Crucial's latest SSD, the P1, is the company's first consumer drive with an NVMe interface and the second Crucial drive to use 3D Quad-Level Cell (QLC) , 4-bit per cell NAND (the first was the 5210 ION SATA enterprise drive). It uses a combination of Micron NAND and a Silicon Motion controller. At launch, the P1 range consists of just two capacities, 500GB and 1TB, but there is a flagship 2TB drive coming soon. Looking after the Micron 64-layer 3D QLC NAND is a Silicon Motion SM2263EN 4-channel controller, and in the case of the 1TB drive there is 1GB of DDR3 cache. Crucial quote Sequential performance figures for the 1TB drive as up to 2,000MB/s for reads and up to 1,700MB/s for writes. The 500GB drive is rated at 1,900MB/s reads and 905MB writes, while the 2TB drive is listed at the same 2,000MB/s reads as the 1TB drive but with a faster write figure of 1,750MB/s. Random 4K performance for the 1TB drive is stated as up to 170,000 IOPS for reads and 240,000 IOPS for writes. The 500GB drive is rated at 90,000 IOPS for reads and 220,000 IOPS for writes, while the 2TB drive has a 250,000 IOPS figure for both reads and writes. Crucial quotes an endurance figure for the 1TB drive of 200 TBW which works out at around 1,093GB of writes per day for the total length of the 5-year warranty Crucial backs the drive with. Quad-Level Cell NAND After SLC (1 bit per cell), MLC (2 bits per cell) and TLC (3 bits per cell), comes the latest leading edge NAND technology, Quad-Level Cell (QLC) with 4 bits per cell. While QLC NAND has obvious advantages in terms of storing more bits of data per cell (an increase of 33% over TLC ) and at lower cost, there are a couple of significant hurdles to overcome in terms of performance and endurance. Performance When NAND is programmed or erased, electrons are sent through an insulator by the application of a voltage. The location and number of electrons is the determining factor for when the current will flow between a source and a sink (voltage threshold) which determines what combination of 1s and 0s (data) is stored in the cell. With SLC NAND this is pretty straight forward as only a 1 or a 0 is being stored. QLC NAND ups the ante considerably as now there are four 0 or 1 states per cell, giving a total of 16 different possible combinations per cell which in turn means 16 different voltage levels to control. As more data is able to be stored in a cell there is a corresponding drop in performance as the cell contents must be read before any writes can commence as the previous bit value in the cell must be recognised first. The corresponding changes in voltage to enable the change in bit state causes instability in the surrounding cells. With QLC NAND's 16 different voltage levels this needs very careful programming to ensure adequate precision during read and write operations. QLC NAND supporting controllers have to have high level error correction capabilities to help overcome these problems, but all these actions mean slower NAND access than previous generations of NAND. To offset the drop in performance due to the length of time it takes to program a QLC NAND block, QLC drives are going to be relying on SLC caching technology a lot more than even TLC NAND does to raise the performance levels of the drive up to what you might expect of an NVMe drive. Endurance Every time data is written to NAND, the cell gets slightly damaged which is why NAND has a finite life or endurance. SLC has a life of around 100,000 P/E (program/erase cycles), MLC 10,000 P/E and TLC approximately 3,000 P/E (thanks to manufacturers striving to keep improving the life of TLC NAND). When QLC was first mooted, the endurance figure often quoted was just 100 P/E, a figure so low it caused many people to question whether the technology would ever see the light of day. As with TLC NAND, manufacturers have been working hard to get the P/E figure of QLC NAND up to more usable levels, Micron currently quotes a figure of 1,000 program/erase cycles. Physical Specifications: Usable Capacities: 1TB NAND Components: Micron 64-layer 3D QLC NAND NAND Controller: Silicon Motion SM2263 Cache: 1GB DDR3 Interface: NVMe 1.3 PCIe 3.0 x4 Form Factor: M.2 2280 Dimensions: 22 x 80 x mm Firmware Version: P3CR010. The drive comes in a box without any image on the drive on the front or back. On the bottom left hand side of the box front is the drive's capacity. The rear of the box has three icons for Data Transfer Software (Acronis True Image HD 2015), Installation Guide and Firmware update which are all available via Crucial’s SSD support web page. The only thing in the box beside the drive is a setup / warranty information pamphlet. Crucial's P1 is