Crucial M4 vs Crucial C300 at 64 GB, 128 GB and 256 GB - BeHardware
>> SSD

Written by Marc Prieur

Published on May 2, 2011

URL: http://www.behardware.com/art/lire/832/


Page 1

The Crucial M4 range



When we published our first SSD report in September 2008, an Intel X25-M 80 GB cost around €500. Since then prices have certainly come down and Intel now markets its Intel SSD 320 Series 300 GB at the same price!

An entry level 128 GB SSD such as those tested in our latest article (SSD 2011 roundup: Crucial M4, OCZ Vertex 3, Intel 510/320) remains quite costly however, at €200. The only solution if you’re on a tight budget is to cut down on capacity, with a 64 GB model for example.


The problem is that only Crucial are currently marketing SATA 6G SSDs at this capacity, with the Vertex 3s and Intel SSD 510s only coming in at 120 GB. Is the Crucial solution a good one? We managed to get hold of an M4 64 GB so as to compare it to its predecessor and bring you up to date on the M4 range, versions 64, 128 and 256 GB.
The M4 range
Micron announced the RealSSD C400 at CES in January. A development of the RealSSD C300, this SSD is known as the M4, Crucial being Micron’s consumer arm. Launched one year earlier, the C300 stood out from other SSDs due to its adoption of the SATA 6 Gbits standard which opens the door to theoretical speeds of 600 MB/s, as against 300 MB/s for SATA 3 Gbits. Combined with aggressive pricing and high end random access performance, this was one of the flagship SSDs of 2010.

The M4 is also based on a Marvell 88SS9174 controller but in a more recent revision than the BJP2 used on the C300. In practice we compared a 256 GB version in the BLD2 revision with another at BKK2 (like the 64 and 128 GB models) and performance was identical. According to Crucial, it’s the firmware rather than the chip revision that makes the difference. The Marvell controller comes with a 256 MB DRAM cache, compared to 128 MB on the C300, and the memory used is now 25nm IMFT, while it was 34nm IMFT before.

Here are the performance figures announced by Crucial:


Read speeds are up 16.9% at equal capacity. Write speeds are up by between 26.7 and 20.9%. Although random writes are also up (+11.1 to 33%), the same can’t be said for random reads: 60,000 IOPs on the C300 but just 40,000 IOPS on the M4.


The Crucial M4 64 GB tested combines a Marvell 88SS9174-BKK2 controller, a Micron DRAM chip for the cache and 8 Micron 29F64G08CFACB flash chips. These chips are engraved at 25nm and combine two 32 Gb dies. The size of the pages and blocks is 4 KB and 1 MB respectively. It’s a surprise to find the BKK2 revision used here as the 128 and 256 GB versions use the apparently more recent BL02 revision. Nevertheless, according to Crucial, the firmware is what makes the difference between the C300 and the M4 and not the controller revision.


On the Crucial M4 128 GB, the DRAM chip from Micron is placed on the back and there are now 16 flash chips, though with the same reference.


Unlike the 64 and 128 GB versions, the Crucial M4 256 GB uses 29F128G08CFAAB flash chips. Their capacity is doubled as they use two 64 Gb dies. This time the pages are 8 Kb and the blocks 2 MB.

On the Crucial M4s, 6.9% of the flash memory is used for wear leveling and internal optimisations, which corresponds to the difference between the size of the flash chips (given on a basis of 1 KB = 1024 bytes) and SSD capacity (given on a basis of 1 KB = 1000 bytes).


Page 2
Sequential speeds

Test protocol
For this test we used the protocol set up for our latest SSD 2011 report that compared the Crucial M4s, OCZ Vertex 3s and Intel SSD Series 510 and 320 models. For more on the protocol we refer you to this page of the report.
Sequential speeds
We started the tests with sequential writes, measured using IOMeter. We tested accesses by 2 MB blocks, in reads and writes. As we have in the other pages of the report, we’ve chosen to display several performance graphs that can be consulted by moving the mouse over the links situated beneath the graph. Here you can view performances on an SATA 6 Gbps port and on an SATA 3 Gbps port.


[ SATA 6G ]  [ SATA 3G ]

As with the C300 range, read performance is fairly similar across the range of M4s whatever their capacity, with the gain on the previous generation varying between 19 and 24%. Write speeds on the other hand vary greatly according to capacity. The gains from the C300 and the M4 are significant, particularly for the low capacity models:

- 64 GB: +44%
- 128 GB: +33%
- 256 GB: +22%

While sequential writes are not the most important area for a 64 GB storage device, improvement of what was the only weak point on the C300 is definitely a positive. Using an SATA 3G interface strongly limits read performance, which drops to around 263-272 MB/s. In writes the impact is notable even on the C300 256, M4 128 and 256 MB models although these SSDs are far from saturating the theoretical bandwidth of the 300 MB/s interface.


Page 3
Random accesses

Random accesses
We then moved on to random accesses, looking at both reads and writes. Of course in standard usage, reads are far more common than writes and when it comes to a system disk, random accesses are of primary importance, especially in intensive multitasking. A storage device’s capacity to process a large number of random accesses, something that is directly linked to access times, is what will save it from slowing down the rest of the PC when it’s asked to access data situated in several places at the same time.

