Western Digital’s VelociRaptor 1TB hard drive

Since the first Western Digital Raptor was released in early 2003, a new version has arrived roughly every two years. The initial incarnation was the first Serial ATA hard drive to spin its platters at 10,000 RPM. With performance that outclassed its 7,200-RPM competition but only 37GB of storage, the Raptor became the original enthusiast’s boot drive.

In the years since, WD has kept the drive’s spindle speed steady while increasing its capacity. A 74GB model soon followed, and the Raptor jumped to 150GB after that. Then, in 2008, the drive shrunk down to a 2.5″ form factor and got a new name: the VelociRaptor. Despite its smaller size, the VelociRaptor retained the original’s 10k-RPM rotational speed and bumped the capacity up to 300GB. Two years later, a 600GB version of the drive was released.

Betcha can’t guess what’s next.

Assuming you’ve read the title, I suppose you can. Western Digital has a new VelociRaptor with an even terabyte of storage capacity. This latest addition to WD’s high-performance desktop line enters a very different market than the one encountered by its forebear. Solid-state drives have gotten a lot faster in the past two years, and crucially, their prices have fallen by about half. At the same time, mechanical hard drive makers have seemingly become less interested in high-performance desktop drives, instead preferring to focus on notebook models and slower products destined for external enclosures.

Is there a place for the VelociRaptor in the evolving PC storage ecosystem? We’ve been testing the drive against its mechanical, hybrid, and solid-state peers to find an answer. Read on to see what we’ve learned.

New upgrades for an old favorite

Western Digital is positioning the VelociRaptor as a high-performance product for enthusiasts, content-creation professionals, and workstations. The Raptor’s real mission has always been enterprise applications, though. As it’s done with past 10k-RPM drives, WD will surely release a version of the new VelociRaptor specifically tailored for servers. Those enterprise aspirations are what drove WD to shrink the original 3.5″ Raptor down to the 2.5″ form factor that has come to define the VelociRaptor.

The VelociRaptor 1TB still slips into 3.5″ drive bays thanks to its IcePack enclosure, which includes the appropriate mounting holes and port placements for the larger form factor. Removing the 2.5″ drive from this sled requires little more than loosening a few Torx screws. One of those screws is covered by a warranty voiding sticker, though; remove with caution.

As much as I love the sinister-looking teeth that line the IcePack, I’d be tempted to run the VelociRaptor naked. Most contemporary cases already come with 2.5″ mounting hardware to accommodate solid-state drives. There’s a certain elegance to the smaller form factor, too, although the VelociRaptor is thicker than the average 2.5-incher. The drive measures 15 mm thick, while SSDs and notebook drives typically have a thickness of 9.5 mm.

The 2.5″ format allows enterprise versions of the VelociRaptor to be packed tightly into rack-mounted servers. Those environments are filled with vibration from adjacent drives, which the VelociRaptor combats with a Rotary Acceleration Feed Forward mechanism that adjusts the height of the drive head based on vibration data collected from linear accelerometers on the circuit board. WD’s NoTouch ramp load feature moves the drive head completely off the platters when the disk is idling. Keeping the drive head from making contact with the platters purportedly reduces wear and lowers the chance of a catastrophic head crash.

Like the Caviar Black 2TB desktop drive, the VelociRaptor 1TB situates its drive head at the end of a dual-stage actuator. The first stage is similar to the actuators on typical hard drives; it gets the drive head into the right zip code. Stage two involves a piezoelectric motor that allows the drive head to zero in on an individual address. If the next address is within the same zip code, the VelociRaptor can jump to it using only the actuator’s second stage.

  VelociRaptor VR200M VelociRaptor 1TB
Interface 6Gbps Serial ATA
Spindle speed 10,000 RPM
Platter size 200GB 333GB
Available capacities 450, 600GB 250, 500GB, 1TB
Cache size 32MB 64MB
Maximum data rate 145MB/s 200MB/s
Idle power 4.3W 4.2W
Read/write power 6.2-6.8W 5.1-5.8W
Idle acoustics 27 dBA 30 dBA
Read/write acoustics 34 dBA 37 dBA
Warranty Five years

The fine-grained control offered by the dual-stage actuator becomes increasingly important as areal densities rise, making the tracks on the platter narrower and the individual bits smaller. We’ve never been able to coax Western Digital into revealing the exact areal density of the platters in its VelociRaptor drives. However, the company did confirm that the new model uses a trio of 333GB platters. Since the old one had 200GB platters, it looks like WD has increased the areal density by roughly 67%. The resulting increase in linear density is what allows the VelociRaptor 1TB to offer a higher maximum data rate than its predecessor; the more bits that pass under the drive head with each revolution, the higher the sequential throughput

On its own, the VelociRaptor’s 333GB platter capacity isn’t all that impressive. WD’s Scorpio Blue manages to squeeze 500GB onto its 2.5″ platters, and the Scorpio Black’s platters pack 375GB. However, those drives spin at only 5,400 and 7,200 RPM, respectively. The VelociRaptor’s media rotates at a much faster 10,000 RPM. At that speed, the outer edge of the platter is moving at around 75 miles an hour.

