by Aleksey Meyev
04/16/2010 | 02:45 PM
We are going to digress from our tradition of comparing hard disk drives of the same capacity within a single review. Not long ago we wrote about 1-terabyte and then about 2-terabyte HDDs, so we already know HDDs of these capacities well enough. But as is often the case, soon after those reviews we came across a number of interesting products. They are too few, however, for us to write two independent reviews, but we do want to tell you about them. That’s why there are both 1- and 2-terabyte HDDs in this article.
Our list of new and interesting HDDs begins with the latest generation of Hitachi’s products. We discussed one model from it in our recent roundup of 1TB drives, so what is so exciting about this one? The fact is Hitachi’s basic firmware for its latest generation of HDDs (both 7K1000.C and 7K2000 series) had proved to be rather poor but the company must be given credit for quickly releasing an update and the HDD in our last review had the improved firmware (version 39C) rather than the basic one (version 25C). Alas, the update was not blameless, either. Recently, we have got a sample with firmware 3EA which proves to be very different from the basic firmware. Thus, we will have the opportunity to see what effect firmware may have on an HDD’s performance and will also give the Deskstar 7K1000.C the chance to improve our opinion of it.
Next goes Samsung’s 2TB drive with 32MB cache from the 5400RPM EcoGreen F3 series. This HDD is interesting for two reasons. First, it is the company’s first 4-platter HDD. It is an achievement, really. Samsung’s previous-generation HDDs used to be always inferior to their opponents in maximum capacity because they had no more than 3 platters whereas most of Samsung’s opponents could use a 4-platter design (and Hitachi had even inherited IBM’s secret knowledge of how to make 5-platter HDDs). Although not a big deal for most users, this inferiority was a serious marketing problem. But it’s not a problem anymore now that Samsung can make maximum-capacity HDDs. The HDD is a power-efficient model as yet, but we hope that 4-platter 7200RPM models from Samsung will follow soon.
Another interesting thing about this product is that the announcing press release came out much later than the HDD itself had hit the shops. Of course, this approach is better than the paper releases of some other computer firms, but an ideal timing of both the actual release and the announcement would be best.
This model was tested in our labs for our very first roundup of 2-terabyte HDDs. It is here for the sake of comparison. To remind you, this HDD has 32 megabytes of cache and a spindle rotation speed of 5900 RPM which is somewhat higher than that of other power-efficient products.
The Barracuda XT is yet another drive we have already tested in our earlier reviews. We take it once again because it is a large-capacity and future-proof model thanks to its SATA 600 interface and 64 megabytes of cache. According to our tests, the new interface is currently nothing more but a marketing feature, though. Moreover, SATA 600 is implemented in today’s mainboards with discrete controllers which are so far from ideal that HDDs connected to them prove to be inferior to HDDs connected to Intel’s South Bridge. We have decided not to benchmark HDDs in SATA 600 mode until the new standard is implemented in mainboard chipsets.
We have not tested this HDD before although it was announced about one year ago. Seagate’s Constellation is the company’s new server-oriented series with a spindle rotation speed of 7200 RPM. Take note that the Constellation series includes 3.5-inch (marked with the suffix ES) as well as 2.5-inch (without “ES”) models.
Seagate has sharply separated its desktop and server HDD series, so we wonder if they are going to have different or, as it was the case with the Barracudas, similar firmware.
The Constellation ES series features 500GB platters and a maximum capacity of 2 terabytes, but the cache buffer is only 16 megabytes large. A server HDD does not need much cache, however, and does not rely much on it, especially at writing, to lower the risk of data loss. These HDDs are available with either SATA 300 (as our model) or SAS 600 interface.
Western Digital keeps on giving us something new on a regular basis. The more new products the company offers, the more market segments it covers and the more frequently its name is repeated in the news.
