7,200-RPM terabytes from Hitachi, Samsung, Seagate, and WD face off

Solid-state disks are easily one of the most exciting new technologies to grace the PC in recent years. The first examples didn’t have the capacity or performance to measure up to their exorbitant price tags, but great strides have been made with subsequent generations. These days, you can pick up a 128GB SSD that will easily trounce a mechanical hard drive for round about $350. That’s still a princely sum, although the value proposition isn’t as questionable as one might suspect.

If you’re planning on dropping more than a grand on a new rig, you’d be foolish not to at least consider pairing a low-capacity solid-state disk with secondary mass storage. But what if you’re spending less? There are cheaper SSDs that serve up 32 and 40GB capacities for around $100. That doesn’t leave a whole lot of room for games or other data on top of a typical operating system and applications payload, though. Besides, for a lot less than $100, you can have your pick of mechanical hard drives that spin a full terabyte of capacity at 7,200 RPM.

We’ve had terabyte hard drives for years now, and thanks to advances in areal density, the latest examples need only two platters. Those who have been following our storage coverage will know that we have a certain affinity for two-platter mechanical drives. While 3.5″ drive makers have stacked as many as five platters over the years, the additional media takes more power to rotate and typically turns up the volume on noise levels. As a result, two-platter drives usually offer the most attractive mix of performance, power consumption, acoustic profile, and overall capacity. Couple that with the lower cost per gigabyte that comes with being a few rungs down on the capacity ladder, and you’ve found the sweet spot for mechanical desktop storage.

For years, this sweet spot was dominated by Western Digital’s 640GB Caviars. Ignited by the original SE16 and capped by an eventual Black variant, WD’s last batch of two-platter Caviars boasted an unrivaled combination of exceptional all-around performance and low noise levels. We went on to recommend the drives in scores of system guides and eventually replaced them with the two-platter, 1TB Caviar Black 6Gbps when it debuted earlier this year.

As much as we like the new Black, I have to admit that I’m not as enamored with it as I was with the old 640GB models. The drive’s performance is excellent. However, the chattering staccato it plays while seeking is noticeably louder than the muted grinding of the 640GB Caviars. The terabyte Black doesn’t make enough noise to be an annoyance unless the rest of your system is already silent, but that’s enough of a chink in the armor to make one wonder if a better option exists. The market certainly isn’t short on alternatives from which to choose.

Hitachi, Samsung, and Seagate have two-platter terabyte offerings, too. Naturally, we had to find out which one is best, so we’ve rounded up the 1TB flavors of the Barracuda 7200.12, Caviar Black, Deskstar 7K1000.C, and Spinpoint F3 in an old-school mechanical throwdown. Keep reading to see which of these drives is worthy of the sweet spot crown.

Introducing the contenders
Unlike the solid-state disk market, the world of mechanical hard drives is dominated by two players: Seagate and Western Digital. In the first quarter of this year, those companies split just over 62% of the hard drive market nearly evenly. Hitachi carved itself 17.6% of the pie, while Samsung was stuck behind Toshiba/Fujitsu with less than 10% of the market.

Despite Samsung’s meager market share, the Spinpoint F3 has been one of the most hotly anticipated desktop drives of the last year thanks to favorable reviews of the last-gen Spinpoint F1. The F3 spent much of its early life painfully out of stock at the few online retailers that even listed it, but the new Spinpoint seems to finally be available with some consistency. Hitachi’s Deskstar 7K1000.C isn’t quite as fresh, but it’s been widely available for a while now. As you might have guessed, the 7K1000.C represents the company’s third-generation terabyte drive to hit 7,200-RPM.

The elder statesman of the bunch is the Barracuda 7200.12, which Seagate released before anyone else was spinning dual 500GB platters at full speed. A premium XT model has since been added to the Barracuda line, but it’s only being offered at 2TB with no plans to reach down to lower capacity points, so the 7200.12 has endured for more than 16 months now. That brings us back nicely to the 6Gbps Caviar Black, which is the youngest entry of the bunch, having popped onto the market just a few months ago.

