by Aleksey Meyev
11/05/2008 | 12:18 PM
Three quarters of a terabyte – is it much or not? We guess it is quite a lot of storage space, especially for a single hard disk, even considering the rapid growth of data amounts. Such a disk can store as many as thirty Blu-ray movies, which is the best consumer video format with outstanding quality. As for storing music and photos, the numbers are going to be just huge. Thus, a 750GB hard disk drive is a universal storage on which you can install your OS and applications as well as keep your home collections of video, audio and photo materials. The cost of storage with HDDs has declined so much that the HDD is now competitive to optical discs in this parameter.
Seagate was the first company to offer a 750GB hard disk drive. It was the superb four-platter model from the Barracuda 7200.10 series. We dedicated a special review to that drive in the fall of that year. Every other company has not only introduced its own 750GB model since then but also released a 1-terabyte drive, reaching a new milestone in terms of storage capacity. The recording density per platter has been steadily increasing, too. The above-mentioned Barracuda 7200.10 had a recording density of 187.5GB per platter whereas modern HDDs have reached a density of 320GB per platter. 375GB platters are already looming in the horizon. So, while the first generation of 750GB drives had four platters, the newer models are based on only three. Talking about the current situation and the close perspective, 1000GB drives are nothing special already and have already declined in price (and the prices of HDDs of smaller capacities have declined even more). The next round of the race is about to begin as Seagate has already announced its ability to introduce 1.5TB drives.
So, it is time to find the highs and lows of the HDDs available today. We have discussed 1TB drives recently, and now we want to deal with 750GB ones. Writing a comparative review is not easy as there are always new products appearing on the market. Once you collect enough material, a new model pops up and another is about to be released. Therefore some newest models are missing in this review, yet we tried to cover as many HDDs as we could.
A 750GB drive is not the largest available, but it is cheaper than a 1TB drive while its capacity should suffice for most applications. So, this storage capacity is all right, but what about speed? You’ll find this out right now.
The series is named after the senior five-platter model which is based on 200GB platters. The HDT721075KLA330 has only four platters, though. If you want to learn more about this series, you can refer to this comparison of different series of hard disk drives from Hitachi.
Interestingly, the Hitachi drive is the only four-platter drive in this review to be equipped with 32 megabytes of cache memory. The other such drives have a 16MB buffer, being inferior to the three-platter models in this respect.
This series signified a fabulous breakthrough on Samsung’s part. The company made a jump from 500GB drives right to 1-terabyte ones, skipping over 750GB models (a 750GB drive was released within the new series headed by a 1TB model), and the F1 series (a telling name) was the first one to be based on 334GB platters. Thus, the 750GB model of the series has three platters only. We wonder if it has the same recording density as the 1TB model that produced superb results in our tests. We’ll check this out in the read speed tests.
Of course, we couldn’t pass by the very first drive with a capacity of 750 gigabytes. We will see how obsolete it has become and if it is far inferior in speed to the newer models. After all, two years of market life is quite a long period for a computer product. Don’t think it is a rarity, though. This drive is still available in shops.
This table contains information about two drives because the enterprise version is almost an exact copy of the desktop model.
The Barracuda ES series had replaced the NL.2 series as including hard disk drives intended for enterprise applications, for example data warehouses. Theoretically, such HDDs are optimized for server loads and feature higher tolerance to vibrations and temperature. We don’t have enough time to check out the tolerance characteristics, but you will see how good this HDD is under server loads.
This drive belongs to the latest, eleventh, generation of Seagate’s Barracudas. The platters of this generation do not break any records: while Samsung’s 1GB drive is based on three platters, Seagate employs four. The reduction of the storage capacity helped use a 3-platter design for the 750GB model, though.
Logically enough, the desktop version of the drive is accompanied with an enterprise version.
It seems to be a twin brother of the previous drive. You’ll see in our tests if they are indeed identical in terms of firmware and speed.