The latest piece of news from Western Digital is the update of the Caviar Black series. It now includes a 1-terabyte model with a 64MB buffer and a SATA 600 interface. Yes, this is WD’s answer to Seagate’s Barracuda XT. We have already seen the 64MB cache buffer in the 2TB model but the new interface is indeed a unique feature of this HDD (there is also the recently updated VelociRaptor series but we will talk about it in more detail in an upcoming review). The new HDD has only two platters and, apparently, a combined heads actuator like in the 2TB model (it means that besides the main electro-mechanical actuator, there is a secondary piezoelectric actuator to fine-tune the head’s position in a narrow (a few tracks wide) zone).
We’ve got three older drives from Western Digital to compare the new one with. They come from different subseries (J7, E8 and E3). Our practice suggests that WD’s subseries can differ a lot, so we will see what difference we have with the 7200RPM products. By the way, we were trying the different subseries to find a model with two rather than three platters – we had a suspicion that such models had come to retail already.
And finally, here is a 2TB model from WD’s power-efficient series. It belongs to the new generation of HDDs with 64MB cache and 4KB sectors (Western Digital calls this “Advanced Format”). As this technology provokes a number of questions and hot arguments, we’d like to briefly repeat the facts we said in our first overview of it.
The Advanced Format is a variant of the Long Data Sector technology that describes the transition of HDDs from 512-byte to 4-kilobyte sectors. Storing 4 KB of data will now require one new sector instead of eight older sectors. As a result, the same amount of data takes less space on the platters while the ECC field gets larger, which means that the HDD can store more data and is more reliable. Western Digital’s Advanced Format implementation supports full emulation of 512-byte sectors: for any electronic devices communicating with the HDD, the latter is represented as having 512-byte sectors, but its platters are actually formatted in 4KB sectors each of which contains eight virtual 512-byte sectors. All the required address translations are performed by the HDD’s electronics and are no concern of the user.
The biggest downside of this emulation is its interaction with Windows XP which, when formatting a hard disk, reserves the first 63 sectors (512 bytes each) and begins the partition at sector 64. As a result, all requests in 4KB data blocks prove to be shifted by 512 bytes relative to the hard disk’s sectors (the real sectors, not the emulated ones), provoking a performance hit at writing. Instead of just writing a single block of data, the HDD has to read two blocks, modify them, and only then write them to the platter.
This problem can be solved in two ways. You can close the HDD’s pins 7 and 8 with a jumper to automatically shift the whole logical structure by 512 bytes. Or you can use the WD Align utility which can be downloaded from the Western Digital website. This tool shifts the already existing partitions on the HDD, aligning its logical structure to the physical sectors.
This alignment provokes some confusion, so here are the facts you should know:
A small note: WD Align has a protection mechanism and will not allow you to align a partition if that is not necessary. So if you have any doubts, use WD Align and forget about the jumper. Do not insert the jumper after WD Align has aligned the drive (or refused to do so because the HDD is all right under Windows Vista/7).
The firmware versions of the tested HDDs are listed in the next table:
You should keep it in mind that the same models of HDDs may perform differently with other firmware.
The following testing utilities were used:
The HDDs were tested with the generic OS drivers. We formatted them in FAT32 and NTFS as one partition with the default cluster size. The HDDs were connected to a mainboard port and worked with enabled AHCI.
As for the WD20EARS with 4KB sectors, we already know how an unaligned HDD with such long physical sectors might perform because we tested the 1-terabyte model. Therefore in this test session disk partitions are aligned for this HDD with WD Align in each test. When benchmarking it in IOMeter’s tests that involve writing, we have two test passes: one is standard (we call it “unaligned”) and the other is “aligned” (the LBA addresses of requests are multiplies of 4 KB and match the HDD’s physical structure ideally). The difference is quite noticeable as you will see. We do not align those tests that do not require that, e.g. the test of sequential reading (aligning 2KB requests to 4KB sectors does not make sense) or multithreaded operations (the request size is 64KB).
We use our internal IOMark tool for low-level tests. Let’s begin with sequential reading.
Let’s compare the drives according to the speed at the beginning and end of the full-capacity partitions created on them.
