Samsung 960 PRO NVMe SSD Review
By David Ramsey
Manufacturer: Samsung Electronics Co., Ltd.
Product Name: SSD 960 PRO M.2
Part Number: MZ-V6P512BW
UPC: 887276185316
Prices: 512GB: $330 (Amazon|Newegg), 1TB: $630 (Amazon|Newegg), 2TB: $1300 (Amazon|Newegg)Full Disclosure: Samsung Electronics Co., Ltd. provided the product sample used in this article.
Samsung’s 950 PRO m.2 PCIE SSD set new benchmark records when we tested it here a few months back at Benchmark Reviews. The performance picture got even better we configured a RAID-0 array with two of these drives. Now they’ve introduced the new Samsung 960 PRO NVMe SSD, with upgraded NAND and a new controller, promising even more spectacular performance.
Specifications
Capacity | 512GB |
Interface | PCIe x4 |
Protocol | NVMe |
Form Factor | m.2 2280 |
Controller | Samsung Polaris |
NAND | Samsung 48-layer V-NAND |
TRIM | Yes |
Max. Read | Up to 3500MB/sec |
Max. Write | Up to 2100MB/sec |
Write Endurance | 400TB |
Warranty | 5 years |
Solid State vs Hard Disk
Benchmark Reviews has championed SSDs over hard disks for many years, as we feel that even when prices were much higher than they are now, the superior performance was worth it. Now that SSD prices have come down dramatically, there’s little reason for any but the most basic computers to use a spinning hard disk as a primary drive.
However, we’re now in the middle of another transition: within the last 18 months or so, standard SATA SSDs have all run up against what used to be the performance province of only the higher-end drives: the bandwidth limitations of the SATA interface. This means that no matter how fast your SSD is, you’re never going to see more than about 550MB/s transfer rates unless you stripe multiple drives together in RAID-0.
Moving from SATA to PCI Express (PCIe) is the obvious solution, but it required different controllers, and many desktop systems, especially enthusiast systems, simply didn’t have the PCI-E lanes to spare. The introduction of Intel’s Skylake architecture added more PCI-E lanes, and the icing on the cake was the introduction of NVMe (Non-Volatile Memory Express) to supplant the older IDE and AHCI protocols, which were designed for spinning hard disks and suffered from efficiency issues with fast SSDs.
The latest crop of PCI-E m.2 SSDs implementing the NVMe protocol promises vast increases in performance. Let’s see how this drive compares.
Closer Look: Samsung 960 PRO m.2 SSD
By moving from the 32-layer V-NAND used in the 950 PRO to 48-layer V-NAND, Samsung was able to dramatically increase the capacity of their m.2 storage products: the top capacity of the 950 PRO, 512GB, now represents the bottom of the lineup for the 960 PRO, which also offers 1 and 2TB models. All are single-sided 2280 form factor (22mm wide, 80mm long) m.2 “sticks”.
The only thing on the back of the PCB is a label.
A small installation and warranty manual is all that’s included with the drive.
It’s worth noting here that as of this writing, even the latest version of Windows 10 doesn’t handle NVMe SSDs in an optimal fashion. For the best performance, especially with high queue-depth random accesses, you should download and install Samsung’s NVMe driver.
SSD Testing Methodology
When we test storage devices, the two main metrics to consider are access time and transfer rate. Simply put, access time is the time is takes the storage device to start delivering data once the request has been received, while transfer rate is how fast (megabytes per second) the data comes once the transfer operation begins. With a hard disk, data transfer cannot begin until the disk’s head servo physically moves the read/write head to the correct track, and the rotation of the disk brings the designated sector under the head. Although modern servos are very fast, in the best case you’re still looking at several milliseconds to do this, while an SSD’s access time is always under a millisecond. The disadvantage is even worse if the data isn’t all in a contiguous space on the disk, since the head will have to be repositioned on the fly, leading to more delays.
Early consumer SSDs actually had slower transfer rates than the best hard disks, although their instantaneous access times more than made up for it. The zenith of consumer hard disk performance was probably reached in 2012 with the release of the Western Digital Velociraptor 1 terabyte disk. Spinning at 10,000RPM, this disk could under ideal circumstances (i.e. a synthetic bandwidth test) reach a sequential transfer rate of over 230MB/s. Keep this figure in mind as you read this review.
SSD Testing Disclaimer
Early on in our SSD coverage, Benchmark Reviews published an article which detailed Solid State Drive Benchmark Performance Testing. The research and discussion that went into producing that article changed the way we now test SSD products. Our previous perceptions of this technology were lost on one particular difference: the wear leveling algorithm that makes data a moving target. Without conclusive linear bandwidth testing or some other method of total-capacity testing, our previous performance results were rough estimates at best.
