by Nikita Nikolaichev
11/07/2011 | 09:48 AM
Last year we tested 500GB hard disk drives which can be an optimal choice for users who don’t need a lot of disk storage. Today, we want to take a look at products which are twice as large. They are generally as fast as their higher-capacity counterparts but have only 2 platters, which makes them quieter, colder and more reliable (however optimized an HDD's mechanics can be, the fewer platters and heads you have, the less likely they are to fail). On the other hand, they are not entry-level products with all the consequences, like their single-platter cousins. 1-terabyte HDDs are also attractive in terms of price/capacity ratio.
Besides other things, we will see if SATA600 brings any tangible benefits and makes a difference compared to the other interface.
We’ve got HDDs from Seagate and Western Digital for this review, so it reflects the market situation we can see today. Hitachi has been bought up by WD whereas Seagate has acquired Samsung’s HDD manufacturing facilities. There are in fact only two competitors left in the desktop 3.5-inch HDD market.
We’re already familiar with this HDD. One of the first HDDs with 500GB platters, it opened Seagate's 12 series along with its 1.5-terabyte cousin. Its spindle rotation speed is 7200 RPM and its buffer is 32 megabytes large. This model hasn't changed much over the year, so we can use it as a baseline for our comparisons.
According to the specs, this HDD is the same as the previous model except for the interface. Their labels are the only external difference, so we expect them to deliver similar performance although the different firmware may change something.
This energy-efficient model has been around for over a year. Like the two previous HDDs, it has two platters, four heads and 32 megabytes of cache, but its spindle rotation speed is set at 5900 RPM.
This model looks much different from the three above because it features denser platters (422 as opposed to 329 gigabits per square inch). It has 4KB sectors whereas all the other HDDs in this review have conventional 512-byte sectors. The manufacturer's website doesn't reveal how many heads this HDD has but we can suppose it’s got three because there's also a 2-terabyte model with six heads in this series. On the other hand, the 1.5-terabyte model from the same series has six heads as well, so the 1-terabyte one may have as many as four.
The cache is 32 megabytes large; the spindle rotation speed is 5900 RPM. By the way, this HDD is categorized as Green rather than LP at the official website. The change of name must be meant to make it more comprehensible (although we guess that Low Power was quite comprehensible, too).
We should also mention the Smart Align technology which, quoting Seagate's info sheet, “resolves Advanced Format misalignment conditions while preserving hard drive performance”. It seems to use read/write buffers to minimize the overhead when writing data blocks smaller than 4 kilobytes.
The first WD1001FALS came to our labs over a year ago but that was a 00J7B0-indexed version with three platters and six heads. It has gone through several revisions and is now 00Y6A0. As far as we know, the latest revision has two platters and four heads, so we can expect some changes in its performance.
Identical to the previous model externally, this one is a new HDD. It differs in cache amount (64 MB) and interface (SATA600).
Like the FALS, this HDD was tested in our last-year roundup. It had the same index of 00Z8A0 then. Its specs haven’t change over the year. It is still positioned as a mainstream product with 7200 RPM and 32 megabytes of cache.
Here’s another pair of HDDs with identical specs but different interfaces. The cache is still 32 megabytes.
We tested a similar HDD in our labs but it was twice the capacity. It is meant for entry-level servers and thus features better mechanics, higher reliability, extra sensors and a technology for quick error correction. As opposed to all other HDDs from Western Digital, this one’s specs explicitly say that it’s based on 2 platters and 4 heads. The spindle rotation speed is a conventional 7200 RPM whereas the buffer is as large as 64 megabytes.
Here is a table with the specs and firmware versions of the HDDs we are going to test:
The following testing utilities were used:
HDDs are tested with generic OS drivers. We format them as one NTFS partition with the default cluster size (for FC-Test we create 32GB partitions), connect them to a mainboard port and enable AHCI. I would like to remind you that we are now using new HDD testing methodology.
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 a disk’s sequential read/write speed on the size of the data block. This test is indicative of the maximum speed a hard disk can achieve.