built on a standard M.2 2280 format with all the components housed on one side of the PCB. Sitting under the label (a plain label without any copper layer) are two packages of Micron 64-layer 3D QLC NAND, a single 1GB DDR3 IC and the Silicon Motion SM2263EN controller. Silicon Motion's SM2263EN is a four channel controller that supports the NVMe 1.3 specification and can be used in conjunction with DDR3, DDR3L, LPDDR3 or DDR4 cache. It also supports full drive encryption with AES and TCG Opal protocols. Crucial’s Storage Executive is a pretty comprehensive SSD toolkit. With it you can check the drive’s S.M.A.R.T data, update the firmware, see how the drive’s capacity is being used, monitor the drive’s operating temperature and overall health, adjust the Over Provisioning and enable the Momentum Cache feature. There's also some options in the software that aren't supported by the P1; resetting the PSID, Flex Capacity (improve's endurance) and Device Self-Test. For testing, the drives are all wiped and reset to factory settings by HDDerase V4. We try to use free or easily available programs and some real world testing so you can compare our findings against your own system. This is a good way to measure potential upgrade benefits. Main system: Intel Core i7-7700K with 16GB of DDR4-3200 RAM, Sapphire R9 390 Nitro and an Asus Prime Z270-A motherboard. Other drives Corsair Force MP500 480GB Corsair Force MP510 960GB Intel Optane SSD905P 480GB Intel SSD760p 512GB Kingston A1000 480GB Plextor M9Pe(Y) 512GB Plextor M8PeG 512GB PNY CS2030 240GB Samsung SSD970 EVO 2TB Samsung SSD970 PRO 1TB Samsung SSD960 PRO 2TB Samsung SSD960 EVO 1TB Toshiba OCZ RD400 512GB Western Digital Black NVMe 1TB Western Digital Black PCIe 512GB Software: Atto Disk Benchmark. CrystalMark 3.0.3. AS SSD. IOMeter. Futuremark PC Mark 8 All our results were achieved by running each test five times with every configuration this ensures that any glitches are removed from the results. Trim is confirmed as running by typing fsutil behavior query disabledeletenotify into the command line. A response of disabledeletenotify =0 confirms TRIM is active. CrystalDiskMark is a useful benchmark to measure theoretical performance levels of hard drives and SSDs. We are using v6.0. to test NVMe drives. In the CrystalDiskMark benchmark, Crucial's P1 doesn't perform too well at a deep queue depth of 32 - particularly when it comes to read performance. However, when it comes to operating at a queue depth where a great deal of the everyday workloads occur, namely a QD of 1, it performs very well. Looking at both benchmark results, it appears that the Silicon Motion controller doesn't have a preference of what type of data it's being asked to work with. The ATTO Disk Benchmark performance measurement tool is compatible with Microsoft Windows. Measure your storage systems performance with various transfer sizes and test lengths for reads and writes. Several options are available to customize your performance measurement including queue depth, overlapped I/O and even a comparison mode with the option to run continuously. Use ATTO Disk Benchmark to test any manufacturers RAID controllers, storage controllers, host adapters, hard drives and SSD drives and notice that ATTO products will consistently provide the highest level of performance to your storage. We are using version 3.5 for our NVMe disk tests. Crucial's official figures for the Sequential performance of the 1TB drive are up to 2,000MB/s for reads and up to 1,700MB/s for writes. As you can see from the ATTO benchmark results, the tested drive confirmed these figures with reads of 2,003MB/s and 1,714MB/s for writes. AS SSD is a great free tool designed just for benching Solid State Drives. It performs an array of sequential read and write tests, as well as random read and write tests with sequential access times over a portion of the drive. AS SSD includes a sub suite of benchmarks with various file pattern algorithms but this is difficult in trying to judge accurate performance figures. While the write score for the P1 is good for an entry level drive, it is however let down by its 901 read score. IOMeter is another open source synthetic benchmarking tool which is able to simulate the various loads placed on hard drive and solid state drive technology. There are many ways to measure the IOPS performance of a Solid State Drive, so our results will sometimes differ from manufacturer’s quoted ratings. We do test all drives in exactly the same way, so the results are directly comparable. We test 128KB Sequential read and write and random read and write 4k tests. The test setup’s for the tests are listed below. Each is run five times. 