Limiting our focus to two-platter, 1TB, 7,200-RPM mechanical hard drives makes this about as much of an apples-to-apples comparison as you’re going to get in the storage world. As you’ve no doubt deduced, each drive has an even terabyte of storage capacity spread across two platters rotating at 7,200-RPM. The Barracuda, Deskstar, and Spinpoint also have 32MB caches and 3Gbps Serial ATA interfaces. Western Digital offers a step up in both categories, equipping its latest Caviar Black with 64MB of cache and next-gen 6Gbps SATA connectivity.

The value of those upgrades is probably dubious at best. For years, Western Digital downplayed the benefit of larger caches, and it only put 32MB into the latest VelociRaptor. The extra cache shouldn’t be a hindrance. Besides, it’s the only part of the drive with a shot at exploiting the 6Gbps interface. The Black’s platters can only sustain transfer rates up to 126MB/s, according to WD’s data sheets, so they can’t saturate a first-gen 1.5Gbps SATA link, let alone the 3Gbps interface used by the others. Unsurprisingly, when we reviewed the Caviar earlier this year, its 6Gbps interface wasn’t much help. The fact that all of today’s test data was collected using a 3Gbps SATA controller shouldn’t unfairly penalize the Caviar.

Although all four drives stack two 500GB platters, the media itself differs slightly from one manufacturer to the next. The key measure here is areal density, which is usually expressed as the number of bits squeezed into each square inch of surface area. A higher areal density allows the drive head to access a given amount of data over a shorter physical distance, enabling faster sequential transfers. There’s a catch: the more tightly the bits are packed, the more difficult it becomes to seek out individual data points. That’s likely why so many drives offer great sequential throughput but mediocre random access times.

Maintaining its streak of one-upmanship, the Caviar Black has a higher areal density than the rest at 400 Gb/in². That’s a substantial advantage over the Deskstar 7K1000.C, which packs only 352 gigabits into every square inch of platter real estate. The Barracuda 7200.12 slots in below the Deskstar with an areal density of 329 Gb/in².

Samsung hasn’t published the areal density of the platters in the Spinpoint F3. We’ve asked the company’s representatives for additional details several times now, but thus far all we’ve been told is that the platters pack 500GB each. For a two-platter terabyte drive? Ya think?

When you line them up, the Caviar Black has the edge on every front except one: price. The extra scratch buys spec sheet bragging rights for your next forum flame war, and more importantly, an additional two years of warranty coverage. You’ll pay $20 less for the ‘cuda and save a cool $25 with Deskstar and Spinpoint—sizable discounts given the fact the cheapest options will set you back just seventy dollars.

With the best specs of the lot, the Caviar is the obvious favorite. Besides, it has a hefty price premium to justify. The pressure is off the other contenders, which are nicely poised for an upset thanks to much lower prices. Without further ado, let’s make our way to the proving grounds to see how these drives stack up.

Our testing methods
Before dipping into pages of benchmark graphs, let’s set the stage with a quick look at other the players we’ve assembled for comparative reference. We’ve called up a wide range of competitors, including a selection of desktop hard drives, traditional notebook drives, Seagate’s Momentus XT hybrid, and a cubic assload of pure solid-state goodness. Below is a chart highlighting some of the key attributes of the contenders we’ve lined up opposite our quartet of two-platter terabytes.

On the SSD front, we’ve collected all the other relevant players, including drives based on Indilinx, Intel, JMicron, Marvell, SandForce, and Toshiba controllers. Although it might not seem like a fair fight, we’ve also thrown in results for a striped RAID 0 array built using a pair of Intel’s X25-V SSDs. The X25-V only runs a little more than $100 online, making multi-drive RAID arrays affordable enough to be tempting for desktop users. Our X25-V array was configured using Intel’s P55 storage controller, the default 128KB stripe size, and the company’s latest 9.6.0.1014 Rapid Storage Technology drivers.

The block-rewrite penalty inherent to SSDs and the TRIM command designed to offset it both complicate our testing somewhat, so I should explain our SSD testing methods in greater detail. Before testing the drives, each was returned to a factory-fresh state with a secure erase, which empties all the flash pages on a drive. Next, we fired up HD Tune and ran full-disk read and write speed tests. The TRIM command requires that drives have a file system in place, but since HD Tune requires an unpartitioned drive, TRIM won’t be a factor in those tests.

