Crucial M4 256GB SSD Review
Test system and software
Intel Core i5-2500K @ 4.8GHz
Be Quiet Dark Rock Pro C1 CPU cooler
4GB DDR3 RAM @ 1600MHz
AMD Radeon HD5670
OCZ Nocti 64GB mSATA
Corsair Performance Pro 256GB
OCZ Agility 3 240GB
OCZ Octane 512GB
Corsair Neutron 240GB
Crucial M4 256GB
Microsoft® Windows 7 Ultimate x64 SP1
ATTO disk Benchmark
CrystalMark 3.0 x64
Blackmagic Disk Speed Test
Autodesk AutoCAD 2012
To ensure the repeatability of the drive’s performance and verify our results, each benchmark was repeated three times and the median result is being displayed in this review.
Real world performance
In this test we have created three file folder copy scenarios and we are copying them between two folders within the drive, recording the time taken for the file transfer to be completed. Our first folder is filled with 870 picture files totaling 1GB, the second folder is filled with 550 MP3 files totaling 2GB and the third folder is home to a single 4.5GB ISO file.
Despite its rather "slow" write speed of 260MB/s, the Crucial M4 actually did quite well in this test. The synchronous NAND flash allows the Marvell controller to handle incompressible data with ease, allowing the M4 to perform very well against competitive products and controllers. Only when handling single large files the Crucial M4 falls notably behind the latest Link-A-Media based Neutron drive.
In our second real world test we are recording the time taken for our OS to fully boot from when we pressed the power on button and the start up times of two very popular and heavy professional applications, the Autodesk AutoCAD 2012 and the Mathworks Matlab. These tests are excellent examples of how an actual system would start and work under real-life circumstances.
The start-up times of the Crucial M4 drive were very good and on par with the rest of the SSD drives, although not notable better either. Nonetheless, it is almost miraculous to see a drive which has been released nearly two years ago competing with designs which are only a few months old.
ATTO / CrystalMark
ATTO Disk Benchmark is an old and simple but still very popular synthetic benchmark used to test the maximum theoretical speed of any storage devices, from RAID controllers and host adapters to HDD and SSD drives.
Being a synthetic benchmark, ATTO only displays the maximum possible speeds a drive may achieve when handling specific strings of data; in our case, compressible 512KB-4096KB packages. This is a good benchmark to examine if the performance of a drive is on par with the manufacturer’s specifications. The performance of the Crucial M4 drive is right on par with the manufacturer's specifications, with the drive delivering up to 506MB/s read and 259MB/s write speeds.
CrystalMark 3.0 x64 is another simple but popular synthetic benchmark used to measure the maximum theoretical speed of hard disk and solid state drives.
The 0×00 fill mode (compressible) mode of the test confirms the test results we received from the ATTO benchmark, with the Crucial M4 SSD displaying only slightly better performance figures; the difference is though well within a statistical error margin. The distinct advantage of the Marvell-based M4 and its synchronous NAND flash is obvious during the actual test, where incompressible data is included and drives based on controllers which rely on data compression take a massive performance hit.
AS SSD / Blackmagic DST
AS SSD is a free piece of software designed to benchmark the performance of SSD drives. It performs two sequential and two random read and write tests and then issues a score.
The Crucial M4 managed to bring a score of 242 over the read test and 310 over the write test, figures which are still very respectable, even by today's standards. However, even though the well-designed Marvell controller and synchronous NAND flash still give the M4 a distinct advantage over value-range SSD drives, the two year old drive can hardly complete with the newer models and especially those based on recently released controllers.
Black Magic Design make a number of industry leading video capture cards, including those which can capture HD material at very high data rates. As a result having a disk with a fast write speed is essential. To aid users in configuring their system for optimal captures Black Magic includes a utility with their hardware which records the real world disk speed and gives the maximum possible framerate for that speed.
In this test, raw read and write performance is all that matters and the benchmark tests the maximum speed a drive can reach while handling compressible data. Naturally, read and write performance figures are on par with the specifications of the drive and the Crucial M4 can read up to 55 and write up to 29 1080 12 bit RGB frames per second respectively.
IOMeter / SiSoft Sandra
IOMeter is another open source synthetic benchmarking tool which is an I/O subsystem measurement and characterization tool for single and clustered systems, originally developed by Intel Corporation who ultimately discontinued their work on the software which today is a free and open source application.
The 4K aligned benchmark of the IOMeter has been "Sandforce territory" for years, as the Sandforce controllers were optimized for this particular type of performance. Despite that, the Marvell-based Crucial M4 did very well in the 4K aligned test as well, yet its particular performance advantage was obvious during the random test, delivering many times better performance than value-oriented drives.
SiSoft Sandra is a very popular software suite which can also be used as a synthetic all-around benchmarking tool. We are using two of the suite’s tests in our article, the File Systems benchmark and the Physical Disks Benchmark.
While most drives are unable to reach their specified performance figures in this test, the Crucial M4 is one of the few which reached a full 500MB/s read speed, exactly the same performance figure as on the drive's specifications.
We perform our power consumption test by installing two digital USB multimeters working as ammeters on the +5V and +12V power lines connected to the drives, checking their current drain at idle and while stressed.
Marvell-based drives usually consume a little more energy than drives using other controllers. The on-board DRAM cache also has a negative impact, even if only minor. Yet, the consumption of the Crucial M4 is measurably lower than that of the Corsair Performance Pro SSD, which also is based on a Marvell controller, placing its energy consumption virtually on par with all other SSD drives.