Seagate carries on its tradition of winning this test, its HDDs being somewhat faster than their opponents with same-density platters. The Barracuda LP probably owes its result to the increased spindle rotation speed, though. Once again we can see that the power-efficient HDDs with 500GB platters are as good as 7200RPM models with 333GB platters (represented by the Caviar Black J7 and E8). Take note of the E3 – this is a dual-platter Caviar Black! Of course, the new FAEX has two platters, too (its low top speed is the results of an unlucky head/platter pair), but we haven’t had any doubts about its design.
And what about the speed of working with the cache buffer?
Hitachi Deskstar 7K1000.C, 1 TB, firmware version 25C
Hitachi Deskstar 7K1000.C, 1 TB, firmware version 3EA
Hitachi’s new firmware improves performance considerably. The HDD now produces smoother graphs without any fluctuations. There is only one shortcoming left: the HDD can do burst reading only when the data block size is a multiple of 256 sectors plus 32 sectors.
Samsung Ecogreen F3, 2TB
The Samsung draws neat and smooth graphs. It can write to the buffer almost as fast as read from it.
Seagate Barracuda LP, 2TB
Seagate Barracuda XT, 2TB
Seagate Constellation ES, 1TB
Seagate’s HDDs differ in this test. The Constellation ES and Barracuda XT deliver smoother graphs but the Barracuda XT has a higher speed with small data blocks. The latter has some problems at burst writing of small data blocks, however, which are the typical load for a hard disk drive.
Western Digital Caviar Black, 1 TB (FAEX Z3)
Western Digital Caviar Black, 1 TB (J7)
Western Digital Caviar Black, 1 TB (E8)
Western Digital Caviar Black, 1 TB (E3)
Western Digital Caviar Green, 2 TB
Well, we’ve finally got some changes in Western Digital’s camp. The dual-platter Caviar Black models both show that their firmware has changed. The HDDs have become faster with large data blocks.
The Hitachi delivers the highest top speed when reading from the buffer (unfortunately, only with data blocks of certain sizes). The HDDs from Samsung and Seagate are the best in terms of writing to the buffer.
IOMeter is sending a stream of read and write requests with a request queue depth of 4. The size of the requested data block is changed each minute, so that we could see the dependence of the drive’s sequential read/write speed on the size of the data block. This test is indicative of the maximum speed the drive can achieve.
The numeric data can be viewed in tables. We will be discussing graphs and diagrams.
There is no struggle in the sequential read test. Seagate’s HDDs are a little (by 10 MBps) faster than same-class opponents in terms of top speed and are slightly ahead at lower loads, too. The Hitachi falls behind the others when processing small data blocks. This seems to be due to a low-performance processor rather than to firmware problems: it is no secret that the HDD processor’ performance is the bottleneck when doing sequential reading in small data blocks (by the way, the load on the main CPU is the highest then, too).
We’ve got the same picture at writing as at reading.
For 10 minutes IOMeter is sending a stream of requests to read and write 512-byte data blocks with a request queue of 1. The total of requests processed by each HDD is much larger than its cache, so we get a sustained response time that doesn’t depend on the HDD’s buffer size.
It is simple with the read response time parameter: the Caviar Black series takes all the places on the podium. The leaders differ between each other, though. The newer E8 is somewhat better than the J7 and takes first place. The new dual-platter models are somewhat worse, the E3’s response being nearly 1 millisecond higher. Interestingly, the dual-platter FAEX is a quarter of a second faster than the similarly designed E3 and seems to really have new read/write heads. We will be sure about that after we test the drive in small traces and using a narrow test zone where the auxiliary piezoelectric actuator is going to deliver higher performance benefits.
Seagate’s Constellation NS is good. This company’s products have not shown such a low response time for long. The Barracuda XT is, on the contrary, poor, being even inferior to the two power-efficient models.
The WD Caviar Green is a disappointment. Its response of 20 milliseconds is too much even for a drive with a reduced spindle speed. We can’t really think of an explanation as its predecessor with ordinary sectors and its 1-terabyte cousin looked better in this test. Perhaps we’ve got a defective sample or a drive from some special series optimized for quiet operation (by lowering the speed of the heads). Whatever the reason, this model will be a poor choice as a system disk or a disk to store a database on. It can do nicely as a passive storage device which only has to deliver a normal sequential speed and consume as little power as possible.