Rather surprisingly, the Barracuda LP ST1000DL002 with its 5900 RPM comes out on top in this test. This is due to its higher recording density thanks to 4KB sectors.
Then, we can note that the HDDs split up into two groups when processing small-size data blocks: the Seagate ones are twice as fast as their WD opponents at data chunks up to 2 KB. The WD drives catch up with the Seagate team at 8KB data blocks and reach their top speeds at 16B data blocks only. The RE4 1003FBYX is the only exception as it delivers its maximum speed as soon as 8KB blocks.
The rather low performance of the Barracuda LP ST31000520AS and Caviar Blue WD10EALX should also be noted here.
The sequential write results do not make up such a nice-looking diagram as we've just seen in the test of sequential reading. The Barracuda 7200.12 ST31000528AS slows down while the Barracuda 7200.12 ST31000524AS is good and the Barracuda LP ST1000DL002, just excellent. The WD drives are slow until 32KB data blocks, the Raid Edition 4 model again accelerating to its top speed sooner than the others.
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.
The HDDs are all close to each other in terms of read response, even the 5900RPM models. The WD drives are somewhat faster overall, though. When it comes to writing, the RE4 WD1003FBYX and the Caviar Black WD1002FAEX are the most responsive whereas the Barracuda 7200.12 ST31000524AS falls behind the main group. The Barracuda LP ST1000DL002 is downright slow due to its 4KB sectors.
What can we learn from these results? First, a lower response time is going to translate into higher performance at random-address operations. The Caviar Black WD1002FAEX and RE4 WD1003FBYX should be singled out in this respect: besides moving their heads around very quickly (which is indicated by their low read response time), they feature very efficient request reordering algorithms (indicated by the difference between the read and write response).
Second, the low result of the Barracuda LP ST1000DL002 is indicative of 4KB sectors. This HDD has to perform three operations to write a single 512-byte data block (read it, recalculate the sector’s checksum, and actually write it) whereas HDDs with 512-byte sectors can write 512-byte data blocks in one operation since the sector’s checksum doesn’t depend on the rest of the disk contents.
Third, a low response time has the downside of higher noise and heat. It means that the Seagate HDDs are likely to be quieter than the WD ones, although this can vary depending on the particular sample.
The results match those of the response time test. The Barracuda LP ST1000DL002 falls back to last place after being first at sequential reading. It is considerably slower than the next worst drive, which is the Barracuda LP ST31000520AS. This is due to the spindle rotation speed of 5900 RPM, of course. The main group is headed by the RE4 WD1003FBYX but the gaps are small there.
It’s more interesting at random writing: having failed the previous test, the ST1000DL002 also fails when writing random-address data blocks smaller than 4 KB (to do such a write, it has to read the data block, wait for the platter to make a full turn, and then perform the write operation proper). However, it is better than the rest of the Seagate HDDs with larger data blocks.
The rest of the results are roughly equivalent to those of the response time test. The Caviar Black WD1002FAEX and RE4 WD1003FBYX are ahead, despite the slump at 128KB data blocks. The other WD drives follow the leaders in a tight group and are joined by the Barracuda LP ST1000DL002. The rest of the Seagate drives go next, the Barracuda 7200.12 ST31000524AS having some problems with deferred writing as its speed isn’t much higher than at random reading.
The Barracuda LP ST1000DL002 has a peak of performance at 1KB data blocks, which must be due to the combination of Smart Align technology with a cache line size of 1 KB.
The reading of a single thread is the same as sequential reading, so the two Barracuda LP products take the first and last lines of the diagram. The 7200.12 series models and the WD drives, except for the Caviar Black WD10EALX, follow close behind the leader.
As soon as there is a second data thread to read, the Barracuda LP ST1000DL002 delivers a much lower performance whereas the Barracuda LP ST31000520AS, which wasn’t that fast when reading a single thread, doesn’t slow down that much here. The 7200.12 series products are both very fast. The RE4 WD1003FBYX is twice as fast as the rest of the WD disks, but can only compete with the energy-efficient product from Seagate. The other WD drives are much slower than their Seagate opponents.