This HDD used to belong to the Caviar SE16 series some time ago but then Western Digital decided to diversify its HDD line-up and offered as many as three Caviar series. This HDD fell into the basic, blue, series. Besides it, there are a high-performance Caviar Black and an economical Caviar Green series. The latter includes 3.5-inch HDDs with a spindle rotation speed of 5400rpm. This is the company’s innovation as all desktop HDDs used to have a spindle speed of 7200rpm for quite a long time. This return to the past is meant to reduce power consumption, which is the main feature of all “green” HDDs. As for the Black series, it is going to deliver the highest performance of all. So far, we have only got a Caviar Blue. The Blue series seems to include rather old models with not-highest recording density that offer an “optimal combination of price and quality” as the marketing folks like to call it.
The Caviar Blue and the server-oriented model from the RE2 series are based on four platters.
Why leave Seagate’s enterprise drives without competition? Western Digital is quite competitive as it produces a pair of a desktop and server model at each step up the technological ladder.
Here is a “green” drive from the server-oriented RE2-GP series. It is funny how Western Digital declares the spindle rotation speed of these HDDs. While this parameter can be easily found among the specs of every other model, it is either not specified for the Green Power disks at all or hidden behind the term IntelliPower that doesn’t offer any specific number. It looks like Western Digital is ashamed of confessing that the platters of these HDDs are rotating at 5400rpm. At least our tests indicate clearly that the spindle speed of this HDD is not variable (as had been rumored in some news articles before the release of the Green Caviar series) but constant and equals 5400rpm.
We’ve got two HDDs from this series for today’s test session: WD7500AYPS and WD7501AYPS. Note that these HDDs feature a 3-platter design but have only 16 megabytes of cache memory.
We received this drive just before we began our test session and the Western Digital website refused to tell anything about it (as opposed to Google). We don’t know if the change of the nomenclature is due to firmware changes, new platters or new electronics or all that together. Anyway, we are more interested in any changes in the speed characteristics of the HDD.
The following testing utilities were used:
We installed the generic OS drivers for the drives and formatted them in FAT32 and NTFS as one partition with the default cluster size. For some tests 32GB partitions were created on the drives and formatted in FAT32 and NTFS with the default cluster size, too. The drives were connected to a Promise SATA300 TX4302 controller installed into a PCI-X slot and switched from the quiet mode (with Advanced Acoustic Management enabled) into the ordinary operating mode when necessary. Of course, the drives are all tested with enabled AHCI.
First we recorded the drives’ data-transfer graphs using WinBench:
The following diagram compares the read speed at the beginning and end of each disk:
Seagate’s new series HDDs deliver the highest read speed. They are the only drives to reach the 100MBps mark, the desktop 7200.11 series proving to be somewhat faster than the server-oriented ES.2. The latter series must have a reduced recording density in order to ensure higher reliability. Anyway, 100MBps is an astounding result. The progress of the HDDs is obvious: the transition to the new generation with higher recording density has increased the performance by a third!
The Samsung is a disappointment. It should have been faster than or at least as fast as Seagate’s HDDs but delivers only 90MBps. Judging by the data-transfer graph, the three platters differ in speed greatly. We seem to have adaptive formatting here. It is possible that Samsung’s 750GB drives are based on platters that did not manage to reach the recording density necessary to achieve a capacity of 1 terabyte on three platters. This is only our supposition, though. This HDD may just be based on other platters that have lower recording density originally.
The results of the RE2-GP series drives from Western Digital are interesting: spinning at 5400rpm, they deliver the same speed as the 7200.10 series drive from Seagate with a higher spindle rotation speed. The newer WD7501AYPS slows down less than the older WD7500AYPS towards the end of the disk. So although they have similar top speeds, the WD7501AYPS is going to have a higher average read speed.
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 highest speed the drive can achieve.
The numeric data can be viewed in tables. We’ll discuss graphs and diagrams. There are two many results, so we divide the graphs of the HDDs in two groups basing on the number of platters (three or four).
IOMeter: Sequential Read results
Seagate’s new HDDs from the 7200.11 and ES.2 series are in the lead again whereas the Samsung shows a considerably lower speed. The Samsung is worse than the others at processing small data chunks. It is outperformed even by the 5400rpm GP series drives from Western Digital. The latter are not so slow, by the way. They are as fast as the Seagate 7200.10 on large data blocks and close to the leaders of this test on small ones.