128KB Sequential Read / Write. Transfer Request Size: 128KB Span: 8GB Thread(s): 1, Outstanding I/O: 1-32 Test Run: 20 minutes per test 4K Sustained Random Read / Write. Transfer Request Size: 4KB Span: 80GB Thread(s): 4, Outstanding I/O: 1-32 Test Run: 20 minutes per test 4K Random 70/30 mix Read/Write. Transfer Request Size: 4KB Span: 80GB Reads: 70% Writes: 30% Thread(s): 4 Outstanding I/O: 2 – 32 Test Run: 20 minutes 128KB Sequential Read / Write Performance As with the ATTO benchmark, we could confirm Crucial's official Sequential read/write speeds of up to 2,000MB/s and up to 1,700MB/s respectively with our 128KB Sequential read/write tests, with tested scores of 2,012MB/s for reads and 1,729MB/s for writes. 128KB Sequential Read v QD A closer look at the Sequential read performance over various queue depths shows a marked improvement in performance at a queue depth of 2, as the drive rises several places in our result graph for that queue depth. 128KB Sequential Write v QD In the Sequential write test the drive was pretty consistent in its level of performance throughout the queue depths. 4KB Sustained Random Read v QD The Crucial P1 didn't perform too well in our sustained 4K random read test, sitting at the bottom of the comparison graphs in all but the QD1 test. 4KB Sustained Random Write v QD In marked contrast to the 4K random read test, the drive performed reasonably well in the 4K sustained random write tests. In our 70/30 read/write mix test, the drive's performance peaked at the QD8 mark at 70,402 IOPS before falling back to finish the test run at 63,949 IOPS. In our read throughput test, Crucial's P1 peaked at 1,705.45MB/s at the 4KB block mark before falling back to finish the test at 1,640.77MB/s. Writes peaked at the 2KB block size at 1,677.56MB/s before fading to finish the run at 1,561.92MB/s. In our throughput tests we couldn't quite match the peak maximum speeds of 2,000MB/s and 1,700MB/s for read and writes respectively as quoted by Crucial. However, the tested write speed was much closer to the official figure than the read speed. Futuremark’s PCMark 8 is a very good all round system benchmark but it’s Storage Consistency Test takes it to whole new level when testing SSD drives. It runs through four phases; Preconditioning, Degradation, Steady State, Recovery and finally Clean Up. During the Degradation, Steady State and Recovery phases it runs performance tests using the 10 software programs that form the backbone of PCMark 8; Adobe After Effects, Illustrator, InDesign, Photoshop Heavy and Photoshop Light, Microsoft Excel, PowerPoint, Word, Battlefield 3 and World of Warcraft. With some 18 phases of testing, this test can take many hours to run. Preconditioning The drive is written sequentially through up to the reported capacity with random data, write size of 256 × 512 = 131,072 bytes. This is done twice. Degradation Run writes of random size between 8 × 512 and 2048 × 512 bytes on random offsets for 10 minutes. It then runs a performance test. These two actions are then repeated 8 times and on each pass the duration of random writes is increased by 5 minutes. Steady State Run writes of random size between 8 × 512 and 2048 × 512 bytes on random offsets for final duration achieved in degradation phase. A performance test is then run. These actions are then re-run five times. Recovery The drive is idled for 5 minutes. Then a performance test is run. These actions are then repeated five times. Clean Up The drive is written through sequentially up to the reported capacity with zero data, write size of 256 × 512 = 131,072 bytes. Overall the P1's performance gets hit very hard during the Degradation and Steady State runs of the PCMark 8's Consistency test as the SLC buffer gets smaller as the drive is filled with data. However, the drive's recovery from the ordeal is very good indeed - although not consistent. PCMark 8’s Consistency test provides a huge amount of performance data, so here we’ve looked a little closer at how the Crucial P1 performs in each of the benchmarks test suites. Adobe Creative Cloud The bandwidth figures for all the test traces in the Adobe Creative CC test run are very low with the accompanying high latency figures. However, the drive's recovery is very good although somewhat erratic in the two Photoshop traces. Microsoft Office As is usual with the Office part of the test, it's the MS Word trace that gets hit hardest during the test run. But as with the other two MS Office traces, Excel and PowerPoint, the recovery for the Word trace is extremely good and consistent. Casual Gaming The P1's bandwidth performance during the two casual gaming tests was very erratic. The