We also have interesting results at writing. WD’s dual-platter Caviar Black have a higher response here, indicating a reduced efficiency of their deferred writing. And comparing Seagate’s HDDs, we can see that the Barracuda LP uses deferred writing very sparingly. The WD Caviar Green behaves in a stable manner, delivering a response time of 20 milliseconds again. This result can be explained easily, though. Our test load operates with 512-byte sectors and this HDD, due to the emulation, has to first read a 4KB data block, modify 512 bytes in it, and then perform a write operation for each write request! The saddest thing is that operating systems perform most of their service operations in such small requests to minimize the risk of losing important data.
Next goes the test of average positioning speed. The drive is being bombarded with read requests like in the response time test, and we calculate the difference between the LBA addresses of the previous and next requests and divide it by the time it took to perform the request. In other words, we have the distance (in gigabytes) the drive’s heads can run through in 1 second.
Besides response time proper, the number of platters is important for this test. The more platters the drive has, the longer distance its heads can run. As you can see, the four-platter 2-terabyte HDDs occupy the podium, ordered according to the speed of their heads. Next go the old 3-platter Caviar Black. Having more platters, they are equal to (or even a little faster than) their higher-density cousins of the newer generation. The Hitachi is poor in this test as it has few platters as well as sluggish heads.
Now we will see the dependence between the drives’ performance in random read and write modes on the size of the data block.
We will discuss the results in two ways. For small-size data chunks we will draw graphs showing the dependence of the amount of operations per second on the data chunk size. For large chunks we will compare performance basing on data-transfer rate in megabytes per second.
Random-address reading produced expected results both with small and large data blocks. There is no need for detailed comments. Each drive’s performance with small data blocks depends on its response time whereas at data blocks of 2 megabytes and larger the sequential read speed becomes the crucial factor.
It is the drive’s response time that largely determines its performance with small data blocks in this random-address writing test but there are a few things we want to discuss in some detail. Take note of the Hitachi pair: they feel more confident with the smallest chunks of data, but their advantage over the Samsung and Seagate Barracuda XT shrinks as the data block size increases. The E3 model stands out among Western Digital’s products. As opposed to the similar FAEX (which still gives us no sign that it has a larger cache, by the way), the E3 is almost as fast on large data blocks as the previous-generation models.
The Green drive from WD is slow when unaligned because it has to do some extra operations due to its 4KB sectors. When aligned, it performs quite normally, its performance matching that of the Seagate Barracuda XT on large data blocks. It only has some problems with the smallest chunks of data where the alignment does not work because the request size is smaller than 4 kilobytes.
The alignment helps the WD Caviar Green on very large data blocks, too. When aligned, it is roughly as fast as the power-efficient models. WD’s Caviar Black products make it clear that HDDs with higher recording density per platter only show their advantage on data blocks larger than 2 megabytes.
In the Database pattern the drive is processing a stream of requests to read and write 8KB random-address data blocks. The ratio of read to write requests is changing from 0% to 100% with a step of 10% throughout the test while the request queue depth varies from 1 to 256.
You can click these links to view the tabled results:
We will build diagrams for request queue depths of 1, 16 and 256.
We’ve got rather odd standings at the shortest queue depth. The Seagate Constellation ES and WD Caviar Black E8 are the leaders in terms of deferred writing. The E8 may be better than the J7 due to some internal improvements (although both use the same version of firmware), but the results of the two dual-platter models are lower than expected.
When the queue depth grows longer, the gaps between WD’s drives with different platters become larger. The newer dual-platter models are also less effective at request reordering. It is not easy to find the leader among their opponents: Seagate’s Constellation ES is only ahead at pure reading or writing. At mixed loads, the Hitachi with new firmware delivers better performance. By the way, the improvement in the latter’s firmware is obvious here.
The Samsung F3 looks best among the power-efficient products. Of course, the Seagate Barracuda LP is very effective at reordering read requests but slows down too much when there are write requests in the queue. The WD Caviar Green behaves like its senior cousins but is hamstringed by its high response time.