Our test results were obtained after each SSD had been prepared using DISKPART or Sanitary Erase tools. As a word of caution, applications such as these offer immediate but temporary restoration of original ‘pristine’ performance levels. In our tests, we discovered that the maximum performance results (charted) would decay as subsequent tests were performed. SSDs attached to TRIM enabled Operating Systems will benefit from continuously refreshed performance, whereas older O/S’s will require a garbage collection (GC) tool to avoid ‘dirty NAND’ performance degradation.
It’s critically important to understand that no software for the Microsoft Windows platform can accurately measure SSD performance in a comparable fashion. Synthetic benchmark tools such as ATTO Disk Benchmark and Iometer are helpful indicators, but should not be considered the ultimate determining factor. That factor should be measured in actual user experience of real-world applications. Benchmark Reviews includes both bandwidth benchmarks and application speed tests to present a conclusive measurement of product performance.
Test System
- Motherboard: MSI Z170A GAMING M7 Socket LGA 1151
- Processor: 4.0GHz Intel Core i7-6700K Skylake CPU
- System Memory: 16GB DDR4 2133MHz
- Operating System: Microsoft Windows 10, with Samsung NVMe driver
Storage Hardware Tested
The following storage hardware has been used in our benchmark performance testing, and may be included in portions of this article:
- Crucial RealSSD-C300 CTFDDAC256MAG-1G1 256GB SATA 6Gb/s MLC SSD
- Crucial m4 CT256M4SSD2 256GB SATA 6Gb/s MLC SSD
- Crucial M550 Solid State Drive515GBCT512M550SSD1
- Crucial MX100 Solid State Drive 512GBCT512MX100SSD1
- Crucial BX100 Solid State Drive 500GB CT500BX100SSD1
- Intel SSD 311 Series Larson Creek SSDSA2VP020G2E
- Intel SSD 320 Series MLC Solid State Drive SSDSA2CW160G3
- Intel SSD 335 Series Solid State Drive SSDSC2CT240A4K5
- Intel SSD 520 Series MLC Solid State Drive SSDSC2CW240A3
- OCZ Agility 2 OCZSSD2-2AGTE120G 120GB MLC SSD
- OCZ Agility 3 AGT3-25SAT3-240G 240GB MLC SSD
- OCZ Vertex 2 OCZSSD2-2VTXE120G 120GB MLC SSD
- OCZ Vertex 3 VTX3-25SAT3-240G 240GB MLC SSD
- OCZ Vertex 3.20 MLC SSD VTX3-25SAT3-240G.20 MLC SSD
- OCZ Vertex 4 VTX4-25SAT3-256G MLC SSD
- OCZ Vertex 450 VTX450-25SAT3-256G MLC SSD
- OCZ Vertex 460VTX460-25SAT3-240G MLC SSD
- OCZ Octane OCT1-25SAT3-512G MLC SSD
- OCZ Vector VTR1-25SAT3-256G MLC SSD
- OCZ Vector 150VTR150-25SAT3-240G MLC SSD
- Patriot Torqx 2 PT2128GS25SSDR 128GB MLC SSD
- WD SiliconEdge-Blue SSC-D0256SC-2100 256GB MLC SSD
Test Tools
- AS SSD Benchmark 1.6.4067.34354: Multi-purpose speed and operational performance test
- ATTO Disk Benchmark 2.46: Spot-tests static file size chunks for basic I/O bandwidth
- CrystalDiskMark 3.0.1a by Crystal Dew World: Sequential speed benchmark spot-tests various file size chunks
- Iometer 1.1.0 (built 08-Nov-2010) by Intel Corporation: Tests IOPS performance and I/O response time
- Finalwire AIDA64: Disk Benchmark component tests linear read and write bandwidth speeds
- Futuremark PCMark Vantage: HDD Benchmark Suite tests real-world drive performance
Test Results Disclaimer
This article utilizes benchmark software tools to produce operational IOPS performance and bandwidth speed results. Each test was conducted in a specific fashion, and repeated for all products. These test results are not comparable to any other benchmark application, neither on this website or another, regardless of similar IOPS or MB/s terminology in the scores. The test results in this project are only intended to be compared to the other test results conducted in identical fashion for this article.
NOTE: The Samsung 960 PRO is a Non-Volatile Memory Express (NVMe) drive. While Windows 10 includes a generic NVMe driver, Samsung recommends the use of their custom NVMe driver with the 960 PRO, and our testing does indeed show dramatic performance improvements with this driver. All our benchmark tests were run with this driver installed.