Seagate’s 7200.12 series is three times as fast as most of the WD drives when reading three data threads and keep on delivering over 20 MB/s. The RE4 WD1003FBYX gets slower by half, joining the low-performance group, but the Caviar Black WD1001FALS speeds up a little, producing the same result as the Barracuda LP ST1000DL002 with two data threads. However, the latter has also improved to keep ahead of its pursuer.
Now each of the five WD drives yields only 7 to 7.5 MB/s whereas the Seagate HDDs seem not to notice the increased load except that the Barracuda LP ST1000DL002 falls somewhat behind.
The Barracuda LP ST1000DL002 is again on top when processing one data thread, but is closely followed by a group of WD drives along with the Barracuda 7200.12 ST31000524AS. The Caviar Blue WD10EALX and the Barracuda LP ST31000520AS are slower than the rest of the HDDs but it is the Barracuda 7200.12 ST31000528AS that is the worst model here, just as in the sequential reading test.
We could expect the Seagate team to deliver fantastic performance like they did in the multithreaded reading test, but their speeds of 60 MB/s (Barracuda 7200.12 ST31000524AS) to 50 MB/s (the rest of the Seagate team), although quite high too, are not impressive compared to the WD drives’ 100 to 105 MB/s. The Caviar Blue WD10EALX’s 96 MB/s doesn’t spoil the overall picture for Western Digital.
When the load grows higher, the gaps between the WD drives get larger, the RE4 WD1003FBYX finding itself on top with a better result than with two data threads. The Barracuda 7200.12 ST31000524AS gets closer to the Caviar Blue WD10EALX.
The standings remain the same when the HDDs process four data threads. The Caviar Black WD1002FAEX and the Barracuda 7200.12 ST31000528AS get slower, but this doesn’t change the overall picture: the Western Digital products are better at multithreaded writing.
The next IOMeter pattern simulates a typical file-server load. The request queue depth changes from 1 to 256. Let's first view the results at a queue depth up to 32 requests.
The Barracuda LP ST31000520AS behaves unlike the other HDDs, reaching its top speed at a queue depth of 4 requests. This optimization makes it somewhat faster under real-life conditions but prevents it from showing a higher performance at longer queue depths.
It is the RE4 WD1003FBYX and Caviar Black WD1002FAEX that have the highest results again. They are followed by the Caviar Blue WD10EALX and Barracuda 7200.12 ST31000524AS whose graphs are close to coinciding except that the Seagate is somewhat slower at a queue depth of 1 and 2 requests. The Barracuda 7200.12 ST31000528AS and WD10EALS are even slower, and the Barracuda LP ST1000DL002 fails the test completely.
We’ve got a somewhat different picture when the queue is as long as 256 requests. Seagate's energy-efficient products exchange their places, the Barracuda LP ST1000DL002 now being indifferent to the queue depth (just like the Barracuda 7200.12 ST31000524AS, though). The Barracuda LP ST31000520AS speeds up suddenly as the queue increases from 32 to 64 requests. The RE4 WD1003FBYX enjoys a clear first place while the Barracuda 7200.12 ST31000528AS and the WD Caviars go neck and neck.
Once again the Barracuda LP ST1000DL002 falls behind. The RE4 WD1003FBYX is on top, followed by the Caviar Black WD1002FAEX.
The graphs diverge when the queue is increased up to 256 requests. The Barracuda 7200.12 ST31000524AS and Barracuda LP ST1000DL002 do not accelerate anymore. The RE4 WD1003FBYX and Caviar Black WD1002FAEX are unrivalled. The rest of the HDDs deliver similar performance.