After HD Tune, we partitioned the drives and kicked off our usual IOMeter scripts, which are now aligned to 4KB sectors. When running on a partitioned drive, IOMeter first fills it with a single file, firmly putting SSDs into a used state in which all of their flash pages have been occupied. We deleted that file before moving onto our file copy tests, after which we restored an image to each drive for some application testing. Incidentally, creating and deleting IOMeter’s full-disk file and the associated partition didn’t affect HD Tune transfer rates or access times.

Our methods should ensure that each SSD is tested on an even, used-state playing field. However, differences in how eagerly an SSD elects to erase trimmed flash pages could affect performance in our tests and in the real world. Testing drives in a used state may put the TRIM-less Plextor SSD at a disadvantage, but I’m not inclined to indulge the drive just because it’s using a dated controller chip.

With few exceptions, all tests were run at least three times, and we reported the median of the scores produced. We used the following system configuration for testing:

You can read more about the hardware that makes up our twin storage test systems on this page of our VelociRaptor VR200M review. Thanks to Gigabyte for providing the twins’ motherboards and graphics cards, OCZ for the memory and PSUs, Western Digital for the system drives, and Thermaltake for SpinQ heatsinks that keep the Core i5s cool.

We used the following versions of our test applications:

• WorldBench 6
• Intel IOMeter 2006.07.27
• Xbit Labs File Copy Test 0.3
• HD Tune 4.01
• Visual Studio 2008 with 03-23-2010 Firefox source
• Call of Duty: Modern Warfare 2
• Crysis Warhead

The test systems’ Windows desktop was set at 1280×1024 in 32-bit color at a 75Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.

Most of the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.

HD Tune
We’ll kick things off with HD Tune, our synthetic benchmark of choice. Although not necessarily representative of real-world workloads, HD Tune’s targeted tests give us a glimpse of a drive’s raw capabilities. From there, we can explore whether the drives live up to their potential.

The second line graph is the same as the first, but the SSDs have been removed for a better look at the mechanical terabyte drives. Among them, Samsung’s Spinpoint F3 manages the highest read speeds across its entire capacity. The drive’s transfer rate oscillates quite a bit more than that of the others, but even in the deepest valleys, the Spinpoint is still a step ahead of its closest competition, the Deskstar 7K1000.C.

Slightly trailing the Samsung and Hitachi drives are entries from Seagate and WD. Looking at average read rates, the Barracuda and Caviar are 11 and 14MB/s slower than the Spinpoint, respectively. That’s a particularly poor start for the Caviar, whose high areal density should give it an advantage with sequential transfers.

Switching gears from reading to writing doesn’t change the picture much for our mechanical terabyte drives. The Spinpoint still leads the Deskstar, albeit by a slightly smaller margin, while the Caviar and ‘cuda remain in third and fourth place, respectively.

As we saw with reads, the Deskstar is the only drive to maintain a smooth transfer rate curve over the extent of its capacity. The Spinpoint’s write speed bounces around seemingly at random, although thankfully without much amplitude. Meanwhile, the Caviar and Barracuda see their write speeds drop in small steps at several points across their respective capacities.

Next up: some burst-rate tests that should test the cache speed of each drive. We’ve omitted the X25-V RAID array from the following results because it uses a slice of system memory as a drive cache.

Notch two wins for the Barracuda 7200.12. Seagate tops our collection of terabyte mechanical drives with a 206MB/s burst speed for both reads and writes. The Spinpoint is more than 10MB/s off the pace, followed by the Caviar and Deskstar.

Our HD Tune tests conclude with a look at random access times, which the app separates into 512-byte, 4KB, 64KB, and 1MB transfer sizes. Let’s start with reads.

The Caviar Black is 1-2 milliseconds ahead of its rivals at three of four transfer sizes and ties for the lead in the fourth. Behind it, the Deskstar and Spinpoint are locked in a dead heat for second place. Only with the 1MB transfer size does the F3 pull itself even with the Caviar for a share of the lead.