AS-SSD Benchmark
Alex Schepeljanski of Alex Intelligent Software develops the free AS SSD Benchmark utility for testing storage devices. The AS SSD Benchmark tests sequential read and write speeds, input/output operational performance, and response times.
AS-SSD Benchmark uses compressed data, so sequential file transfer speeds may be reported lower than with other tools using uncompressed data. For this reason, we will concentrate on the operational IOPS performance in this section.
The 960 PRO returned amazing read results on the 4K-64 thread portion of this benchmark– by far the highest we’ve ever recorded.
Samsung 960 PRO Test Results
As the chart below shows, the 960 PRO’s read performance on this section of the benchmark was astonishing, over three times better than the 950 PRO’s results. Samsung seems to have really done their homework with the new Polaris controller.
In the next section, Benchmark Reviews tests transfer rates using ATTO Disk Benchmark.
ATTO Disk Benchmark
The ATTO Disk Benchmark program is free, and offers a comprehensive set of test variables to work with. In terms of disk performance, it measures interface transfer rates at various intervals for a user-specified length and then reports read and write speeds for these spot-tests. There are some minor improvements made to the 2.46 version of the program that allow for test lengths up to 2GB, but all of our benchmarks are conducted with 256MB total length. ATTO Disk Benchmark requires that an active partition be set on the drive being tested. Please consider the results displayed by this benchmark to be basic bandwidth speed performance indicators.
512GB Samsung 960 PRO ATTO Benchmark Results
With a blistering 3.4GB peak read speed, the 960 PRO is the fastest storage device Benchmark Reviews has ever tested in this benchmark, besting even the 950 PRO RAID!
The 950 PRO RAID still wins the maximum write speed contest, though.
In the next section, Benchmark Reviews tests sequential performance using the CrystalDiskMark 3.0 software tool…
CrystalDiskMark 3.0 Tests
CrystalDiskMark 3.0 is a file transfer and operational bandwidth benchmark tool from Crystal Dew World that offers performance transfer speed results using sequential, 512KB random, and 4KB random samples. For our test results chart below, the 4KB 32-Queue Depth read and write performance was measured using a 1000MB space. CrystalDiskMark requires that an active partition be set on the drive being tested. Benchmark Reviews uses CrystalDiskMark to illustrate operational IOPS performance with multiple threads. In addition to our other tests, this benchmark allows us to determine operational bandwidth under heavy load.
CrystalDiskMark uses compressed data, so sequential file transfer speeds are reported lower than with other tools using uncompressed data. For this reason, we will concentrate on the operational IOPS performance in this section.
CrystalDiskMark 3.0 reports single-threaded sequential speeds reaching 1589MB/s reads and 1468MB/s writes. 4K tests at a queue depth of 32 produced 762MB/s read and 571MB/s write performance.
512GB Samsung 960 PRO SSD CrystalDiskMarkResults
The chart below summarizes 4K random transfer speeds with a command queue depth of 32. In this one section of the CrystalDiskmark test, the 960 PRO falls behind its predecessor in read speeds, although it overtakes it in write speeds. Still, note the staggering sequential read and write speeds in the CrystalDiskMark results above.
In the next section, we continue our testing using Iometer to measure input/output performance…
Iometer IOPS Performance
Iometer is an I/O subsystem measurement and characterization tool for single and clustered systems. Iometer does for a computer’s I/O subsystem what a dynamometer does for an engine: it measures performance under a controlled load. Iometer was originally developed by the Intel Corporation and formerly known as “Galileo”. Intel has discontinued work on Iometer, and has gifted it to the Open Source Development Lab (OSDL). There is currently a new version of Iometer in beta form, which adds several new test dimensions for SSDs.
Iometer is both a workload generator (that is, it performs I/O operations in order to stress the system) and a measurement tool (that is, it examines and records the performance of its I/O operations and their impact on the system). It can be configured to emulate the disk or network I/O load of any program or benchmark, or can be used to generate entirely synthetic I/O loads. It can generate and measure loads on single or multiple (networked) systems.
To measure random I/O response time as well as total I/O’s per second, Iometer is set to use 4KB file size chunks over a 100% random sequential distribution at a queue depth of 32 outstanding I/O’s per target. The tests are given a 50% read and 50% write distribution. While this pattern may not match traditional ‘server’ or ‘workstation’ profiles, it illustrates a single point of reference relative to our product field.
