type of Microsoftï¿½ Windows Vista PC from multimedia home entertainment systems and laptops to dedicated workstations and high-end gaming rigs. Regardless of whether the benchmarker is an artist or an IT Professional, PCMark Vantage shows the user where their system soars or falls flat, and how to get the most performance possible out of their hardware. PCMark Vantage is easy enough for even the most casual enthusiast to use yet supports in-depth, professional industry grade testing.
FutureMark has developed a good set of hard disk tests for their PCMark Vantage Suite. Windows users can count on Vantage to show them how a drive will perform in normal day to day usage scenarios. For most users these are the tests that matter since many of the old hat ways to measure performance have become ineffective to measure true Windows performance.
HDD1 - Windows Defender
HDD2 - Gaming
HDD3 - Windows Photo Gallery
HDD4 - Vista Startup
HDD5 - Windows Movie Maker
HDD6 - Windows Media Center
HDD7 - Windows Media Player
HDD8 - Application Loading
Moving into real-world software performance, the Extreme II fills the chart very well and delivers performance in many tests that's higher than port Vector and 840 Pro in equal capacity sizes.
This has a direct relationship to SanDisk's advanced NAND. 4-corner performance when isolated can only show so much, putting it all together in mix mode, real-world use is difficult to tune. Vantage is a read heavy test, but in using typical real-world software traces, software we all use every day.
Benchmarks - PCMark Vantage - Drives with Data Testing
For a complete breakdown on the Drives with Data Testing please read this article. You will be able to perform this test at home with the files provided in the article - full instructions are included.
- Brief Methodology
SSD's perform differently when used for a period of time and when data is already present on the drive. The purpose of the Drives with Data testing is to show how a drive performs in these 'dirty' states. SSD's also need time to recover, either with TRIM or onboard garbage collection methods.
Drives with Data Testing - 25%, 50%, 75% Full States and Dirty / Empty Test
Files needed for 60 (64GB), 120 (128GB), 240 (256GB)
60GB Fill - 15GB, 30GB, 45GB
120GB Fill - 30GB, 60GB, 90GB
240GB Fill - 60GB, 120GB, 160GB
Empty but Dirty - a test run just after the fill tests and shows if a drive needs time to recover or if performance is instantly restored.
Measuring real-world performance with 50% of the NAND capacity full resulted in nearly identical performance to OCZ's Vector. At 75% full, Extreme II was a little faster than the current Vector 120GB, but down slightly to the 840 Pro.
Benchmarks - BootRacer
BootRacer - System Boot Time
Version and / or Patch Used: 4.0
Developer Homepage: Greatis
Product Homepage: BootRacer
Download here: http://www.greatis.com/bootracer/download.htm
Note: In this test we use the Lenovo W530 Mobile Workstation loaded with an operating system and several program files. The data on the drive at the time of the test is 45GB. The second test, 50GB Free, was ran with the drives filled with block data until only 50GB of free capacity remained.
Boot Racer's Time to Desktop is a measurement of time of how long it takes to restart a computer. As you can see the Lenovo W530 takes around the same amount of time with all of our SSD's.
The SanDisk Extreme II is right in the middle of all the results.
Benchmarks - DiskBench
DiskBench - Directory Copy
Version and / or Patch Used: 18.104.22.168
Developer Homepage: Nodesoft
Product Homepage: DiskBench
Download here: http://www.nodesoft.com/diskbench/download
Note: In this test we use the Lenovo W530 Mobile Workstation and a SuperSSpeed S301 SLC 128GB SSD to move a 15GB block of data to and from the target drive. This is part of our real-world test regiment. Roughly 45GB of data resides on the target drive before the '15GB Block' is transfer. The 15GB Block is the same data we built for the Data on Disk Testing and is a mix of compressible and incompressible data.
The only weak point we found in all of our tests is was when writing our 15.2GB directory to the Extreme II. Both 840 Pro and Vector with their 2xnm flash were faster in this test, but Extreme II performed a lot like the Plextor M5 Pro. The Extreme II easily outperformed M500, though.
Benchmarks - Power Testing
Bapco MobileMark 2012 1.5
Version and / or Patch Used: 2012 1.5
Developer Homepage: http://www.bapco.com
Test Homepage: http://www.bapco.com
MobileMark 2012 1.5 is an application-based benchmark that reflects usage patterns of business users in the areas of office productivity, media creation and media consumption. Unlike benchmarks that only measure battery life, MobileMark 2012 measures battery life and performance simultaneously, showing how well a system design addresses the inherent tradeoffs between performance and power management.
In our battery life test, the Extreme II 120GB delivered significantly more notebook time on battery power than both the 840 Pro and Vector. The drive fell a little behind the only retail 2.5" LSI SandForce B02 stepping we have in our lab, but the battery life was very close to it.
This is a big win for SanDisk considering all of the work that went into improving SF-2281's battery life. We generally don't see products based on Marvell hardware getting anywhere close to this level.
PCMark Vantage HDD Test - Power Draw
Here we see that the how Extreme II comes so close to B02 levels. B02, a.k.a ADATA SX900 128GB has lower idle power consumption, but Extreme II isn't far behind.
We're in an odd time for SSD's right now. With 25nm products still on warehouse shelves, we can't say those products are a thing of the past just yet. We know their days are numbered and as they move into history, a number of existing good products will be gone forever. Their replacements, the 20nm parts, just aren't as good as the products they are replacing, at least not with ONFi 20nm, not yet and most likely not anytime soon. Flash Forward Ltd, Toshiba / SanDisk 19nm process flash has a significant performance lead over IMFT, Intel / Micron 20nm flash. When assembled in a 128GB class SSD, the performance difference is massive. I suspect this is why Intel never launched a 335 Series SSD in 120GB capacity, and it's obvious that the Crucial M500 120GB has less than desirable write performance.
In Q3 2012 Toshiba shipped 33% of the world's NAND flash, a number that most likely had the SanDisk shipments rolled into it. Even if the SanDisk NAND shipments weren't included in the Trend Focus data, 33% Toggle either from Toshiba or SanDisk seems about right for what we see in the performance SSD market. Given the performance differences, that means a lot of companies had better trade in their paddle for an outboard, because the creek they are in is getting deep.
Today isn't going to be love on Extreme II day either. We would really like to see SanDisk tune up the performance a bit, mainly when transferring large amounts of data to the drive from another SSD. We're knick picking a little bit, but this drive is worthy of really diving in to find any tiny detail to complain about. We have to write about something, right?
In this capacity size it really boils down to this. The Samsung 840 Pro 128GB has awful write latency when worked hard and Vector as we know it today won't be around by Q4 this year. If OCZ uses 20nm flash, well we know how that will go over, if they go Toggle than the Vector 150 will be a competitor. The original Vector is still around so we can't exactly remove its performance crown. The SanDisk Extreme II 120GB performs at the same level as both of these class-leading products, but doesn't have the latency issue and isn't already slated for replacement. It's like when you wife tells you that you can buy a new (insert new toy here), but you have to wait for three months.
With the performance crown talk dialed back a bit, the SanDisk Extreme II 120GB is a solid product that is stable, ships with a nice accessory package, has a good price and is a strong performer. It definitely deserves to be included with the hyper-performance SSD's in this capacity size.
Check back later in the day. We have the performance results of the 240GB and 480GB Extreme II up next.