Things don’t look as rosy for the Barracuda, which matches the Deskstar and Spinpoint at the 64KB transfer size but falls a millisecond behind in the rest.

Caviar tastes sweeter with random writes. The terabyte Black is out in front of the competition at all four transfer sizes. The Deskstar and Spinpoint trade second place back and forth. The Samsung is quicker at the smallest and largest transfer sizes, and the Hitachi comes out on top in the middle two.

Once more, the Barracuda 7200.12 has the slowest access times of our four contenders. The ‘cuda squeaks by the Deskstar with the 1MB transfer size, but it’s at least a millisecond off the pace otherwise.

File Copy Test
Since we’ve tested theoretical transfer rates, it’s only fitting that we follow up with a look at how each drive handles a more typical set of sequential transfers. File Copy Test is a pseudo-real-world benchmark that times how long it takes to create, read, and copy files in various test patterns. We’ve converted those completion times to MB/s to make the results easier to interpret.

Windows 7’s intelligent caching schemes make obtaining consistent and repeatable performance results rather difficult with FC-Test. To get reliable results, we had to drop back to an older 0.3 revision of the application and create our own custom test patterns. During our initial testing, we noticed that larger test patterns tended to generate more consistent file creation, read, and copy times. That makes sense, because with 4GB of system memory, our test rig has plenty of free RAM available to be filled by Windows 7’s caching and pre-fetching mojo.

For our tests, we created custom MP3, video, and program files test patterns weighing in at roughly 10GB each. The MP3 test pattern was created from a chunk of my own archive of ultra-high-quality MP3s, while the video test pattern was built from a mix of video files ranging from 360MB to 1.4GB in size. The program files test pattern was derived from, you guessed it, the contents of our test system’s Program Files directory.

Even with these changes, we noticed obviously erroneous results pop up every so often. Additional test runs were performed to replace those scores.

The Spinpoint’s pack-leading sequential transfer rates in HD Tune were no fluke. In FC-Test, the F3 hits much higher file creation speeds than its rivals with all three file sets. The Spinpoint’s advantage ranges from 18 to 32MB/s depending on the file set, with the Barracuda slotting into second place across the board. Hot on its heels is the Caviar Black, which has a substantial lead over the 7K1000.C.

Impressively, the Spinpoint F3 is actually the fastest mechanical hard drive of the bunch. Not even WD’s flagship Caviar Black 2TB or the latest VelociRaptor can match the Samsung’s file creation speeds.

When tasked with reading the same files it created, the Spinpoint continues to churn through file sets quicker than the competition. The gaps to second and third place are much smaller this time around, as the Caviar and Deskstar trade blows in a battle for second best.

In a bit of a surprise, the Barracuda stumbles mightily—and consistently—with the MP3 and program file sets. We’ve seen the ‘cuda fare much better with these file sets when running Intel’s storage controller drivers, but we’re using the Microsoft AHCI drivers built into Windows 7 this time around. For what it’s worth, the 7200.12 isn’t the first storage device to trip up in FC-Test with the Microsoft drivers.

Copy tests are particularly challenging because they ask drives to switch back and forth between reads and writes. The Spinpoint was the fastest in the individual read and write tests, so its place at front of the pack in the copy tests should surprise no one.

We actually have a three-way tie for second place in the MP3 file set, and the Deskstar and ‘cuda hold onto that tie for the rest of the copy tests. That duo proves a little too quick for the Caviar, which falls into last place with the program files and video file sets.

In case you hadn’t noticed, the Spinpoint happens to be the fastest mechanical drive overall in FC-Test’s copy, er, tests. The VelociRaptor VR200M does have faster read speeds, but it has only a fraction of the F3’s capacity at several times the cost.

File copy speed
Although FC-Test does a good job of highlighting how quickly drives read, write, and copy different types of files, the app is antiquated enough to completely ignore the command queuing logic built into modern hard drives and SSDs. FC-Test only uses a queue depth of one, while Native Command Queuing can stack up to 32 I/O requests when asked. To get a better sense of how these drives react when moving files around in Windows 7, we performed a set of hand-timed copy tests with 7GB worth of documents, digital pictures, MP3s, movies, and program files. These files were copied from the drive to itself to eliminate any other bottlenecks.