All of our SSD tests used Iometer 1.1.0 (build 08-Nov-2010) by Intel Corporation to measure IOPS performance. Iometer is configured to use 32 outstanding I/O’s per target and random 50/50 read/write distributionconfiguration: 4KB 100 Random 50-50 Read and Write.icf. The chart below illustrates combined random read and write IOPS over a 120-second Iometer test phase, where highest I/O total is preferred:
While random IOPS are significantly off the pace set by the Samsung 950 PRO, they’re more than double the rate set by the very best SATA drives.
In our next section, we test linear read and write bandwidth performance and compare the speed of the RD400 SSD against several other top storage products using the AIDA64 Disk Benchmark.
AIDA64 Disk Benchmark
Many enthusiasts are familiar with the Finalwire AIDA64 benchmark suite, but very few are aware of the Disk Benchmark tool available inside the program. The AIDA64 Disk Benchmark performs linear read and write bandwidth tests on each drive, and can be configured to use file chunk sizes up to 1MB (which speeds up testing and minimizes jitter in the waveform). Because of the full sector-by-sector nature of linear testing, Benchmark Reviews endorses this method for testing SSD products, as detailed in our Solid State Drive Benchmark Performance Testing article. One of the advantages SSDs have over traditional spinning-platter hard disks is much more consistent bandwidth: hard disk bandwidth drops off as the capacity draws linear read/write speed down into the inner-portion of the disk platter. AIDA64 Disk Benchmark does not require a partition to be present for testing, so all of our benchmarks are completed prior to drive formatting.
Linear disk benchmarks are superior bandwidth speed tools because they scan from the first physical sector to the last. A side affect of many linear write-performance test tools is that the data is erased as it writes to every sector on the drive. Normally this isn’t an issue, but it has been shown that partition table alignment will occasionally play a role in overall SSD performance (HDDs don’t suffer this problem).
512GB MZ-V6P512BW SSD Read Results
As shown above, average read performance was over 2.5 gigabytes per second, with one odd blip near 3GB/s.
AIDA64 linear write-to tests were next…
512GB Samsung 960 PRO SSD Write Results
Write performance was very even: often, with SSDs, you’ll see initial very high performance followed by a sudden, dramatic drop as the data written overflows an SLC cache or other buffer. Here, the drop from a maximum of 1.9GB/s to an average of about 1.8GB/s is pretty small.
Comparing the average linear read and write speeds of recently-tested SSDs, we see that the 960 PRO handily beats its 950 PRO forebear, but loses out to the 950 PRO RAID.
Linear tests are an important tool for comparing bandwidth speed between storage products, serve to highlight the consistent-bandwidth advantages of SSDs, which don’t suffer the performance drop-off that HDDs do as the test proceeds away from the fast outer edge of the disk.
In the next section we use PCMark Vantage to test real-world performance…
PCMark Vantage HDD Tests
PCMark Vantage is an objective hardware performance benchmark tool for PCs running 32- and 64-bit versions of Microsoft Windows 7. PCMark Vantage is well suited for benchmarking any type of Microsoft Windows 7 PC: from multimedia home entertainment systems and laptops, to dedicated workstations and high-end gaming rigs. Benchmark Reviews has decided to use the HDD Test Suite to demonstrate simulated real-world storage drive performance in this article.
PCMark Vantage runs eight different storage benchmarks, each with a specific purpose. Once testing is complete, results are given a PCMark score while and detailed results indicate actual transaction speeds. Since it simulates real-world consumer workloads, Vantage gives much more weight to read speeds, and fast iOPS are not as important as they would be in a server or other business environment. With an overall score of 99679 (Toshiba NVMe driver), the RD400 showed good performance, but as you can see from the chart below, the use of the Windows NVMe driver increases the overall score by over 60%.
In this benchmark, the 960 PRO returns an overall score that’s virtually identical to the 950 PRO, and substantially better than the 950 PRO RAID. Individual scores are shown below:
512GB Samsung 960 PRO PCIe SSD PCMark Vantage Results (Windows driver)
As you can see, in the PCMark Vantage benchmark, the Toshiba driver returns lower scores in every single test, and the difference is dramatic in some tests– for example, importing pictures is over three times faster using the Windows driver.
In the next section, I share my review conclusion and final product rating.
Samsung 960 PRO NVMe SSD Conclusion
IMPORTANT: Although the rating and final score mentioned in this conclusion are made to be as objective as possible, please be advised that every author perceives these factors differently at various points in time. While we each do our best to ensure that all aspects of the product are considered, there are often times unforeseen market conditions and manufacturer changes which occur after publication that could render our rating obsolete. Please do not base any purchase solely on our conclusion, as it represents our product rating specifically for the product tested which may differ from future versions.