The Caviar Black WD1002FAEX wins this test, featuring excellent deferred writing algorithms. Being as fast as the RE4 WD1003FBYX at pure reading and writing, it is much faster under mixed loads. The 7200.12 series disks deliver the same performance at any reads/writes ratio whereas the energy-efficient models cope well with this test. The LP ST31000520AS isn’t much slower than the WD drives whereas the Barracuda LP ST1000DL002 catches up with the ordinary WD disks at writing. That’s an excellent result for 5900RPM HDDs.
The request queue getting longer, the RE4 series model gets better. The Caviar Black WD1002FAEX doesn’t enjoy such an overwhelming advantage at mixed loads as before. The Caviar Blue WD10EALS, Caviar Blue WD10EALX and Caviar Black WD1001FALS draw neat graphs without any defects whereas the 7200.12 series products are good until 60% writes but slow down at 80% writes.
The Barracuda LP ST1000DL002 performs very well and even catches up with the WD team at 100% writes. The Barracuda LP ST31000520AS is good at pure reading and writing but doesn’t like mixed loads.
When the queue is as long as 16 requests, Seagate’s 7200.12 series HDDs compete with the regular products from WD at reading but fail to do so at high percentages of writes. Meanwhile, the Barracuda LP ST1000DL002 behaves differently. Its caching algorithms are optimized for this load, as opposed to the Barracuda LP ST31000520AS which slows down even at 10% writes.
The WD drives behave exactly as at a queue depth of 4 requests. The RE4 WD1003FBYX and Caviar Black WD1002FAEX are unrivalled in this test.
The Caviar Black WD1002FAEX and RE4 WD1003FBYX are on top again although the slump at 90% is now more conspicuous. The standings of the other HDDs have changed. They have got closer to each other, and the Barracuda LP ST1000DL002 is just as good as others.
The RE4 WD1003FBYX and Caviar Black WD1002FAEX are in the lead again. WD’s regular drives deliver similar performance. The Barracuda 7200.12 ST31000528AS and the Barracuda LP ST31000520AS join the group of WD drives, although have completely different graphs. The Barracuda LP ST1000DL002 and 7200.12 ST31000524AS do not do very well here.
For this test two 32GB partitions are created on the hard disk and formatted in NTFS. A file-set is then created, read from the disk, copied within the same partition and copied into another partition. The time taken to perform these operations is measured and the speed of the HDD 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.
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.
When reading large, ISO-like files, the HDDs almost reach their top sequential read speeds, so we don’t see anything new in this test. The Install pattern isn’t informative, either: every HDD has the same result except for the Caviar Blue WD10EALX, Barracuda LP ST31000520AS and Barracuda 7200.12 ST31000524AS. When reading the smaller files of the MP3 pattern, the losers are the same but the rest of the HDDs have a leader (it’s the RE4 WD1003FBYX). The data-transfer rates are not much lower than the sequential read speeds of these HDDs, though.
Switching to the Programs pattern wasn’t that easy, though, as the speed dropped by half. We see all of the WD drives go ahead, being obviously better under hard loads. The same goes for the Windows pattern except that the Caviar Black WD1002FAEX is somewhat faster than the rest of the WD drives. The Barracuda LP ST31000520AS takes last place in both cases, but the other Seagate drives are close to it.
We’ve got the same results as in our sequential writing test.
Seagate is on the losing side again. The Barracuda 7200.12 ST31000524AS is the only model to deliver acceptable performance while the rest of them are much slower than in the previous test. The WD drives have remained almost as fast as with the ISO pattern.
Seagate has problems again. The RE4 WD1003FBYX is in the lead while the Caviar Blue WD10EALX is the slowest of the WD drives. The latter is joined by the Caviar Black WD1001FALS in the Programs and Windows file-sets.
The Barracuda LP ST1000DL002 and RE4 WD1003FBYX are best at copying large files irrespective of whether the two data threads are near or far from each other. The Barracuda 7200.12 ST31000524AS is good, too. Despite its problems with writing, the Barracuda 7200.12 ST31000528AS is in the middle of the standings while the last places belong to the Caviar Blue WD10EALX and Caviar Black WD1001FALS.