A standard hard drive with an access time of 10ms won’t manage more than 100 operations per second (IOPS), while SSDs have access times of between 0.13 and 0.23ms, or 7800 to 4300 IOPS. A huge difference! As in the previous test we measured random accesses on an Intel SATA 6G port and, in addition, on an Intel SATA 3G port.

The tests were carried out with 1, 2, 4, 8, 16 and even 32 simultaneous accesses (QD1 at 32 in the graphs). This gives us an understanding of an SSD’s capacity to process these accesses in parallel, with the ideal scenario being that performance is doubled between 1 and 2, 2 and 4 and so on. While it’s definitely worth checking this out, we shouldn’t lose sight of the fact that in standard usage, the level of simultaneous accesses is somewhere between 1 and 4.

Mo /s : [ 6G - Reads ]  [ 3G - Reads ]  [ 6G - Writes ]  [ 3G - Writes ]

IO/s : [ 6G - Reads ]  [ 3G - Writes]  [ 6G - Reads ]  [ 3G - Writes ]

Random reads are pretty much comparable across the C300 range, except with a very high level of simultaneous commands, where performance on the 64 GB version drops off. The Crucial M4s are slower overall here than the C300s and although the 64 GB and 128 GB versions perform at similar levels (except at QD16 where the C300 64 GB is also down on the 128 GB version), the 256 GB version is slower.

The first reason for this is that 25nm IMFT memory is slower than 34nm IMFT when it comes to reading a page - 65µs compared to 75µs. This explains the drop in performance on the M4 64 and 128 GB models in comparison to the C300 64 and 128 GB versions.

Performance on the 256 GB version can be explained by the difference in architecture of the memory chips used. On the 64 and 128 GB versions, the chips used are organised with pages – the smallest readable unit – of 4 KB, whereas the 256 GB version is constituted of 8 KB pages. To read the 4 KB blocks required in this test, the 256 GB version therefore has to access a full 8 KB page.

Using the SATA 3G interface does of course have a negative impact on performance but not as big an impact as with sequential writes as absolute speeds are lower.

Write performance is better than read performance. When you write data randomly to an SSD, the SSD actually writes the data sequentially, while reorganising its internal allocation table to make the addresses known by the OS on the storage system (the LBAs) correspond to the right memory cells. The Crucial M4s are a good deal faster here than the C300s and the higher the SSD capacity the better the level of performance with several simultaneous commands. As with sequential writes, speeds are limited by the SATA 3G interface although we’re far from its theoretical maximum.


Page 4
Practical tests: Files

Practical tests: Files
Moving onto the practical tests, we started by looking at write and read speeds for various groups of files. These groups were composed in the following way:

- Extra large: 731.17 MB on average
- Large files: 5.2 MB on average
- Medium sized files: 800.88 KB on average
- Small files: 48.78 KB on average

We used a RamDisk as the source or the target for reads or writes on the SSD. In view of the rapidity of recent SSDs and so as to obtain results that are subject to as little variation as possible, we used Robocopy with an in-house piece of software which enables us to carry the tests out repeatedly.


[ 6G - Reads ]  [ 3G - Reads ]  [ 6G - Writes ]  [ 3G - Writes ]

In reads all the C300s are on a par. Slower with small files, the M4s come back into things with medium sized files and then take the lead with large and extra large files, with speeds of up to 355 MB/s on the 64 and 128 GB versions. The 256 GB version is the slowest of the M4s overall with slightly lower performance on reads of small, medium and extra large files. Switching to 3G has an impact on the speeds of all transfers but is most noticeable at highest speeds on the largest files.

As expected, write performance varies greatly according to SSD capacity. The M4 64 GB has it over the C300 64 GB in all areas but the M4 128 and 256 GB models fall behind the C300 128 and 256 GB SSDs when it comes to writes of small files. They are faster in the other cases. This time using the SATA 3G interface only has a limited impact but strangely the M4 128 is more affected than the C300 256. The M4 256 GB is most impacted by switching to SATA 3G as it gives the highest performance.


Page 5
Practical tests: Applications

Practical tests: Applications
To finish with we carried out the purely practical tests, namely various timed operations after installation of Windows 7 (64 bit) on each of the SSDs:

- Boot Windows 7
- Start up 3D Studio Max
- Start up 3D Studio Max + Photoshop + Word + Excel
- Launch of a game in Civilization V
- Launch of a game in Crysis 2
- Installation of Photoshop CS 5

We did our best to limit the variations as much as we could, however as the timing was carried out by hand and the results themselves can vary slightly, there is a definitely a small margin of error here. As you can see, the results for the various SSDs are very close one to the next and there was no point in trying to test for the differences between SATA 3G and SATA 6G: therefore only the results at 6G are given.

For comparison purposes we added a hard drive, the very rapid Western Digital Caviar Black 2 TB, and the score obtained on an 8 GB RamDisk! With sequential speeds measured at over 5 GB/s and 100,000 IOPS in random 4KB accesses (QD1), the RamDisk is 10-15 x faster than the best SSD!