The results are still very interesting at the longest queue depth. The old Caviar Black drives from WD are still ahead of the dual-platter models, enjoying a large lead. The new models have smoother graphs here, though. They do not slow down that much at 90% writes. The Caviar Black E8 is competing with the Seagate Constellation ES for top place, being overall better due to higher performance under mixed loads.
As for the power-efficient products, the Samsung EcoGreen F3 loses its ground. Its performance does not scale up as much as that of its opponents at such a long request queue. The Seagate Barracuda LP, on the contrary, speeds up rapidly and takes a leading position despite its very modest deferred writing.
Winding up this part of our tests, we will show you diagrams with each drive’s performance at five different request queue depths.
Hitachi has made some obvious progress in writing firmware. The new version produces smoother graphs and is free from the problem when the HDD slows down at a longer request queue. NCQ is obvious. Deferred writing, even though not very effective, can also be seen. If these algorithms were more efficient, the HDD’s performance would scale up better at longer queue depths.
The Samsung EcoGreen F3 does not show us anything new. Its behavior is similar to the company’s earlier products and its performance is low under most loads. It is only at long queue depths that this HDD is good at reading. But it is slow at mixed loads then – slower even than at the smaller queue depths. It looks like the HDD allots all its resources to reading and there are no cache lines left for deferred writing.
Oddly enough, the power-efficient Seagate Barracuda LP and the server Constellation ES behave in a similar manner whereas the Barracuda XT is completely different. The first two models boast a highly impressive algorithm of read request reordering but slow down as soon as there are any writes to be processed. Both models also speed up at small queue depths when there are more writes than reads in the queue. It looks like the HDD is monitoring the load and decides when it is the right moment to allot its resources for writing. And there are not so many resources, by the way: they are used up even at the smallest queue depth. As a result, there is no effect from the increased queue depth at high percentages of writes.
The Barracuda XT behaves in a demure, classic way. It has all the necessary optimization algorithms but they all work with rather average efficiency.
These diagrams illustrate the changes in firmware of Western Digital’s Caviar Black series introduced together with the 500GB platters. The algorithms have changed noticeably, and not for the better as concerns server loads. The request reordering is now less effective, and the deferred writing has become less aggressive. Interestingly, the FAEX with 64MB buffer is not far better than its E3 cousin: the difference between the 3-platter E3 and E8 is larger.
We don’t quite understand the purpose of these changes. We suspect that there are fewer but larger cache lines in the newer HDDs. This should have a positive effect on performance under loads typical of desktop computers rather than servers.
When unaligned, HDDs with Advanced Format technology suffer a terrible performance hit in this test due to the lack of deferred writing. But when aligned, the WD Caviar Green with 64MB cache behaves just like other HDDs from WD. Although it does not set any records, we cannot find any real fault with it, either.
The drives are tested under loads typical of servers. The names of the patterns are self-explanatory. The results are presented as performance ratings which are calculated as the average speed of the drive under every load.
We have abandoned the old Workstation pattern because PCMark offers more detailed and up-to-date tests.
We don’t see any surprises here. The results might be predicted if you know the read response time of each HDD. And to rank the HDDs up even more accurately, you could look up what HDDs were more effective at long request queue depths in IOMeter: Database. So, the two top places are occupied by the three-platter models WD Caviar Black and Seagate Constellation ES.
The File-Server pattern has some write requests and does not produce as clear a picture as the previous pattern. As you can see, the Seagate Barracuda LP and Constellation ES do not speed up much at queue depths up to 32 requests long, unlike the other HDDs. The Samsung shows a mediocre performance. We know of its scalability problems from IOMeter: Database, but now we’ve got one more confirmation of that fact. Take note of the Hitachi: the newer firmware works better at high loads. The aligned WD Caviar Green is competitive to the other power-efficient products.
Overall, the WD Caviar Black series is in the lead, the 3-platter models with older firmware enjoying a considerable advantage.