recovery of the drive for the Battlefield 3 trace was very good indeed, but the recovery of the World of Warcraft trace less so. Just like the Consistency test, PCMark 8’s Standard Storage test also saves a large amount of performance data. The default test runs through the test suite of 10 applications three times. Here we show the total bandwidth performance for each of the individual test suites for the third and final benchmark run. Crucial's P1 performs well in PCMark 8's Standard Storage test, with strong bandwidth figures for both Photoshop tests and the InDesign test run. For the long term performance stability test, we set the drive up to run a 20-minute 4K random test with a 30% write, 70% read split, at a Queue Depth of 256 over the entire disk. The 1TB Crucial P1 averaged 27,832 IOPS for the test with a performance stability of 82.1% - which is superb for a drive in this class. To test real life performance of a drive we use a mix of folder/file types and by using the FastCopy utility (which gives a time as well as MB/s result) we record the performance of drive reading from & writing to a 256GB Samsung SSD850 PRO. We use the following folder/file types: 100GB data file. 60GB ISO image. 60GB Steam folder – 29,521 files. 50GB File folder – 28,523 files. 12GB Movie folder – 24 files (mix of Blu-ray and 4K files). 10GB Photo folder – 621 files (mix of png, raw and jpeg images). 10GB Audio folder – 1,483 files (mix of mp3 and .flac files). 5GB (1.5bn pixel) photo. The Crucial P1 handled the larger file sizes of our real life file transfers well, apart from when it was reading the 100GB data file where the performance slumped to 191MB/s. To get a measure of how much faster PCIe NVMe drives are than standard SATA SSDs, we use the same files but transfer to and from a 512GB Toshiba OCZ RD400: Taking the standard SATA SSD out of the equation, in our real life file transfers using another NVMe drive to transfer the data to and from saw huge jumps in bandwidth performance, particularly when dealing with larger file sizes of some of our tests. It's been a long time coming but at long last Crucial has entered the NVMe SSD fray with the P1, a range of drives that forsakes MLC or TLC NAND to use the latest NAND evolution, 3D Quad-Level Cell (QLC) NAND supporting 4-bits per cell. At launch the P1 range consists of just two capacities, 500GB and 1000GB (1TB). Both these drives are built on a single-sided format but these will be joined shortly by a 2TB flagship drive that uses both sides of the PCB to house the NAND chips. Performance wise Crucial quotes official Sequential speeds for the 1TB drive at up to 2,000MB/s for reads and up to 1,700MB/s for writes. We could confirm these figures with the ATTO benchmark as our P1 sample produced a read figure of 2,003MB/s with writes coming in at 1,714MB/s. Random 4K performance for the 1TB drive is quoted as up to 170,000 IOPS for reads and up to 240,000 IOPS for writes. In our 4K testing we couldn't get anywhere near that read figure, the best we saw was 67,309 IOPS. On the other hand when it came to tested 4K random writes we got pretty near that maximum 240,000 IOPS figure, with 235,545 IOPS at a QD of 8. Like many of today's SSDs, Crucial's P1 uses an SLC buffer to boost performance but with QLC NAND this is needed more than ever. Crucial's caching solution is called Hybrid-Dynamic Write Acceleration technology. It's labelled Hybrid-Dynamic because as well as a static allocation, the technology can use up to 14% of the drive's capacity if it has the free space available. The problem is that the cache size shrinks from 100GB when the drive is empty to around 12GB as the drive fills with data, meaning there is a corresponding loss in performance. The drive's label doesn't contain a copper layer too, so the P1 does get rather toasty when really being pushed. Although it's unlikely to get pushed as hard in real life as it does during strenuous benchmarking, it might be an idea that to make sure there is sufficient cooling around the drive to safeguard against any thermal throttling. We don't yet have a buy link for the P1 but we have been told to expect pricing around £185. Pros Good performance while the drive is reasonably empty. Crucial Storage Executive Software. 5-year warranty. Cons Performance drops in relation to the drive being filled with data. Gets pretty hot when really pushed. KitGuru says: It's good to see Crucial finally enter the NVMe SSD market and it's interesting that they've chosen a QLC NAND equipped SSD drive to be their first model in the market space.
KitGuru KitGuru.net – Tech News | Hardware News | Hardware Reviews | IOS | Mobile | Gaming | Graphics Cards