It is interesting to compare HDDs from the same generation but different series (desktop and enterprise). For example, the WD Caviar Blue is always a little faster than the RE2. Seagate’s 7200.10 and ES are equals whereas the Seagate 7200.11 has a higher top speed but is worse with small data chunks.
IOMeter: Sequential Write results
We’ve got the same overall picture at writing. The different behavior of Western Digital’s RE2-GP drives can be noted: the newer WD7501AYPS has sudden performance slumps with 1KB data chunks at reading and with 16KB data chunks at writing. This cannot be attributed to the interaction with the controller’s driver because, as we already know quite well, such driver-provoked slumps can only occur with 4KB data blocks (as indicated by the results of the Seagate Barracuda ES.2, for example).
In this test IOMeter is sending a stream of requests to read and write 512-byte data blocks with a request queue of 1 for 10 minutes. The total number of requests processed by the HDD is over 60 thousand, so we get a sustained response time that doesn’t depend on the HDD’s buffer size.
Seagate’s new series HDDs win this test at reading, enjoying a considerable lead over their opponents. The Western Digital WD7500AYPS is expectedly last due to its low spindle rotation speed. Surprisingly, the updated model from the same series has better results at both reading and writing, outperforming some of the 7200rpm models.
The HDDs from Samsung and Hitachi are in the lead at writing, indicating efficient deferred writing algorithms. The old drives from Seagate (7200.10 and ES series) and Western Digital’s RE2 are obviously poor here. Take note that every enterprise model is somewhat worse than the corresponding desktop model at writing: the efficiency of caching algorithms is lowered in them to ensure higher reliability.
Now we’ll see the dependence between the drives’ performance in random read and write modes on the size of the data block size.
We will discuss the results of the disk subsystems at processing random-address data in two versions. 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 depending on data-transfer rate in megabytes per second. This approach helps us evaluate the disk subsystem’s performance in two typical scenarios: working with small data chunks is typical for databases. The amount of operations per second is more important than sheer speed then. Working with large data blocks is nearly the same as working with small files, and the traditional measurement of speed in megabytes per second becomes more relevant.
IOMeter: Random Read, operations per second
The drives deliver similar speeds with small data chunks but a few models stand out. Seagate’s new models have high results while the Western Digital WD7500AYPS is rather slow. The updated WD7501AYPS is as fast as the other HDDs as if its spindle rotation speed were 7200rpm.
IOMeter: Random Read, MBps
The new Barracudas are unrivalled with large data chunks although the Samsung is good, too. The updated GP series model from Western Digital is surprisingly good again as it is not only faster than its mate but also outperforms Seagate’s 7200.10 and ES series drives and is almost as fast as the Western Digital RE2. But all these drives have a spindle rotation speed of 7200rpm rather than 5400rpm!
IOMeter: Random Write, operations per second
Seagate’s 7200.11 is good again. The enterprise ES.2 model is not so good at writing small data chunks, although outperforms most of the four-platter models. Surprisingly enough, the Hitachi competes with the leader and the Samsung follows the leaders closely. And the biggest surprise is that the WD7501AYPS is nearly as fast as the Samsung!
IOMeter: Random Write, MBps
Seagate’s new HDDs are in the lead when writing large data chunks, the enterprise version being somewhat slower than the desktop one as usual. The Western Digital Caviar Blue takes third place in tough competition with its RE2 series mate and with the Samsung drive.
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 the following links to view the tabled results:
We will build diagrams for request queue depths of 1, 16 and 256.
The Seagate 7200.11 (its server version shows middling performance again), Samsung and Hitachi are fighting for first place. We guess the Seagate should be considered the overall winner. Having the same speed of writing, it is ahead of its opponent under mixed loads (when there is about the same share of writes and reads) and at reading.
The WD7501AYPS is outstanding again as it takes fourth place, enjoying a considerable lead over all the four-platter models, except for the Hitachi, and outperforming the Seagate ES.
The Seagate 7200.11 is an obvious leader now that the load is increased. Its server version from the ES.2 series has improved its position relative to the other HDDs. Having the best speed at pure reading, it has a performance slump at 30-40% writes. Its pure writing speed is not what a leader should deliver, either.