Like the Scorpios and most of WD’s latest models, the VelociRaptor’s platters use Advanced Format. This standard replaces the 512-byte sectors of old with 4KB ones that make more efficient use of the drive media. The larger sectors dedicate fewer blocks to address and error correction data, leading to a 7-11% increase in storage capacity, according to WD.

The last upgrade for the VelociRaptor is an increase in cache size from 32 to 64MB. That cache will be the only part of the drive capable of making use of the 6Gbps Serial ATA interface, which offers much more bandwidth than the drive’s maximum data rate.

Five-year warranty coverage is to be expected on a premium hard drive like the VelociRaptor, and WD doesn’t disappoint. The longer warranty is particularly notable given the recent trend toward shorter coverage for mainstream drives. Despite their premium prices, only a handful of SSDs can match the VelociRaptor’s five-year coverage.

Our testing methods

We have a full suite of performance results for literally dozens of different drives, but today, we’ve narrowed the field to focus largely on how the VelociRaptor’s performance compares to mechanical alternatives. The only drive that’s really comparable to the new VelociRaptor is its predecessor, the VR200M, which we’ve tested alongside a selection of 7,200-RPM desktop models. We’ve also thrown in a handful of 2.5″ notebook drives, including two generations of Seagate’s Momentus XT hybrids, which combine mechanical platters with NAND caches.

Since the VelociRaptor 1TB is set to cost $320, we couldn’t resist adding a handful of SSDs in the same price range. We’ve included four 240-256GB models based on some of the most popular controller configurations on the market right now. Our testing methods and systems haven’t changed, so the VelociRaptor’s scores can be compared to those in any of our storage reviews dating back to last September.

If you’re familiar with our test methods and hardware, the rest of this page is filled with nerdy details you already know; feel free to skip ahead to the benchmark results. For the rest of you, we’ve summarized the essential characteristics of all the drives we’ve tested in the table below.

  Interface Cache Spindle speed Areal density Flash controller NAND
Corsair Force Series 3 240GB 6Gbps NA NA NA SandForce SF-2281 25-nm Micron async MLC
Corsair Force Series GT 240GB 6GBps NA NA NA SandForce SF-2281 25-nm Intel sync MLC
Crucial m4 256GB 6Gbps 256MB NA NA Marvell 88SS9174 25-nm Micron sync MLC
Samsung 830 Series 256GB 6GBps 256MB NA NA Samsung S4LJ204X01 2x-nm Samsung Toggle DDR
Hitachi Deskstar 7K3000 3TB 6Gbps 64MB 7,200 RPM 411 Gb/in² NA NA
Seagate Barracuda 3TB 6Gbps 64MB 7,200 RPM 625 Gb/in² NA NA
Seagate Momentus 5400.4 25GB 3Gbps 8MB 5,400 RPM 204 Gb/in² NA NA
Seagate Momentus XT 500GB 3Gbps 32MB 7,200 RPM 394 Gb/in² NA* 4GB SLC
Seagate Momentus XT 750GB 6Gbps 32MB 7,200 RPM 541 Gb/in² NA* 8GB SLC
WD Caviar Black 1TB 6Gbps 64MB 7,200 RPM 400 Gb/in² NA NA
WD Caviar Black 2TB 6Gbps 64MB 7,200 RPM 400 Gb/in² NA NA
WD Scorpio Black 750GB 3Gbps 16MB 7,200 RPM 520 Gb/in² NA NA
WD VelociRaptor VR200M 600GB 6Gbps 32MB 10,000 RPM NA NA NA
WD VelociRaptor 1TB 6Gbps 64MB 10,000 RPM NA NA NA

We used the following system configuration for testing:

Processor Intel Core i5-2500K 3.3GHz
Motherboard Asus P8P67 Deluxe
Bios revision 1850
Platform hub Intel P67 Express
Platform drivers INF update 9.2.0.1030

RST 10.6.0.1022

Memory size 8GB (2 DIMMs)
Memory type Corsair Vengeance DDR3 SDRAM at 1333MHz
Memory timings 9-9-9-24-1T
Audio Realtek ALC892 with 2.62 drivers
Graphics Asus EAH6670/DIS/1GD5 1GB with Catalyst 11.7 drivers
Hard drives Corsair Force 3 Series 240GB with 1.3.2 firmware