Windows 7 start-up was timed from the appearance of the load screen to full load of desktop (hourglass disappears). The results are very close, with just 0.2 seconds between the slowest and the fastest SSD. This is much faster than a hard drive.

The results for launching 3D Studio Max shows a small advantage for the C300s over the M4s, whatever the capacity, which is probably due to the faster 4K random accesses on the C300s. The M4s offer a very good level of performance, in among the leading group of those in our latest report.

It’s in combining the launch of 3D Studio Max, Photoshop, Word and Excel, where the storage device is pushed hardest in terms of intensive multitasking, that the gaps in comparison to a standard hard drive are the most significant. This combined launch is 37.2s slower than 3DS alone on the HDD, but in the worst case scenario adds just 0.8s on an SSD! Note once again a slight advantage for the C300s over the M4, the M4 256 GB being the slowest.

The RamDisk does give peformance gains but they remain quite low given the gigantic difference in performance on paper in comparison to SSDs. Accessing data in double quick time is all very well but when it doesn’t simply consist of reads/writes, this data still has to be processed by the system!



Page 6
Practical tests: Applications (cont.)

Practical tests: Applications (cont)

Now we move on to loading saved games, using Crysis 2 and Civilization V. The differences between the SSDs are very small and the gain (while it does exist) slight over a standard hard drive. Loading data quickly is all well and good but you still have to process it and this is what takes most time when you load up a game already underway. The gain offered by a Ramdisk is moreover quite slight given its level of performance.

Of course SSDs do reduce any in-game load times, definitely a positive. In some games such as WoW, they also make quite a difference in terms of load times, but we weren’t able to include this title in our test because of the random nature of loads linked to the fact that the number of characters in a certain place can vary.


This brings us to the final test, a timed installation of Photoshop CS5 from the archive downloaded from the Adobe site. This installation is split into two stages, decompression of the archive and then the installation itself. Note here there is some difference between the SSDs, notably the performance of the C300 64 GB drops off and is down on the hard drive because of its reduced sequential write speeds. The M4 64 GB does better, with the 128 and 256 GB versions even further out front. Once again, the Ramdisk adds very little.


Page 7
Energy consumption

Energy consumption
Finally we measured energy consumption with a clip ammeter. Energy consumption was taken at idle and in one of the most intensive SSD loads, sequential writes.


At idle, the M4s draw less power than the C300s, though only slightly. In load, the M4 64 and 128 GB models draw more power than their C300 equivalents. On the other hand, the M4 256 GB does very well and is on a par with the 128 GB version and therefore with significantly lower consumption than the C300 256 GB. However these figures need to be put together with write speeds to give a complete view of power draw. The following graph shows energy consumption for each 100 MB written:


By combining write speed and energy consumption during writes, you can see that the M4s are more efficient than the C300s across the board (at identical capacities). The M4 256 GB benefits from using chips with two 64 Gb dies, while the 128 and 64 GB versions use chips with two 32 Gb dies.


Page 8
Conclusion

Conclusion
Although SSDs are now much more affordable than in the past with most sales around the €100 – 120 marker, you wouldn’t know this from the fact that manufacturers constantly send us the 256 or 300 GB versions of their latest models for testing. By bringing out a 64 GB version of its M4, Crucial is for the moment the only manufacturer to address this price band, with no affordable Vertex 3 or Intel SSD 510 models yet available.

As we’ve seen in our practical tests of the 64 GB versions, neither the C300 nor the M4 models do badly when compared with the higher capacity versions. While the C300 suffered at times from its low seqential write speeds (73 MB/s), the M4 gives another 44% here (106 MB/s) making it a small capacity SSD with no major drawbacks.


When it comes to a disk system, a Crucial M4 64 GB will therefore do just as well as a 128 or 256 GB version, as long as there’s space for all your data! It may well be expedient to combine it with a hard drive for the storage of your photo, audio and video media or the games you use least, given the storage space such items require.

This article also gave us a comparison across almost the whole Crucial C300 and Crucial M4 ranges, with the exception of the M4 512 GB. Those who were expecting this new generation to give a substantial gain in performance will be disappointed. While sequential reads and writes are up, random accesses are only superior when it comes to writes. Random reads are down, especially for the 256 GB version. This is down to the change in structure linked to the introduction of 25nm 8 GB dies, something we identified when they were announced at the beginning of 2010.

In practice however, these differences do not transpire as current SSDs simply no longer constitute the machine’s bottleneck. In almost all situations they can supply the CPU with data more quickly than it can process it (pure file copying being excluded).

The only area in which improvement can be expected is therefore in terms of the capacity / price ratio. 25nm memory hasn’t yet fulfilled its promise here, as while the M4s aren’t more expensive than the C300s, in contrast to the pricing by the competition, we’re still waiting for Crucial to reduce M4 pricing in the medium term. The announcement of 20nm MLC flash NAND from IMFT and 19nm from Toshiba over the last few weeks should moreover ensure that the capacity / price ratio race continues into 2012.


Copyright © 1997-2014 BeHardware. All rights reserved.