The multithreaded tests simulate a situation when there are one to four clients accessing the hard disk at the same time – the clients’ address zones do not overlap. We will discuss diagrams for a request queue of 1 as the most illustrative ones. When the queue is 2 or more requests long, the speed doesn’t depend much on the number of applications. You can also click the following links for the full results:
The Hitachi drive with newer firmware delivers a fantastic result. We would praise it loudly if it were not for one thing. This HDD is as fast as 139 MBps at four data threads, which looks improbable. And if you take a look into the table, you will see amazing data-transfer rates like 160 and 170 MBps. So, this is obviously an error of the benchmark. Even with a narrow test zone and highly optimized firmware, the HDD has to suffer a performance hit, however small it may be.
The considerable fluctuations of speed at different request queue depths are hard to explain, too. We have read through all available results carefully and found nearly all of Hitachi’s latest-generation HDDs with updated firmware to behave like that. The Seagate Constellation ES, according to the table, also manages to be faster at multiple threads than at one thread. The distribution of speed among the threads does not provoke any suspicions, by the way. Every thread is being read at about the same rate. Summing it up, we have to be skeptical about these results and are trying to find the cause of this problem and ways to solve it.
As for the more credible results, the pair of WD’s latest-generation HDDs (with two platters inside) are somewhat faster than the others. The Samsung EcoGreen F3 shows the typical behavior of this company’s recent models: it copes well with two and three data threads but slows down to 10 MBps at four threads.
There are no extraordinary results at multithreaded writing except that the leading Seagate Constellation ES looks suspicious at long request queue depths, but we can explain this by its well-developed caching. Its performance at a request queue depth of 1 is excellent, but within reasonable limits. It is the power-efficient HDDs and the Hitachi (with both firmware versions) that suffer the biggest performance hit from having to write multiple data threads.
For this test two 32GB partitions are created on the drive and formatted in NTFS and then in FAT32. A file-set is then created, read from the drive, copied within the same partition and copied into another partition. The time taken to perform these operations is measured and the speed of the drive is calculated. The Windows and Programs file-sets consist of a large number of small files whereas the other three patterns (ISO, MP3, and Install) include a few large files each, the ISO pattern containing the largest files.
You should be aware that the copying test not only indicates the speed of copying within the same HDD but is also indicative of the latter’s behavior under complex load. In fact, the HDD is processing two data threads then, one for reading and another for writing.
This test produces too much data, so we will only discuss the results achieved in NTFS in the Install, ISO and Programs file-sets. You can use the link below to view full results:
Western Digital’s team win the test of writing in FC-Test, the models with 500GB platters being somewhat faster than their cousins. The Caviar Green must be noted especially as it copes with large files excellently but slows down when processing small files. It is downright poor in the Programs pattern: writing those numerous small files involves extra read operations for this HDD.
Seagate’s HDDs are kind of disappointing. They do not look fast against their opponents that also have 500GB platters. The firmware version does not affect the performance of the Hitachi drive.
We’ve got expected results at reading: the HDDs are ranked up according to their sequential read speed when reading large files. When reading small files, all of them slow down proportionally. The Hitachi with older firmware used to lose its speed too quickly on small files, but this problem is solved in the newer firmware version.
Copying depends on a variety of factors, yet we can note some general rules. Quite expectedly, Western Digital’s Caviar Green does not feel confident processing small files. The Hitachi is not fast with them, either, irrespective of its firmware. The Seagate Barracuda XT is not as good at copying as its server-optimized cousin.
PCMark 2005 has the same tests as the 2004 version (not only in names, but also in results as we have seen a lot of times), so we only discuss one test from PCMark 2004 which is not available in the 2005 version. It is called File Copying and measures the speed of copying some set of files. The other PCMark 2004 results can be learned from the table below. The PCMark 2005 tests are:
The final result is the average of ten runs of each test are available here.
This test agrees with the results of FC-Test but a couple of things must be mentioned specifically. First, the new WD Caviar Black series enjoys a considerable advantage over their opponents. And second, the firmware update helps the Hitachi drive quite a lot, raising its performance to an acceptable level.