It is more difficult to name the drives that take second and third places. The Samsung and Hitachi cope best with writing, but reading is more important for databases. Mixed loads are important too, and these drives both have a serious slump under mixed loads. Thus, second place must be given to the WD7501AYPS, the sensation of this test session. This HDD produces outstanding results indeed.
When the load grows even higher, we see a top three consisting of Seagate’s two new models and the updated Western Digital RE2-GP. The Hitachi is fourth, the Samsung is fifth, and the others are far behind.
Now let’s check out diagrams that show a graph with five queue depths for each of the HDDs. This representation is indicative of the efficiency of the drive’s firmware algorithms.
The Hitachi 7K1000 behaves like most other HDDs from this brand. It features effective deferred writing and good NCQ algorithms.
The Samsung is similar to the Hitachi but its NCQ algorithms are somewhat more effective and deferred writing is less aggressive.
Seagate’s early 750GB drives have a very phlegmatic temperament. The desktop 7200.10 and the server-oriented ES have a minimum of deferred writing and request reordering. We should note, however, that the server-oriented version is better at very high loads.
The new generation shows the results of the developer’s work. The new firmware features superb requests reordering. Perhaps Seagate has applied some technologies from Maxtor whose products used to boast excellent NCQ implementation? Anyway, requests reordering and deferred writing are top class in the new firmware. They would be just ideal if it were not for the slump at 30% reads.
Take note that while both models perform requests reordering with the same efficiency, deferred writing is more effective in the desktop 7200.11. The server-oriented Barracuda ES.2 probably doesn’t take too much data into the cache at writing in order to ensure higher reliability: these data are going to be irrecoverably lost in case of a power failure!
And here is another pair of identically behaving drives. Their results are better than those of Seagate’s older models but not by much. Deferred writing has low efficiency, and there is almost no requests reordering.
The older WD7500AYPS already indicated that Western Digital had done a good job on improving the firmware. This model reorders requests better than the above-discussed drives from the same maker. The updated WD7501AYPS has even higher efficiency of requests reordering (it is as efficient as that of Seagate’s new models) and features more efficient deferred writing. Judging by the increased read speed, the drive’s mechanical components have also been improved.
The multi-threaded tests simulate a situation when there are one to four clients accessing the hard disk at the same time. The depth of the outgoing request queue is varied from 1 to 8, and the address zones of the applications (called Workers in IOMeter) do not overlap.
You can follow the links below to see tables with results, but we’ll discuss diagrams for a request queue of 1 as the most illustrative ones. When the queue is longer, the speeds depend but little on the number of applications.
The drives with the highest recording density expectedly take top places when reading one thread but the situation changes dramatically when there are two read threads. Western Digital’s drives are the favorites now. We’ve got the same performance with any of the four models, including the two drives from the RE2-GP series that have a lower spindle rotation speed. The Hitachi and the new drives from Seagate pass the test successfully whereas the older drives from Seagate and the Samsung suffer a terrible performance hit.
The drives all slow down proportionally at three threads, the Hitachi being the only drive to have a heavier performance hit than the others. The addition of a fourth thread provokes changes in the standings. It looks like Seagate’s 7200.11 and ES.2 drives catch a second wind: they win first places as the consequence. The older drives from Seagate and the Samsung feel equally bad at any number of threads other than one.
It is different with multithreaded writing. We’ve got new leaders: the Samsung and the Hitachi. The former even accelerates when switching from one to two threads. Then, the new drives from Seagate are surprisingly slow. They have the lowest speed here, being outperformed by the older models from Seagate. Take note that the server versions are somewhat faster than their desktop counterparts.
The Samsung is 15MBps faster than the closest pursuer at three and four threads. Interestingly, Seagate’s 7200.11 and ES.2 are again somewhat faster at four than at three threads.
The drives are tested under loads typical of servers and workstations.
The names of the patterns are self-explanatory. The Workstation pattern is used with the full capacity of the drive as well as with a 32GB partition. The request queue is limited to 32 requests in the Workstation pattern.