Corsair Force Series GT 240GB with 1.3.2 firmware

Crucial m4 256GB with 0009 firmware

Samsung 830 Series 256GB with CXM03B1Q firmware

Hitachi Deskstar 7K3000 3TB with MKA0A580 firmware

Seagate Barracuda 3TB with CC47 firmware

Seagate Momentus XT 500GB with SD22 firmware

Seagate Momentus XT 750GB with SM12 firmware

WD Caviar Black 1TB with 05.01D05 firmware

WD Caviar Black 2TB with 01.00101 firmware

WD Scorpio Black 750GB with 01.01A01 firmware

WD VelociRaptor VR200M 600GB with 04.05G04 firmware

WD VelociRaptor 1TB with 04.06A00 firmware

Power supply Corsair Professional Series Gold AX650W
OS Windows 7 Ultimate x64

Thanks to Asus for providing the systems’ motherboards and graphics cards, Intel for the CPUs, Corsair for the memory and PSUs, Thermaltake for the CPU coolers, and Western Digital for the Caviar Black 1TB system drives.

We used the following versions of our test applications:

Some further notes on our test methods:

  • To ensure consistent and repeatable results, the SSDs were secure-erased before almost every component of our test suite. Some of our tests then put the SSDs into a used state before the workload begins, which better exposes each drive’s long-term performance characteristics. In other tests, like DriveBench and FileBench, we induce a used state before testing. In all cases, the SSDs were in the same state before each test, ensuring an even playing field. The performance of mechanical hard drives is much more consistent between factory fresh and used states, so we skipped wiping the HDDs before each test—mechanical drives take forever to secure erase.

  • We run all our tests at least three times and report the median of the results. We’ve found IOMeter performance can fall off with SSDs after the first couple of runs, so we use five runs for solid-state drives and throw out the first two.

  • Steps have been taken to ensure that Sandy Bridge’s power-saving features don’t taint any of our results. All of the CPU’s low-power states have been disabled, effectively pegging the 2500K at 3.3GHz. Transitioning in and out of different power states can affect the performance of storage benchmarks, especially when dealing with short burst transfers.

The test systems’ Windows desktop was set at 1280×1024 in 32-bit color at a 75Hz screen refresh rate. 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 — Transfer rates

HD Tune lets us present transfer rates in a couple of different ways. Using the benchmark’s “full test” setting gives us a good look at performance across the entire drive rather than extrapolating based on a handful of sample points. The data created by the full test also gives us fodder for line graphs.

To make the graphs easier to interpret, we’ve greyed out the SSDs. They’ve also been left off our line graphs to make things easier to read. The other results have been colored by drive maker, with the VelociRaptor 1TB set apart from WD’s other offerings in a brighter shade of blue.

Score one for the VelociRaptor—sort of, and only just. The latest VelociRaptor just edges out the Barracuda 3TB to top the mechanical field in HD Tune’s read speed test. As the line graph illustrates, the VelociRaptor offers higher transfer rates than the ‘cuda over about the last quarter of their respective capacities. The old VelociRaptor must settle for third place among mechanical drives, 35MB/s behind the leader.

Well, the leader of the mechanical pack, anyway. The VelociRaptor 1TB is blown away by the SSDs, whose average sequential read speeds are 2.5 times faster.

HD Tune’s write speed test produces similar results for the mechanical drives. The VelociRaptor remains in the lead, trailed by the Barracuda 3TB.

Although the SSDs don’t enjoy as much of an advantage as they did in the read speed test, they still outclass the VelociRaptor. The slowest of the solid-state drives is 80MB/s ahead of the WD drive, and the Samsung 830 Series is 2.4 times faster.

HD Tune’s burst speed tests are meant to isolate a drive’s cache memory.

The VelociRaptor 1TB’s 64MB cache is very quick, allowing the drive to leap into fourth place overall. It’s not the only mechanical drive to post higher burst speeds than some of the SSDs, though. The Barracuda has a higher write burst speed, in fact.

HD Tune — Random access times

In addition to letting us test transfer rates, HD Tune can measure random access times. We’ve tested with four transfer sizes and presented all the results in a couple of line graphs that include the Samsung 830 Series SSD for reference. We’ve also busted out the 4KB and 1MB transfers sizes into bar graphs that include all the SSDs.

There’s no getting around the fact that the access times for SSDs are two orders of magnitude lower than they are for mechanical drives. Even the Momentus XT hybrids get in on the action up to the 64KB transfer size. 1MB random reads prove too large for their solid-state caches, though. The SSDs are down to one order of magnitude ahead in the 1MB test, as well.

Against its strictly mechanical peers, the VelociRaptor is unmatched. The 10k-RPM spindle speed pays big dividends here, putting the VelociRaptors in a different class than their 7,200-RPM competition. The new model is slightly quicker than the old one across the board.