Pity the poor hardware reviewer, who, upon having bestowed his highest accolades to other PCI-E SSDs just a few short months ago, is now faced with a dramatically faster product. While the performance increases we used to expect with each new generation of Intel processor have become quite modest, immeasurably so in some cases, PCI-E SSDs continue to raise the performance bar with innovative new controllers and NAND. Samsung’s leading the way here with their vertically-stacked “V NAND” chips, which allows them to produce a 2TB, single-sided m.2 drive, and their Polaris controller, which wrings immense performance out of the memory, without the stutters, lags, and dropoffs I’ve become used to seeing in SSD testing.
But getting the absolute top level of performance comes with a commensurate price: compared to a SATA 3 SSD, a 512GB Samsung 960 PRO commands a 20%-25% price premium, while at the terabyte level it’s more like 200%-250%. And it’s just silly at the 2TB level. But the Samsung 950 PRO has 300%-500% of the performance of a SATA SSD…on our benchmarks.
And therein lies the rub, as they say: to what degree do these amazing benchmark numbers translate into a perceptible, real-world advantage? And the answer is “Not that much”. Do you remember the first time you used an SSD based machine, and how lighting-fast it seemed next to a hard disk-based computer? You’re not going to get that “WOW” moment switching from a SATA SSD to a PCIE SSD, benchmark numbers notwithstanding.
That said, it’s still very impressive what Samsung has done with the 960 PRO. Their new Polaris controller, a 5-core chip with 8 data channels was introduced several months ago with their OEM SM961 drives, paired with their new 48-layer V-NAND chips, provides a dramatic performance increase over the previous generation 950 series, outgunning even a RAID-0 array of 950 PROs in some of our tests, without the performance problems the RAID array had with heavily-queued random accesses.
The top tiers of computer performance are always expensive, but remain a viable niche even though the real-world benefits are marginal: NVIDIA still manages to sell Titan X graphics cards, and Intel unapologetically charges over $1,500 for the Core i7-6950X consumer CPU. But while these items offer only marginal performance increases over their more mainstream stablemates, a m.2 PCIE SSD offers a solid 300%-500% performance improvement over a SATA SSD. My sole complain about this drive is that I’d like to see Samsung offer some migration utility for users moving their Windows installations from their current drives to this new drive.
The real-world performance benefits of the 960 PRO may not be perceptible for most users, but for the enthusiast who must have the very best, well, the Samsung 960 PRO is the very best.
At the time of this article, the Samsung 960 PRO NVM Express M.2 Solid State Drive was available online for: 512GB: $330 (Amazon|Newegg), 1TB: $630 (Amazon|Newegg), 2TB: $1300 (Amazon|Newegg). NOTE: prices for these drives are currently extremely volatile!
Pros:
+ Fastest. SSD. Ever.
+ 5-year warranty
Cons:
– Substantial price premium over SATA SSDs
– No included migration software
Ratings:
- Performance: 9.75
- Appearance: 8.25
- Construction: 9.75
- Functionality: 9.25
- Value: 8.00
Final Score: 9.0 out of 10.
Excellence Award: Benchmark Reviews Golden Tachometer Award.
COMMENT QUESTION: Which brand of SSD do you trust most?
4 thoughts on “Samsung 960 PRO NVMe SSD Review”
Which brand of SSD do you trust most?
Samsung. My 850 evo, is running with the speed of the ram, in rapid mode. That’s fast enaugh! 2900 mb/sec, seq.write , on a8-7600, chipset a88x, ram at 1800 mhz.
This was definitely an upgrade to the 256GB 950 Pro I had not all that much for the 950 Pro 512GB. Hopefully we eventually start to see price savings with the additional layers.
Confused. How is it that a 3 to 5 times “performance improvement” equates to imperceptible real world benefits. Won’t windows load faster? Won’t game zone load times be slashed?
If not, is it xpoint we should be expecting to achieve these goals?
Easy: because most of the data transfers you’ll make are quite small, in the kilobytes-to-hundreds of kilobytes range.
Say you’ve got a 250-kilobyte block of data to read. With a SATA SSD that can do 550 megabytes per second, you can get that data in about 1/2200th of a second.
Now with a spiffy new m.2 NVME SSD that can sustain 2.5 gigabytes per second, that data transfer take 1/10,000th of a second. Can you tell the difference, sitting in your chair?
Of course, there will aways be use cases where the difference _is_ perceptible. But booting Windows or loading game zones isn’t just about raw data transfer rates; there’s a lot of computation going on.
Comments are closed.