Through a quirk of SSD technology, the Crucial M4 64GB has limited write speeds compared to the larger M4 models. The question is whether it’s still a worthwhile upgrade for users with limited storage needs and/or budget restrictions.
September 13, 2012 by Lawrence Lee
| Product | Crucial M4 64GB CT064M4SSD2 2.5″ SSD |
| Manufacturer | Crucial Technology |
| Street Price | US$70 |
We’re currently in a sort of golden age for solid state drives with every well-known flash memory brand taking a crack at the market. With so many models available, selecting an SSD is arduous, but on the bright side, the intense competition has made them more affordable. The current crop of high performance consumer SSDs are faster, more reliable, and about half the price compared to those from a year ago. Despite the substantially lower pricing, they are still more expensive per byte than a standard magnetic hard drive by an order of magnitude.
That said, many users don’t need a lot of capacity for everyday use, assuming personal media like music and video, etc. are stored in the cloud or on secondary/shared hard drives in desktops, enclosures, or NAS type devices. A fully updated Windows 7 install uses about 15GB, give or take, and the most common applications are megabytes in size rather than gigabytes, with some notable exceptions like Microsoft Office, Adobe Photoshop, and of course, high-end games. In many cases, a small, cheap SSD like the Crucial M4 64GB, can make for a very nice upgrade.
Crucial first dipped their toe in the performance SSD market a couple of years ago with the acclaimed C300, one of the first SSDs to use the new SATA 6 Gbps interface. The M4 is the next iteration based mainly on the same architecture with a slightly updated Marvell 88SS9174 controller and smaller 25 nm NAND Flash chips rather than the 32 nm packages found in the C300.
Crucial M4 64GB (CT064M4SSD2): Specifications (from the product data sheet) | |
| NAND | Micron MLC |
| Controller | Integrated 8-channel single chip |
| Interface | SATA 6Gb/sec (compatible 3Gb/sec) |
| Average Access Time | < .1 ms |
| Sequential Read (up to) | 415 MB/sec (SATA 6Gb/sec) |
| Sequential Write (up to) | 95 MB/sec (SATA 6Gb/sec) |
| Random 4k READ | 40,000 IOPS |
| Random 4k WRITE | 20,000 IOPS |
| 2.5-inch SSD dimensions (L x W x H) | 100.5 x 69.85 x 9.50 mm |
| 2.5-inch SSD weight | 75g |
| MTBF | 1.2 Million Hours |
| Drive Endurance | 36TB=20GB per day for 5 years |
| Warranty | Limited 3 Year Warranty |
Looking at the specifications you will see some disappointing numbers. The write performance is listed as only 95 GB/s maximum for sequential writes and 20,000 IOPS for random writes, fairly pitiful for a SATA 6 Gbps drive. Some small capacity SSDs have poor performance because manufacturers burden them with more affordable controllers and/or slower NAND chips but this doesn’t apply to the M4. The larger versions have more respectable write speeds yet use the same internal components. The issue boils down to the design of the controller and the type of NAND chips used across the entire M4 line.
Crucial M4 Model Comparison: Performance Specifications | ||||
| Specification | 64GB | 128GB | 256GB | 512GB |
| Sequential Read (up to) | 415 MB/sec | 415 MB/sec | 415 MB/sec | 415 MB/sec |
| Sequential Write (up to) | 95 MB/sec | 175 MB/sec | 260 MB/sec | 260 MB/sec |
| Random 4k Read | 40,000 IOPS | 40,000 IOPS | 40,000 IOPS | 40,000 IOPS |
| Random 4k Write | 20,000 IOPS | 35,000 IOPS | 50,000 IOPS | 50,000 IOPS |
The M4 family uses 25 nm synchronous NAND modules with two 32Gb dies per chip for a combined size of 64Gb (8GB) per chip. So the 64GB model has 8 chips while the 128GB model sports 16. The Marvell controller has 8 channels that can be written to simultaneously and they’re all populated, one chip per channel in the 64GB version. Unfortunately this isn’t enough to to fully exploit each channel’s write potential — the 128GB model, with twice as many chips per channel, has specified write speeds about 80% higher, while the 256GB and 512MB versions are faster still. The question is how much does this effect the speed of the 64GB drive in real world use and whether it’s still a worthy upgrade over a magnetic hard drive.
|
|
Crucial M4 64GB (CT064M4SSD2) box and contents.|
|
The drive.There are three versions of the Crucial M4 64GB that use the familiar 2.5 inch 9.5 nm thick form factor (Crucial also offers two slim 7 mm models) with the only difference being the accessories. Like many manufacturers, they offer a desktop kit with a 3.5 inch drive adapter, a laptop kit with a SATA to USB adapter for data transfer and imaging, and a less expensive barebones, almost OEM model which is what we have today. All that’s included is a small instruction sheet and the drive itself in an antistatic bag, immobilized by a thin plastic shell.