These different disk subsystem loads produce the same result: WD’s 1-terabyte Caviar Black series can only be challenged by the Barracuda XT. The Hitachi with updated firmware does not look hopeless, yet its performance is still not high enough to compete with the leaders. Take note of the WD Caviar Green: it is quite competitive to the other power-efficient products, being overall better than the Samsung F3.
Seagate’s 7200RPM drives win the Virus Scan test which is highly sensitive to caching. The Samsung EcoGreen F3 is on the losing side, again.
PCMark 2005 seems to use large files for the copying test: the results are very close to the maximum speeds these HDDs can deliver. The Samsung should be noted for its excellent performance here: it is ahead of the other power-efficient products as well as of WD’s 3-platter Caviar Black models.
The Seagate Barracuda XT boasts the highest overall score. Its server-oriented cousin Constellation ES is third. Second place goes to the WD Caviar Black with 64MB cache.
To make this part of our test session complete, we are going to run the latest version of PCMark called Vantage. Compared with the previous versions, the benchmark has become more up-to-date and advanced in its selection of subtests as well as Windows Vista orientation. Each subtest runs ten times and the results of the ten runs are averaged.
Here is a brief description of each subtest:
Basing on these subtests, the drive’s overall performance rating is calculated.
We’ve got different leaders in the different tests, but we will just point out the most interesting facts. First, neither the Hitachi nor the Seagate Constellation ES could prove their phenomenal capabilities in the multithreaded Windows Defender test. The Hitachi is also inferior to its opponents when it comes to processing small files, being closely followed by the power-efficient products then.
The overall ratings are not surprising at all. The Seagate Barracuda XT is followed by the four WD Caviar Black drives. The Seagate Barracuda LP wins in the sector of power-efficient products.
Next goes our homemade test of defragmentation speed. We created a very defragmented file system on a 32GB partition of a disk by loading it with music, video, games and applications. Then we saved a per-sector copy of the disk and now copy it to the disk we want to test. Next we run a script that evokes the console version of the Perfect Disk 8.0 defragmenter and marks the time of the beginning and end of the defragmentation process. We run this test with AHCI enabled. For more information about it, you can refer to this article.
There are no unexpected results in this test, yet we’ve got something to talk about. First of all, we can note the good performance of the Samsung which is but slightly slower than WD’s Caviar Black that occupy the podium. The Hitachi with new firmware takes only one fourth of the time the old firmware needed to complete this test. The mediocre performance of the WD Caviar Green can be explained by its 4-kilobyte sectors (as you can guess, this test is largely about processing small data requests), but the poor results of the Seagate Barracuda LP and Constellation ES are hard to explain.
Now we are going to show you one more interesting test in which we use WinRAR version 3.8 to compress and then uncompress a 1.13GB folder with 8118 files in 671 subfolders. The files are documents and images in various formats. These operations are done on the tested drive. This test depends heavily on CPU performance, but the storage device affects its speed, too.
Most of our drives pass the test with similar results. There are only three losers that spent half a minute more than the others: the Seagate Barracuda XT, the WD Caviar Black J7 (the very first series) and the Hitachi with original firmware.
This is a very vivid illustration of how hard it is for HDDs with 4KB sectors to write lots of small files. That is the single serious downside of the new technology, but we cannot neglect it.
WD’s Caviar Black enjoy an overwhelming advantage, the older models being somewhat faster than the newer ones. The Samsung joins the group of leaders, which is a success for that drive.
You can refer to our article called Hard Disk Drive Power Consumption Measurements: X-bit’s Methodology in Depth for details on this test. We will just list the specific modes we measure the power consumption in:
Let’s check out each mode one by one.
The HDDs split up into three groups in terms of startup current. As expected, the power-efficient products need less power although the Seagate Barracuda LP is an exception. Its startup power consumption is as high as that of the most voracious drives which include the two other drives from Seagate and the new WD Caviar Black series.
The WD Caviar Green behaves unnaturally in the idle mode. Its power consumption is too high and the oscillograms suggest that the HDD is not idle at all. We don’t know what this HDD was doing all the time but it did not get to sleep even after an hour of being in idle mode.
The other power-efficient HDDs are all right: they take top places, the Samsung being the most economical one in terms of 12V consumption.