The results are presented as performance ratings. For the File-Server and Web-Server patterns the performance rating is the average speed of the drive under every load. For the Workstation pattern we use the following formula:
Rating (Workstation) = Total I/O (queue=1)/1 + Total I/O (queue=2)/2 + Total I/O (queue=4)/4 + Total I/O (queue=8)/8 + Total I/O (queue=16)/16.
The Seagate ES.2 and Western Digital’s WD7501AYPS are the best drive at high loads. The latter has certainly become better since the update – you just somehow forget about its low spindle rotation speed. The Seagate 7200.11 is very good at low loads. By the way, the previous-generation pair of disks from Seagate behave in the same manner (the server version is better at high loads while the desktop version, at low loads) but their absolute results are lower, of course. The desktop version is always better in the pair of Western Digital’s RE2 and Caviar Blue drives.
Let’s see what performance rating we have.
Our rating system prefers drives that do well under low loads (such loads have a higher weight in the formula), so the Seagate 7200.11 takes first place. That’s why the WD7501AYPS outperforms the Seagate ES.2 and takes second place. The Hitachi is a disappointment as it is even inferior to the WD7500AYPS that has modest firmware and a spindle speed of 5400rpm.
When there are no write requests, Seagate’s newer drives are in the lead, the desktop version being better than the enterprise one. Third place is shared among the three drives from Western Digital (the WD7500AYPS is out of the competition) and the Hitachi. The Samsung isn’t far worse, either. It is slower at high loads only.
So, third place goes to the Western Digital RE2 according to our rating system. The other drives contending for third lose only by a few points, though.
Seven out of the ten drives produce similar results in the Workload pattern. The WD7500AYPS is last, the Seagate 7200.11 is first, while the Samsung is second. This is the load the Samsung drive proves to be good at.
The performance ratings are interesting as Seagate’s previous-generation drives have third and four places. Besides the WD7500AYPS, the group of losers now includes the Hitachi and the newest Seagate ES.2.
When the test zone is limited to 32 gigabytes, the Seagate ES.2 steps up to second position while the RE2-GP series drives from Western Digital both slow down. Their lower spindle rotation speed shows up here.
The standings of this test depend on sequential speeds of the drive whereas firmware is not important. The firmware algorithms can only help distinguish a drive among others that are equal to it in terms of sequential speed.
For this test two 32GB partitions are created on the disk and formatted in NTFS and then in FAT32. After that a file-set is created. It is then read from the disk, copied within the same partition and then copied into another partition. The time taken to perform these operations is measured and the speed of the disk 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.
We’d like to note that the copying test is indicative of the drive’s behavior under complex load. In fact, the HDD is working with two asynchronous threads (one for reading and one for writing) when copying files.
This test produces too much data, so we will only discuss the results achieved with the Install, ISO and Programs file-sets in NTFS. You can use the links below to view all the results:
The Samsung and Hitachi cope best with creating files but the two RE2-GP drives from Western Digital compete with the leaders when creating large files (the updated RE2-GP is expectedly faster a little). Seagate’s drives surprise us with their low results: the pair of new-generation models is even slower than their predecessors, obviously due to some flaws in their firmware.
The Samsung turns in a good performance at reading, too. It gives way to Seagate’s 7200.11 and ES.2 when processing very large files, but Seagate’s HDDs perform worse with smaller files. The Samsung, on the contrary, is very good irrespective of the size of the files. The standings of the HDDs when processing large files resemble what we saw in the sequential read test. With smaller files, the sequential speed alone is not enough. Therefore both RE2-GP from Western Digital outperform Seagate’s HDDs in the Programs pattern while the Hitachi takes second place.
The Samsung is the best when copying within the same partition. The performance of the other drives depends on the size of the files. The previous-generation drives from Seagate and the 7200rpm drives from Western Digital are good with large files while the Hitachi goes ahead with small files. The current generation of Seagate’s drives is slow, probably due to their low write speed.
The Samsung confirms its superiority in FC-Test in the Copy Far subtest.