We run this test on SSDs in a factory fresh and simulated used state since there are often performance differences between those two conditions. To put our SSDs into a simulated used state, I run our IOMeter workstation access pattern with 256 concurrent I/O requests for 30 minutes before launching into a second batch of copy tests.

IOMeter creates a massive test file that spans the entirety of a drive’s capacity, and deleting it to make room for a batch of copy tests nicely puts solid-state disks into a tortured used state. What we’ve essentially done here is filled all of an SSD’s flash pages, subjected the drive to a punishing workload with a highly-randomized access pattern, and then marked all of the flash pages as available to be reclaimed by garbage-collection or wear-leveling routines.

Mechanical hard drives aren’t subject to the block-rewrite penalty that causes SSD performance degradation as flash pages become occupied, so there’s no difference between their fresh- and used-state performance. We’ve double-checked to be sure. To avoid confusing the issue, we’ve omitted the fresh-state copy speeds of the SSDs in the graph below.

Another test that leans heavily on sequential throughput, another win for the Spinpoint F3. The Samsung drive transfers our eclectic file set 11MB/s faster than its nearest rival, the 7K1000.C. The Caviar Black and Barracuda 7200.12 aren’t far behind the Deskstar, but they’re well short of the Spinpoint.

Perhaps more impressive than the Spinpoint’s continued domination is the fact that it’s faster than both VelociRaptors and the Caviar Black 2TB. Not bad for $70.

Application performance
We’ve long used WorldBench to test performance across a broad range of common desktop applications. The problem is that few of those tests are bound by storage subsystem performance—a faster hard drive isn’t going to improve your web browsing or 3ds Max rendering speeds. A few of WorldBench’s component tests have shown favor to faster hard drives in the past, though, so we’ve included them here.

Among our application tests, only WorldBench’s Photoshop and Nero components show much separation between the terabyte drives, let alone the rest of our results. In Photoshop, the Spinpoint reigns supreme again, just ahead of the Caviar. The Deskstar and Barracuda are a little slower, but both bounce back in Nero. The 7K1000.C actually leads the quartet in that test, while the other three finish within four seconds of each other.

Boot and load times
Our trusty stopwatch makes a return for some hand-timed boot and load tests. When looking at the boot time results, keep in mind that our system must initialize multiple storage controllers, each of which looks for connected devices, before Windows starts to load. You’ll want to focus on the differences between boot times rather than the absolute values.

This boot test starts the moment the power button is hit and stops when the mouse cursor turns into a pointer on the Windows 7 desktop. For what it’s worth, I experimented with some boot tests that included launching multiple applications from the startup folder, but those apps wouldn’t load reliably in the same order, making precise timing difficult. We’ll take a look at this scenario from a slightly different angle in a moment.

The Caviar Black finds some redemption when booting Windows 7, which it does a full two seconds faster than the Spinpoint and ‘cuda. Another half second passes before the Deskstar finishes the boot process.

A faster hard drive is not going to improve frame rates in your favorite game (not if you’re running a reasonable amount of memory, anyway), but can it get you into the game quicker?

In our level load tests, the big dogs split the lead. The ‘cuda is a smidgen quicker than the Caviar in Modern Warfare 2 but a fraction of a second behind in Crysis Warhead. Bronze goes to the Spinpoint in both games.

I’m a little surprised to see the Deskstar with such an atrocious Crysis load time, but the 7K1000.C was consistently sluggish through multiple test runs. The drive is a full eighteen seconds slower than the Spinpoint and nearly nine seconds behind a 5,400-RPM notebook hard drive.

Disk-intensive multitasking
TR DriveBench allows us to record the individual IO requests associated with a Windows session and then play those results back on different drives. We’ve used this app to create a new set of multitasking workloads that should be representative of the sort of disk-intensive scenarios folks face on a regular basis.

Each workload is made up of two components: a disk-intensive background task and a series of foreground tasks. The background task is different for each workload, but we performed the same foreground tasks each time.