TESTING
Our samples were tested according to our standard hard drive testing methodology. As of mid-2008, we have been conducting most acoustics tests in our own 10~11 dBA anechoic chamber, which results in more accurate, lower SPL readings than before, especially with <20 dBA@1m SPL.
Two forms of hard drive noise are measured:
- Airborne acoustics
- Vibration-induced noise.
These two types of noise impact the subjective
perception of hard drive noise differently depending on how and where the drive
is mounted.
Both forms of noise are evaluated objectively and
subjectively. Airborne acoustics are measured in our anechoic chamber using a lab reference
microphone and computer audio measurement system. Measurements are taken at a distance of one meter from the top
of the drive using an A-weighted filter. Vibration noise is rated on a scale
of 1-10 by comparing against our standard reference drives.
As of late-2011, we have been conducting performance testing. A combination of timed real-world tests is used to represent a workload of common activities for a boot drive including loading games, running disk-intensive applications, copying files, and installing programs. Synthetic tests are also run to better judge the performance across the entire span of the drive.
Summary of primary HDD testing tools:
- HD Tune Pro
– Benchmarking tool for storage devices. - CrystalDiskMark – Benchmarking tool for storage devices.
- SPCR’s Audio Audio
Recording/Analysis system using SpectraPlus
and other utilities - SPCR Anechoic Chamber
- Custom-built HDD power
measurement and Vibration test tools
Key Components in LGA1155 Heatsink Test Platform:
- Intel Core i5-2400 Sandy Bridge core, LGA1155, 3.1 GHz, 45nm, 95W TDP, set to 1.6 GHz to emphasize differences in the performance of storage devices.
- Intel DP67BG ATX motherboard.
P67 chipset. - Asus
EAH3450 Silent graphics card. - OCZ Platinum Extreme Low Voltage DDR3 memory. 2 x 2 GB, DDR3-1333 in dual channel.
- Seasonic X-400 SS-400FL
400W ATX power supply. Passively cooled - Microsoft
Windows 7 Ultimate operating system – 64-bit
Performance Test Tools:
- Call
of Duty: World At War – PC game - Far
Cry 2 – PC game - ExactFile –
file integrity verification tool - TrueCrypt
– file/disk encryption tool - 3DMark Vantage Installer
- Cyberlink PowerDVD 10 Installer
Benchmark Details
- Boot: Time elapsed between pressing the power button to the desktop and the Windows start sound playing (minus the average time to get to the “loading Windows” screen, 11 seconds on our test system)
- COD5: Combined load time for “Breaking Point” and “Black Cats” levels.
- Far Cry 2: Load time for one level.
- ExactFile: Creating a MD5 check file of our entire test suite folder.
- TrueCrypt: Creating a 10GB encrypted file container.
- 3DMark Vantage: Install time, longest interval between prompts.
- PowerDVD 10: Install time, longest interval between prompts.
- Small File Copy: Copy time for a variety of small HTML, JPEG, MP3, ZIP, and EXE files.
- Large File Copy: Copy time for 4 AVI files, 2 x 700MB and 2 x 1400MB
in size.
A final caveat: As with most reviews, our comments
are relevant to the samples we tested. Your sample may not be identical. There
are always some sample variances, and manufacturers also make changes without
telling everyone.
Ambient conditions at time of testing were 10.5 dBA and 22°C.
Real World Performance
A Windows 7 image loaded with our test suite was cloned to a 50GB partition
at the beginning of each drive after a complete format. Our entire
test suite was run start to finish three times with a defragmentation (SSDs and hybrid drives excluded) and reboot
between runs.