PCMark04 benchmarks drives in four different modes: Windows XP Startup is the typical disk subsystem load at system startup; Application Loading is the disk activity at sequential starting-up and closing of six popular applications; File Copying measures the HDD performance when copying a set of files; the General Usage parameter reflects the disk activity in a number of popular applications. These four parameters are used to calculate the overall performance rating.
We ran each test ten times and averaged the results.
Hitachi drives used to dominate this test for a long time, but now the Hitachi is only second to the Samsung. The last drive in the top three is the server-oriented RE2 from Western Digital.
Well, the Hitachi regains its leadership in Application Loading. Take note that we’ve got the same top three as in the previous test. Interestingly, the WD Caviar Blue, the desktop mate of the WD RE2 which is third, is only second to last and has a much lower speed.
The standings in this subtest are the same as in FC-Test: the Samsung is first while the new drives from Seagate are slow, taking last place. It seems like something important is missing in their firmware: Seagate’s drives used to be strong in copying tests.
The Hitachi is in the lead again, followed by the Samsung. This time the Seagate 7200.11 and ES.2 drives pushed the Western Digital RE2 from third to fifth position.
The Samsung has the highest overall score whereas the Hitachi is only second. Take note that both drives from Western Digital’s RE2-GP series are quite competitive to most of their opponents that have a higher spindle rotation speed.
PCMark05 is an updated version of the previous benchmark. Instead of File Copying, there is now a File Write trace. A new trace called Virus Scan is added. Its name is self-explanatory.
Again, we performed each test ten times and averaged the results.
The leaders are the same as in the previous version of the benchmark but the Hitachi doesn’t lose its first place now.
The leading trio remains the same in the Application Loading subtest. The Hitachi is in the lead while the Western Digital RE2 is ahead of the Samsung, even though by a small margin.
There is nothing new in this subtest, too. The standings are exactly like in the previous version of PCMark.
Here is the new load. The high speeds indicate that this subtest checks the drive’s caching ability. The Samsung copes best with this load. Most of the other drives go close to each other. The WD7501AYPS makes use of its superb firmware to be just as fast as most of its opponents. The Western Digital RE2 finds itself among the losers for some reason.
This is another new load. The Samsung is in the lead again. The other places of the podium go to the new-generation drives from Seagate – the subtest seems to emulate the writing of rather large files. The Hitachi and WD Caviar Blue turn in good performances, too.
The overall result is logical: the Samsung and Hitachi take first places, enjoying a considerable lead over the others. Third place goes to the Seagate ES.2 that outperformed its desktop mate from the 7200.11 series.
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 is run 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.
Before proceeding to the results, we want to note one curious thing. We first benchmarked the WD7501AYPS with AHCI disabled. Having found our mistake, we retested the drive with AHCI enabled. And then we compared the results and found them to be very interesting. So, we offer both results for you to see the difference.
Strangely enough, the multithreaded load is handled better by those drives that had terrible results in the multithreaded read test, i.e. by the Samsung and Seagate Barracuda ES. Third place goes to the Hitachi. Seagate’s new-generation drives are slower than their predecessors due to different firmware – their higher recording density doesn’t help here.
The WD7501AYPS produces the same result with and without AHCI.
The gamer’s choice is obvious: the speeds of the Samsung, Hitachi and Seagate 7200.11 are similar and higher than those of the other drives.
AHCI provides a performance gain of over half a megabyte per second here.
Photographers should appreciate the Samsung and the Seagate 7200.11. The Hitachi rolled back to the bottom of the diagram, giving way to the Western Digital Caviar Blue.
AHCI slows the drive down somewhat here.
The Seagate 7200.10 proved to be the best drive for booting up Windows Vista. It is ahead of its server version as well as Seagate’s new-generation drives. The latter found this test to be too hard and lose to almost every other drive.
AHCI provides a performance gain of nearly 2MBps here. This helped the drive from Western Digital overtake the Seagate ES.2.
Samsung’s second place could be expected but the top position of the Seagate ES is a surprise. The desktop Barracuda 7200.10 isn’t brilliant whereas Seagate’s new-generation drives are in last places, being inferior even to the 5400rpm RE2-GP series drives from Western Digital.