The files are smaller and the Seagate drives are slow again, except that the ST31000524AS is somewhat faster than the WD1001FALS.
Seagate’s HDDs are in the bottom half of the results table again, even though the Caviar Black WD1001FALS and Caviar Blue WD10EALX aren’t much faster, either. The Barracuda LP ST31000520AS is downright slow while the RE4 WD1003FBYX and Caviar Black WD1002FAEX take top places.
The picture is already familiar to us. The Barracuda LP ST1000DL002 is the slowest drive this time around.
We don’t see many changes here except that the Barracuda 7200.12 ST31000524AS has caught up with the Caviar Black WD1001FALS and Caviar Blue WD10EALX.
Here is yet another high-level test. Intel NAS Performance Toolkit contains a set of traces to benchmark any storage device, not only NASes. It uses the standard OS mechanisms of accessing storage devices, so the test platform and the software installed on it affect the results.
By the way, Windows 7 uses caching for writing data to a disk and reports to the test tool that data has been written before it is actually written. Therefore the time an HDD takes to pass the test may be smaller than the amount of time the HDD actually spent, and the results may be higher than in the low-level tests.
The Caviar Blue WD10EALS is somewhat surprisingly ahead in the read test although the rest of the HDDs are close to each other, the Barracuda LP ST31000520AS being the only exception. When copying a whole folder, three disks from WD and the Seagate Barracuda 7200.12 ST31000528AS go ahead while the Barracuda LP ST1000DL002 falls behind.
The WD team wins once again, the Barracuda LP ST1000DL002 being the only Seagate drive to offer any resistance in the file copy test.
WD’s drives are unrivalled again. The Barracuda 7200.12 series are inferior to the 5900RPM Barracuda LP models.
This is in fact a sequential reading test.
The Barracuda 7200.12 ST31000528AS slow down again when switching from reading to writing.
The RE4 WD1003FBYX, Caviar Black WD1002FAEX and Caviar Blue WD10EALS take the podium at multithreaded load. They are followed by the energy-efficient HDDs together with the Barracuda 7200.12 ST31000524AS and Caviar Blue WD10EALX. The Barracuda 7200.12 ST31000528AS and Caviar Black WD1001FALS are the slowest drives here.
As opposed to IOMeter, the WD drives are ahead at multithreaded reading, the Caviar Blue WD10EALS being the winner. This is due to the difference in load: the data threads are close to each other with NASPT whereas IOMeter puts them 100 gigabytes apart from each other.
The Caviar Blue WD10EALS isn’t good for office applications, though. The RE4 WD1003FBYX is expectedly on top.
The results are close, yet the standings do not look right to us after the earlier tests.
Now we are going to show you another interesting test in which we use WinRAR version 3.91 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 HDD. This test depends heavily on CPU performance, but the storage device affects its speed, too.
The standings are close in the compression test, the Caviar Black WD1001FALS and Barracuda 7200.12 ST31000524AS being somewhat faster than the others. The HDDs differ more when uncompressing the archive: the Barracuda LP ST1000DL002 is a clear loser while the RE4 WD1003FBYX and Caviar Black WD1002FAEX are ahead of the rest of the HDDs.
Compared with the previous versions, the Vantage version of PCMark is 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 HDD’s overall performance rating is calculated.
The Seagate HDDs are on the losing side here. Despite its rather good performance in the rest of our tests, the Barracuda LP ST1000DL002 fails in PCMark, being last in every subtest save for Application Loading. The Caviar Black WD1001FALS doesn’t do well, either. It shares the last-but-one place with the Barracuda 7200.12 ST31000528AS in the overall standings. The RE4 WD1003FBYX, Caviar Black WD1002FAEX and Caviar Blue WD10EALS compete among themselves and occupy the top of the podium.
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 that disk and now copy it to a disk we want to test. We run a script that evokes the integrated defragmenter of Windows 7 and marks the time of the beginning and end of the defragmentation process. For more information about this test, you can refer to this article.