Average times were collected for comparison.
 |
In our loading tests, read speed is key and according to the specifications, the 64GB version of the M4 is just as fast in this regard as the larger models. There doesn’t appear to be any compromise in performance with the M4 64GB coming in second overall
 |
Our application tests showcased the disparity in read and write performance. In the ExactFile file integrity check test, it was as snappy as all our recently tested high performance SSDs. In the TrueCrypt test, where an encrypted file container is created (written to the drive), it was substantially slower.
 |
In our file copy test, the M4 64GB lagged behind the other SSDs tested by a noticeable margin. It was even defeated by the VelociRaptor 1TB, a high performance 10,000 RPM hard drive.
 |
The M4 produced a similar result in our installation tests, edged out all the models compared except the older VelociRaptor 600GB.
 |
To accurately represent the overall real world performance of the drives, we gave each model a proportional score in each benchmark series (loading, application, file copy, and installation) with each benchmark set equally weighted. The scale has been adjusted so that among the drives compared, a perfectly average model would score 100 points.
With its handicapped write speeds, the Crucial M4 64GB finished slightly behind the Corsair Force 180GB, a first generation SandForce drive from almost two years ago, though it did manage to beat the fastest hard drives available, the VelociRaptors from WD.
Synthetic Test Results
Though our timed benchmark tests do a fair job of simulating performance in real world situations, it doesn’t tell the whole story. Synthetic tests like HD Tune and CrystalDiskMark help fill the gap. Note: on SSDs, a full format was conducted before running these tests.
|
|
|
|
HD Tune’s main benchmark clearly illustrates the disproportionate read and write speeds. In sequential read performance, the M4 64GB came very close to the Kingston HyperX 240GB, the fastest SandForce drive we’ve tested. In sequential writes, its average speed was atrocious, well below 100 MB/s which is slower than current 7200 RPM hard drives. Access times were excellent though.
|
|
CrystalDiskMark uncovered more of the same. Using a block size of 512K and a random data set, its sequential and random read speeds were excellent, challenging the top SSDs compared. Writes speeds were behind by 40% to 65% depending on the drive.
|
|
Random read/writes with the smaller 4K block size gave us similar results as well. Good reads, poor writes.
Energy Efficiency
|
|
The energy efficiency of the Crucial M4 64GB was poor compared to other SSDs. Its power consumption was more typical of a 5400 RPM notebook hard drive. As a notebook upgrade, don’t expect any noticeable battery life improvement.
Noise
As solid state drives have no spinning platters or moving parts of any kind, they are effectively silent storage devices. It is possible that there could be a tiny bit of electronic noise (typically a high pitched squeal) being emitted, either intermittently depending on task, or continuously, but the Crucial M4 64GB was completely silent. In fact, the only SSD we’ve ever tested that made any audible noise was the Zalman S Series 128GB model which produced an odd high frequency squeal whenever it was accessed.
FINAL THOUGHTS
The Crucial M4 64GB is arranged with only one NAND chip connected to each of the eight channels provided by its Marvell controller, preventing each channel from reaching its full write potential. The obvious alternative to this strategy is to use only four channels with two NAND chips each but this would have decreased overall performance, not just writes. For SSDs, read speed is generally more important, particularly on small capacity models, so it’s hard to argue with their reasoning. Manufacturers want to use the latest, most densely-packed NAND chips for their high-end products and prefer uniform production lines using the same parts to keep costs down. Unfortunately this means that smaller SSD models sometimes end up getting the shaft.
If we look purely at the reads tests, the M4 64GB is fairly competitive with modern SSDs like the best SandForce SF-2281 drives and the Samsung 830 Series. However, the limited write speeds has a substantial effect on all around performance, making it look like a last generation product ported onto a SATA 6 Gbps interface to keep up appearances. Despite this, it’s still a stronger overall performer than any hard drive on the market, even 10,000 RPM models like the WD VelociRaptor. If you’re looking for an SSD primarily for quicker loading times, that is the one area at which the M4 64GB excels.
Unfortunately the pricing scale doesn’t favor low capacity SSDs either — the M4 64GB doesn’t provide much bang for the buck. Its US$70 street price is comparable to various 60GB SandForce drives but bigger SSDs are much cheaper per byte and faster as well. The larger 128GB version of the M4 for example, costs only $30 more and has much higher specified write speeds than that of the 64GB model. Unless your budget is severely restricted, it’s advisable to pay for the upgrade even if the extra capacity will largely go unused.
Many thanks to Crucial Technology for the M4 64GB solid state drive.
* * *
SPCR Articles of Related Interest:
Kingston HyperX 3K 240GB vs. Samsung 830 Series 128GB
Corsair Force GS 240GB: SandForce with Toggle-Mode NAND
Western Digital Red 3TB & 1TB Hard Drives
ADATA XPG SX910 128GB Solid State Drive
WD VelociRaptor 1TB and Scorpio Blue 500GB
Hitachi Deskstar 7K3000 2TB Hard Drive
* * *