We don’t know what the developer of the benchmark means by video editing but this process needs NCQ very much: the speed grows up by a half with enabled AHCI!
Working with video seems to be a very specific load. At least we’ve got the same leaders here. The results indicate that all the action takes place in the buffer memory as neither drive can work with the platters at such a high speed. Seagate should find the piece of code that was responsible for such superb optimization and add it into the new series.
AHCI is profitable again, providing a 50% performance gain.
This test has the same leaders, which are now closely followed by the Hitachi. Seagate’s new-generation drives are slow still. Firmware optimizations are a complex thing: having improved the performance in server tests, these drives have also begun to lose in PCMark.
AHCI is advantageous, producing a considerable performance gain.
The traditional leaders – the Samsung, Hitachi and Seagate 7200.11 – regain their leadership when loading applications.
We guess this test has also proved to you that AHCI is important.
The overall scores provide a general picture of performance, although do not reveal the weak and strong points of each drive. So, the Samsung is the winner and is followed by the Seagate ES which was especially good in this test. The newer ES.2 takes last place, behind both drives from the Western Digital RE2-GP series.
AHCI didn’t improve performance in every test, but was overall profitable. The overall performance increase is 18%. We guess that’s quite enough for you to remember to turn AHCI on.
Next goes our homemade test of defragmentation speed. We created a very defragmented file system on a 32GB partition of a hard 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 HDD we want to test. The tested HDD is connected to the mainboard’s SATA controller whose operation mode (AHCI/Standard SATA) is controlled from the mainboard’s BIOS. Next we run a FC-Test 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.
Thus, each HDD is tested twice: with the controller’s AHCI support enabled and disabled. You can refer to a dedicated article for details about this test.
The results follow below. Unfortunately, we couldn’t test both enterprise HDDs from Seagate in it:
This test may be questionable because its load is very specific. The Samsung wins it, especially with disabled AHCI. The WD RE2 and Hitachi performed well, too. On the losing side are both drives from Seagate (the newer 7200.11 took more time to defragment than the 7200.10) and the Western Digital Caviar Blue.
You can refer to our article called Hard Disk Drive Power Consumption Measurements: X-bit’s Methodology Indepth for details on this test. We’ll just list the specific modes we measure the power consumption in:
Before we discuss the results, we’d want to add a few words about the temperature factor. The temperature of the platters is what the service life of the platters and the accuracy of reading data depends on. But how can it be measured? If we measure it physically, with an external thermocouple, there is a question of the specific location. The top, bottom and side panels of the case and the PCB with the electronics can be used as such a location but it may be the top panel that is the hottest spot of one HDD whereas another HDD will be the hottest at its PCB. Measuring in multiple locations will produce a lot of numbers but won’t answer the question, what drive is hotter and needs more cooling.
Alternately, we could measure the temperature with the software, using the integrated thermal sensors. However, we can’t be sure about the accuracy of the sensor (every hardware tester is somewhat cautious about data offered by the manufacturer) and these sensors may be located at various places in the HDD. One maker measures the temperature of the heads, another measures the temperature of the platters, and a third maker will install the sensor into the electronics. How can such different data be compared?
Therefore we offer you the simplest method. According to the laws of physics, all the energy consumed by the HDD transforms into heat and is dissipated in the air. It may transform into heat in the chips, in the motor, in the heads actuator, but it is heat anyway. Thus, you can learn what HDD is hotter by comparing power consumption. Of course, the resulting temperature of the HDD and its components depends on cooling, but cooling is the responsibility of the end user. Our tests will show you how much heat must be dissipated from each specific HDD model.
Unfortunately, we don’t have the results for the Seagate Barracuda ES.2 but we suppose they are going to be similar to those of the Seagate 7200.11 because the electronical and mechanical sections of the two models are in fact identical.
Seagate’s previous-generation models and the economical WD7500AYPS prove to consume the most from the 12V rail. The updated version of Western Digital’s drive consumes less. The Samsung needs far less power than the others. Most of the drives have the same consumption from the 5V rail, which powers the electronics, but the Hitachi consumes about 50% more than the others.