In the foreground, we started by loading up multiple pages in Firefox. Next, we opened, saved, and closed small and large documents in Word, spreadsheets in Excel, PDFs in Acrobat, and images in Photoshop. We then fired up Modern Warfare 2 and loaded two special-ops missions, playing each one for three minutes. TweetDeck, the Pidgin instant-messaging app, and AVG Anti-Virus were running throughout.

For background tasks, we used our Firefox compiling test; a file copy made up of a mix of movies, MP3s, and program files; a BitTorrent download pulling seven Linux ISOs from 800 connections at a combined 1.2MB/s; a video transcode converting a high-def 720p over-the-air recording from my home-theater PC to WMV format; and a full-disk AVG virus scan.

DriveBench produces a trace file for each workload that includes all IOs that made up the session. We can then measure performance by using DriveBench to play back each trace file. During playback, any idle time recorded in the original session is ignored—IOs are fed to the disk as fast as it can process them. This approach doesn’t give us a perfect indicator of real-world behavior, but it does illustrate how each drive might perform if it were attached to an infinitely fast system. We know the number of IOs in each workload, and armed with a completion time for each trace playback, we can score drives in IOs per second.

Below, you’ll find an overall average followed by scores for each of our individual workloads. The overall score is an average of the mean performance score in each multitasking workload.

DriveBench doesn’t produce reliable results with Microsoft’s AHCI driver, forcing us to obtain the following performance results with Intel’s 9.6.0.1014 RST drivers. We couldn’t get DriveBench to play nicely with our the X25-V RAID config, either, which is why it’s not listed in the graphs below. The app will only run on unpartitioned drives, so we tested drives after they’d completed the rest of the suite.

Samsung scores well in our disk-intensive multitasking benchmark, leading the Barracuda by roughly the same margin that the ‘cuda enjoys over the Caviar Black. The Deskstar finds itself bringing up the rear again, although mercifully only by a slim margin.

Let’s break down the overall average into individual test results to see if anything stands out.

The Spinpoint predictably looks pretty good overall, but it falls to the bottom of the heap with our transcoding workload. WD captures the top spot among terabyte mechanicals with that workload, and the Caviar spends most of its time chasing the Spinpoint and Barracuda with the others. However, it doesn’t deal with the virus scanning workload as gracefully as the other drives.

There are no surprises with the Deskstar, at least. It’s the slowest terabyte with almost every workload and never claws higher than third place.

Curious to see whether removing the multitasking element of these tests would have any bearing on the standings, I recorded a control trace without a background task.

Now that’s interesting. With our secondary workload taken out of the equation, the Caviar Black vaults into the lead just a sliver ahead of the Barracuda 7200.12. At a greater distance, the Spinpoint and Deskstar fill out the final two slots among our terabyte contenders.

DriveBench lets us start recording Windows sessions from the moment the storage driver loads during the boot process. We can use this capability to take another look at boot times, again assuming our infinitely fast system. For this boot test, I configured Windows to launch TweetDeck, Pidgin, AVG, Word, Excel, Acrobat, and Photoshop on startup.

Our two-platter foursome is tightly bunched in the Windows 7 startup workload. The Caviar edges out the ‘cuda again, and the Spinpoint and Deskstar aren’t far behind.

IOMeter
Our IOMeter workloads are made up of randomized access patterns, presenting a good test case for both seek times and command queuing. The app’s ability to bombard drives with an escalating number of concurrent IO requests also does a nice job of simulating the sort of demanding multi-user environments that are common in enterprise applications.

SSDs are orders of magnitude faster than mechanical hard drives in this test, and that makes graphing the results rather challenging. In this first set of graphs, I’ve omitted the SSDs so we can actually see what’s going on with the other drives. Keep scrolling for an epic solid-state beatdown.

Through all four access patterns, the Caviar Black 1TB maintains a notably higher transaction rate than any of its terabyte foes. This is expected behavior given Western Digital’s long history of repurposing Caviar desktop drives for enterprise duty as RE models—or is it the other way around? Either way, the Caviar is clearly better equipped than the other terabyte models to handle demanding multi-user loads.