Western Digital’s E3 has some mysterious activity on the 5V line, too. At the same time, it is phenomenally economical in terms of 12V consumption. Although we can see a general trend towards lower consumption from the 12V line with the WD products, the difference of 1.5 watts from the very similar Caviar Black FAEX is a serious one. We can only suspect that the HDD entered a very deep sleep mode, shutting down everything save for the platters drive.
The Hitachi is good. For all its 7200 RPM, it needs less than 5 watts when idle. Well, nearly every HDD deserves our praise here. For example, the Seagate Constellation ES consumes a mere 5.4 watts – a fantastic result compared with the first generation of 7200 RPM drives.
Random reading is when the HDD turns off its cache and moves its heads around. Therefore, the results differ only on the 12V line. The Samsung saves a lot of power, the WD Caviar Green following it closely. The Seagate Barracuda LP is, on the contrary, not very good, consuming almost as much as the 7200RPM Seagate Barracuda XT and WD Caviar Black E3. The 3-platter models consume more in the Caviar Black series. Quite expectedly, an HDD has to spend more power to rotate three rather than two platters.
The Hitachi leaves us perplexed. It has only two platters and its heads do not seem to be very quick, so what does it need so much power for?
Random writing is overall similar to random reading in terms of power consumption although each HDD consumes less power due to caching. The WD Caviar Black E3 is especially good, being almost as economical as the 2TB power-efficient models. The Seagate Barracuda XT is good, too, for an HDD with four platters rotating at 7200 RPM. It is the first-generation WD Caviar Black that consume the most power of all.
The Hitachi competes with the 5400RPM drives in terms of power consumption at sequential operations. And it does so successfully thanks to the low consumption on the 5V line. The WD Caviar Black E3 is far more economical than its predecessors as well as its cousin with 64MB buffer. The Seagate Barracuda XT is the most uneconomical drive here, although it differs but slightly from the 3-platter WD Caviar Black models.
Finally, we can note that modern HDDs all have a power consumption of 10 watts and lower. The reduction of spindle speed produces a 25% reduction in power consumption (and, accordingly, heat dissipation).
Now, let’s do some summarizing. We will go through the tested products in alphabetic order.
The Hitachi 7K1000.C had very odd basic firmware with low performance, but now it looks decent enough. It breaks no records and is not completely free from shortcomings, but the 3EA firmware makes this HDD a normal option for the home computer of an undemanding user.
Samsung is making first steps in producing HDDs with four platters. The EcoGreen F3 is the first such product and we cannot expect something extraordinary from it. Yet we must acknowledge that it is inferior to its opponents, especially under server loads. On the other hand, it has low power consumption and features fast processing of files, which makes it a good choice as a passive data storage device.
The Seagate Constellation is a nice surprise to us. It is a really good server-oriented HDD with firmware optimized for server loads and with rather fast heads (which is something very rare for Seagate’s recent products). The Barracuda XT is still a good choice as an HDD for home computers whereas the Barracuda LP can do well enough in low-load servers or as a power-efficient disk with rather high performance.
Western Digital’s Caviar Black series is somewhat ambiguous. It looks like the transition to new platters has not been easy and the developer’s main goal was to keep performance at the same level rather than increase it. This goal has largely been achieved although the HDDs have become slower in server applications (or the company just wants to separate its top-performance server and desktop series more). Yes, the model with 64MB buffer and SATA 600 is the new acme of the series (among the 1-terabyte models), but we guess the WD Caviar Black E3 (with two 500GB platters inside as well) looks even more interesting. The latter delivers roughly the same performance, especially when it comes to typical home applications, but consumes less power.
The WD Caviar Green with 64MB buffer is one more drive from Western Digital to sport 4-kilobyte sectors. We guess we will see more and more such products in the future. You should not be wary of them as their performance is absolutely normal. The software emulation of 512-byte formatting in the HDD’s firmware has only one serious downside: the drive’s performance lowers when processing very small (smaller than 4 KB) data blocks. Of course, you should not forget to align the HDD’s partitions if you are going to use it in older OSes that do not natively support 4KB sectors.