Take note that this test shows the maximum currents. The overall consumption of each drive is changing in a complex manner and each drive takes a different amount of time to start up. Therefore you can’t measure the output power basing on these numbers.
Both RE2-GP drives from Western Digital are in the lead in idle mode: their reduced spindle rotation speed helped reduce the consumption on the 12V line. The updated WD7501AYPS is just brilliant as it needs little power from both 12V and 5V lines. As a result, it consumes 1.5 times less than its series mate and 2-2.5 times less then the 7200rpm models in idle mode. The Samsung and the new Seagate 7200.11 are the best among the fast drives: their electronics and mechanical parts both consume less power than in the other drives.
Random reading is the most power-consuming mode. There is nothing to optimize: the heads are moving wildly on the platters. It is clear that the Hitachi and the previous-generation drives from Seagate (which have almost identical results) are on the losing side. It is the mechanical components that consume the most, but the Hitachi also has voracious electronics. The two drives from Western Digital with a reduced spindle rotation speed are in the lead. The others are similar, with a total consumption of almost 11W.
The deferred writing optimizations help reduce power consumption in the random write mode. Of course, we have the same winners, but they consume more from the 5V line than the Seagate 7200.11 and the Samsung do. The latter are the best among the desktop drives: the new generation features not only the highest-capacity platters and better speed but also very economical components. The Hitachi and the Seagate ES are the worst drives in this test, consuming a lot from the 12V line.
The heads do not move much at sequential reading and the consumption on the 12V line drops down. As a result, the economical RE2-GP drives from Western Digital consume almost the same amount of power from the 12V and 5V lines. On the whole, we’ve got the same leaders and losers, but note that the consumption of the electronics from the 5V line contributes a lot to the overall result.
It’s the same thing with writing. Although the standings do not change, the Seagate 7200.11 and Samsung are very close to the economical models from Western Digital thanks to their low consumption on the 5V line.
Now we’ll try to sum all the results up.
The Hitachi Deskstar 7K1000 is just good. Like many other drives from this firm, it copes well with server loads, features superb deferred writing, processes small files quickly and can compete and beat its opponents in PCMark. But on the whole, the recording density of its platters is not high enough to be really competitive to the current generation of HDDs. The components should also be updated: the highest power consumption is not the result one could be proud of.
The Samsung F1 is very good. Yes, it failed the server tests as its firmware couldn’t cope with such load. But in most tests that resemble the typical load of home and office computers it was the winner or in the top three. This HDD also proved to have modest power consumption, so it is going to be rather cool. It is not without flaws, though. Besides its low server performance, it might do better in the multithreaded reading test. And its top speed should have been closer to that of the 1TB model of the same series.
Seagate’s HDDs should be discussed according to the specific generation. The generation of the Barracuda 7200.10 and Barracuda ES was the first to reach the capacity of 750GB but their performance is now disappointing. They took last places in most of our tests and may only be appealing now due to their low price.
The current generation, represented by the Barracuda 7200.11 and Barracuda ES.2, are the opposite of their predecessors. Their speed characteristics are outstanding. Seagate solved the problem of low speed of multithreaded reading and low performance in server tests. These HDDs were good in every test and were leaders in many of them, delivering terrific sequential read speeds. They are also among the best in terms of power consumption. So, these are two universal HDDs that are equally suitable for server applications and for desktop computers.
Western Digital’s Caviar Blue and RE2 turn to be average products that are neither record-breakers nor losers. As opposed to them, the Western Digital RE2-GP series deserves special attention because it includes two models behaving in a different manner. The WD7500ABYS delivers modest performance but features low power consumption (thanks to its spindle rotation speed of 5400rpm). The WD7501ABYS consumes little power too, but delivers higher performance. It is competitive to the other HDDs in the server tests, being even close to the leader in some of them. So, Western Digital did something miraculous to the firmware of this HDD. If your server needs a cold but rather fast HDD, this one should certainly be considered.
That’s all we’ve got to tell you today. We only want to add that this test session left a good overall impression because the developers not only increase the density of the platters but also work on the firmware. If they keep on doing like that, the upcoming generation of dual-platter 750GB drives is going to be tremendously fast!