The Deskstar and Barracuda are closely matched, and I’m not sure there’s a definitive winner between them. Samsung’s in the mix, too, but only in web server access pattern. With the other three, the Spinpoint’s performance levels off and actually begins to drop as we progress beyond 32 concurrent I/O requests. The web server access pattern is made up exclusively of read operations, while the others have a mix of reads and writes, suggesting that the F3’s problem lies with the latter.

For those looking to justify an SSD purchase, behold:

Um, yeah. Let’s move on.

Power consumption
For our power consumption tests, we measured the voltage drop across a 0.1-ohm resistor placed in line with the 5V and 12V lines connected to each drive. We were able to calculate the power draw from each voltage rail and add them together for the total power draw of the drive. Drives were tested while idling and under an IOMeter load consisting of 256 outstanding I/O requests using the workstation access pattern.

The Deskstar consumes less power than the other desktop terabytes at idle, yet it’s the thirstiest when hammered by our IOMeter load. Only fractions of a watt separate the Spinpoint and ‘cuda as they swap places between their idle and load states. The Caviar’s power consumption is consistently on the high side, though.

Noise levels
Noise levels were measured with a TES-52 Digital Sound Level meter 1″ from the side of the drives at idle and under an HD Tune seek load. Drives were run with the PCB facing up.

Our noise level and power consumption tests were conducted with the drives connected to the motherboard’s P55 storage controller.

I’ve consolidated the solid-state drives here because they’re all completely silent. The SSD noise level depicted below is a reflection of the noise generated by the rest of the test system, which has a passively-cooled graphics card, a very quiet PSU, and a nearly silent CPU cooler.

That’s right. Not only is the Spinpoint largely the fastest drive of the terabyte bunch, it’s also the quietest at idle and under load. Not by insignificant margins, either. The F3 is a full decibel and a half quieter than the Caviar at idle and three decibels quieter than its closet competitor while seeking.

From a few feet away, it’s very difficult to detect a difference in idle noise levels between these four 1TB drives. The Deskstar and Caviar are noticeably louder when chewing through seek loads, though. Fortunately, there’s a way to quiet the drives down.

Most mechanical hard drives have an Automatic Acoustic Management (AAM) value that can be set between 128 and 254. Manipulating this setting tends not to affect idle noise levels, but it can dramatically impact seek noise and access times. To get an idea of the sort of performance and acoustic range available with our collection of mechanical drives, we’ve tested the seek noise level and random access time of each at the extremes of the AAM scale. By default, all of the terabyte drives had AAM disabled or set to 254, which is the most aggressive seek setting.

The seek noise levels of the Caviar and Deskstar can be reduced dramatically by fiddling with their AAM settings. Most impressive is the Western Digital drive, which goes from loudest to quietest among our selection of 7,200-RPM terabytes. Sliding the AAM slider down to 128 cuts more than three decibels off the 7K1000.C’s seek chatter. Interestingly, though, there’s little change in seek noise when the AAM settings are tweaked on the Spinpoint and ‘cuda.

So what’s the silence cost on the performance front? Quite a lot if you’re a Caviar Black or Deskstar 7K1000.C. Even the Spinpoint loses a few milliseconds, despite enjoying only a minimal drop in noise levels. The Barracuda’s seek times are just as impervious to AAM meddling as its noise levels.

The value perspective
After spending pages rifling through a stack of performance graphs, it’s time to broaden our horizons a little and take each drive’s price into consideration. First, we’ll look at capacity per dollar.

To establish an even playing field for all the contenders, we’re using Newegg pricing for all the drives. Mail-in rebates weren’t included in our calculations. Rather than relying on manufacturer-claimed capacities, we gauged each drive’s capacity by creating an actual Windows 7 partition and recording the total number of bytes reported by the OS. Having little interest in the GB/GiB debate, I simply took that byte total, divided by a Giga (10⁹), and then by the price. The result is capacity per dollar that, at least literally, is reflected in gigabytes.

No surprises here. With equal capacities, it all comes down to price with our terabyte mechanicals. The Deskstar and Spinpoint are $5 cheaper than the Barracuda, putting them comfortably in the lead. The terabyte Caviar Black’s relatively high price does it no favors here.