The RE4 WD1003FBYX wins this test by a large margin. The Barracuda 7200.12 ST31000524AS and Barracuda 7200.12 ST31000528AS are second and third, respectively. The Caviar Blue WD10EALX and Caviar Black WD1001FALS are somewhat slower than the leaders while the Barracuda LP ST31000520AS delivers the performance you can expect from a 5900RPM drive.
This test illustrates Smart Align technology. You can see that the Barracuda LP ST1000DL002 performs worse when the partition is not aligned (the result is green in the diagram). However, the performance hit is not as huge as in our earlier reviews. So, Smart Align is helpful indeed.
You can refer to our Hard Disk Drive Power Consumption Measurements: X-bit’s Methodology in Depth for details on this test. Here is a list of the specific modes we measure the power consumption in:
The high power draw of the Barracuda LP ST1000DL002 is noticeable in the diagram. It is not cold when idle. The Barracuda LP ST31000520AS consumes much less, for example. The rest of the Seagate disks are economic, too. This is one of the few tests where less is better.
The standings do not change much when the HDDs are accessing sequentially located data. There are some changes in the ranks of the WD products while the Barracuda 7200.12 ST31000524AS turns out to be less economic than the Caviar Blue WD10EALS.
We've got a different situation at random-address loads: the Barracuda LP ST1000DL002 still consumes about 7 watts when writing but needs as much as 10 watts when reading, which is the worst result in the test. The Barracuda LP ST31000520AS remains the most economical drive whereas the RE4 WD1003FBYX, despite its impressive performance, is not the hottest one. The Caviar Blue WD10EALS is yet another product that is both fast and economic.
The newer products with SATA600 have proved to be generally faster than their counterparts with the older interface. On the other hand, we can't name an overall winner as all of the tested products have performed quite well. Unfortunately, we couldn't benchmark Barracuda XT and ES series drives for this review, which may provoke the false impression that Seagate is much worse than Western Digital, but the positioning of HDDs should also be taken into account.
The RE4 WD1003FBYX is very good. It hardly has any rivals under high loads but you have to pay for this. We mean, literally. It is twice as expensive as both the old and new Barracuda LP disks, for example. It is going to be the perfect choice as a high-capacity RAID disk which is fast but not necessarily as fast as SAS disks. This is the application it was developed for, after all. But when it comes to home computers, the RE4 WD1003FBYX doesn’t look optimal. It may fall behind other HDDs at medium and low loads typical of home applications while costing much more than them. By the way, if storage capacity is not a critical factor (e.g. you use a different disk to store your data files on), you may want to prefer an SSD which is going to be even more expensive but also much faster. What we're driving at is that if you don't know exactly why you need the RE4 WD1003FBYX of all the models available, you may want to find a more optimal disk for your particular system.
We’d recommend the Caviar Black WD1002FAEX as a very fast HDD. It is occasionally faster even than the RE4 WD1003FBYX but costs considerably less. It is clearly better than the older Caviar Black WD1001FALS in most of the tests, so we don’t think the latter is worth buying.
Comparing the Caviar Blue WD10EALS and the Caviar Blue WD10EALX, we’d prefer the newer HDD, too. The WD10EALX is as fast as the previous-generation model but features an excellent price/capacity ratio. It is a well-balanced product offering a rather high performance at a more or less affordable price. Take it if you just need a good HDD.
Well, we can say the same about the 7200.12 series from Seagate. Far from impressive in our tests, they are actually not much slower than the Caviar Black series in most of them. Moreover, they were much faster than their opponents in IOMeter’s multithreaded reading test which seems to be the closest to real-life loads (it is the same as reading several files scattered around the hard disk).
And if you need a passive storage device for your multimedia collection, the Barracuda LP is going to be a perfect choice. The ST1000DL002 model is quite impressive at low loads and, thanks to Smart Align technology, has very few, if any, problems because of its 4KB sectors. The Barracuda LP ST31000520AS, on its part, was the slowest but also the most economical disk in this review. Thus, it has lower cooling requirements.