Overall performance per dollar is up next, but before we get there, we need to come up with an overall performance score for each drive. Using a single number to represent a drive’s performance across a range of different benchmark tests can be tricky business. After reading through numerous papers on the subject, we’ve settled on calculating a harmonic mean of all the results you’ve seen today. A harmonic mean can be useful for quantifying overall performance for a benchmark suite when individual test results can be compared to a reference baseline, and it’s not prone to being skewed by the fact that we have performance differences of several orders of magnitude in some cases. We just happen to have a full suite of results normalized to a performance baseline provided by an ancient 2.5″, 4,200-RPM IBM Travelstar mobile drive, and as you’ll see in a moment, the harmonic mean generates an overall score that nicely tracks with expectations based on the performance we’ve observed thus far.

I should note that we considered using an arithmetic average to calculate our overall score. However, this simple mean is easily skewed by the enormous performance gaps in IOMeter and HD Tune’s random access time tests, which are several orders of magnitude larger than the performance deltas in the other tests. The resulting overall score doesn’t track with expectations based on the performance we’ve already quantified. Weighting the average to account for those orders-of-magnitude differences would have been arbitrary at best, so we’ve settled on a harmonic mean, which seems to provide useful results.

Our overall score includes individual results for DriveBench and IOMeter rather than the averages we presented in the first set of value graphs. There are five DriveBench multitasking loads and four IOMeter access patterns, giving us a total of 19 test results from which to calculate the harmonic mean. This collection of tests is a little biased towards random access patterns rather than sequential transfers, but we think it strikes a good balance for drives that will store a system’s OS and applications. The power-efficiency and noise-level results have been left out to keep this a strictly performance-per-dollar affair.

Because they had to sit out at least one of the tests that make up our overall average, the PX-128M1S and X25-V RAID array haven’t been included in the graphs below. We wouldn’t recommend the former, anyway, and with two drives at its disposal, the RAID config would’ve had an unfair advantage—you know, like it’s had all day already.

Our collection of terabyte offerings clusters near the bottom of the pile in terms of overall performance. The Caviar Black actually takes the lead ahead of the Spinpoint here, no doubt aided by its strong performance in IOMeter, which is a large component of this overall measure. Behind the Spinpoint, the Barracuda and Deskstar are effectively tied.

SSDs obviously have a huge advantage when we consider overall performance. However, capacity is an equally important component of any storage device. We’ve divided each drive’s overall performance score by its cost per gigabyte to get a look at overall performance per dollar per gigabyte. Try saying that five times fast.

Samsung returns to the front of the field in our bar graph, but the real action is in the scatter plot, which nicely illustrates the intersection of performance and cost per gigabyte for each drive. The mechanical drives are clustered in the lower left corner of the plot, and they’re pretty close with the exception of the VelociRaptors. If you squint, you can see that the Spinpoint offers a small step up in performance over the Deskstar at the same cost. The Caviar’s performance score is higher, but a loftier price moves that particular data point a little to the right on the plot.

Another way to look at this data is to divide each drive’s performance by the cost of a complete system built around it. The aim here is to determine whether spending a little (or a lot) more makes sense when the price premium is absorbed as part of the cost of a complete system. The step up from a $70 drive to a $95 one is hardly daunting to start, and once you factor in the cost of a complete build, the price difference practically disappears.

For our system price calculations, we’ve used our test rig as the inspiration for a base config, to which the price of each drive will be added. Our example system includes a Core i5-750, a P55-based ASUS P755D-E motherboard, 4GB of DDR3-1333 memory, a passively-cooled Radeon HD 4850, Antec’s Sonata III enclosure, and Windows 7. Its base price is $844.94, although you’ll probably want to tack on the cost of secondary mass storage for configurations that will use an SSD.

Clearly, the large amounts of additional performance available with an SSD start to make some sense in the context of a larger system budget.

Between terabyte mechanical drives, the Caviar Black offers slightly better overall performance than its rivals at a marginally higher price (at least within the context of a complete build). The Spinpoint leads the others, albeit barely.