Things get a little bit, well, weird in HD Tune’s 512-byte random write test. The SSDs continue to dominate, of course, and the read-only caches of the Momentus XT hybrids are of no help. A number of the Western Digital drives have higher access times than one might expect given their performance in the other tests, though. The VelociRaptor 1TB is one of them, but its predecessor is unaffected. Somehow, the Barracuda 3TB manages SSD-like access times, which makes me think it’s caching small writes in its DRAM cache.

In the tests with larger transfer sizes, the VelociRaptor 1TB continues to excel. It has much lower access times than the 7,200-RPM alternatives and even a big edge over the old VelociRaptor in the 1MB test. The solid-state drives remain out of reach.

TR FileBench — Real-world copy speeds

Concocted by resident developer Bruno “morphine” Ferreira, FileBench runs through a series of file copy operations using Windows 7’s xcopy command. Using xcopy produces nearly identical copy speeds to dragging and dropping files using the Windows GUI, so our results should be representative of typical real-world performance. We tested using the following five file sets—note the differences in average file sizes and their compressibility. We evaluated the compressibility of each file set by comparing its size before and after being run through 7-Zip’s “ultra” compression scheme.

  Number of files Average file size Total size Compressibility
Movie 6 701MB 4.1GB 0.5%
RAW 101 23.6MB 2.32GB 3.2%
MP3 549 6.48MB 3.47GB 0.5%
TR 26,767 64.6KB 1.7GB 53%
Mozilla 22,696 39.4KB 923MB 91%

The names of most of the file sets are self-explanatory. The Mozilla set is made up of all the files necessary to compile the browser, while the TR set includes years worth of the images, HTML files, and spreadsheets behind my reviews. Those two sets contain much greater numbers of smaller files than the other three. They’re also the most amenable to compression.

We’ve tested the SSDs in a used state, after running a 30-minute IOMeter workload. The IOMeter test file consumes the SSD’s entire capacity, and we delete it right before running FileBench. SSDs are typically slower in this state than they are fresh from a secure erase that empties all the available flash pages. We haven’t found a substantial difference in the performance of mechanical drives between these states.

The new VelociRaptor takes a bite out of the Crucial m4 in the Mozilla and TR tests, but it can’t catch the other SSDs. Those drives open up bigger leads in the other tests, which are made up of fewer, larger files. In the MP3, RAW, and movie tests, the VelociRaptor stays within striking distance of another SSD, this time the asynchronous SandForce setup that underpins Corsair’s Force Series 3.

 

WD’s latest VelociRaptor is much faster than the old 600GB model no matter the file set. It runs away from the mechanical drives in the TR and Mozilla tests, too. However, the VelociRaptor is neck-and-neck with the Barracuda when copying our mid-sized MP3 and RAW files.

TR DriveBench 1.0 — Disk-intensive multitasking

TR DriveBench allows us to record the individual IO requests associated with a Windows session and then play those results back as fast as possible on different drives. We’ve used this app to create a set of multitasking workloads that combine common desktop tasks with disk-intensive background operations like compiling code, copying files, downloading via BitTorrent, transcoding video, and scanning for viruses. The individual workloads are explained in more detail here.

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 with each multitasking workload.

Utter domination for the SSDs. Their I/O rates are in the neighborhood of four times higher than the best mechanical drive, which turns out to be the VelociRaptor 1TB—again. The new VelociRaptor has a comfortable lead over its elder sibling and a substantial advantage over all the other mechanical drives, including the hybrids. Indeed, the VelociRaptor 1TB nearly doubles the score of the Barracuda 3TB and Caviar Black 1TB.

Let’s break down DriveBench’s individual tests to see if we can find any interesting storylines.

The VelociRaptor 1TB’s trouncing of its mechanical competition is consistent. However, it pales in comparison to the performance of the SSDs. Looks like the Barracuda 3TB, which fared well in our real-world file copy tests, doesn’t like performing typical desktop tasks at the same time. It plunges to last place in the file copy test, pushing about a quarter of the IOps of the latest VelociRaptor.

TR DriveBench 2.0 — More disk-intensive multitasking

As much as we like DriveBench 1.0’s individual workloads, the traces cover only slices of disk activity. Because we fire the recorded I/Os at the disks as fast as possible, solid-state drives also have no downtime during which to engage background garbage collection or other optimization algorithms. DriveBench 2.0 addresses both of those issues with a much larger trace that spans two weeks of typical desktop activity peppered with multitasking loads similar to those in DriveBench 1.0. We’ve also adjusted our testing methods to give solid-state drives enough idle time to tidy up after themselves. More details on DriveBench 2.0 are available on this page of our last major SSD round-up.

Instead of looking at a raw IOps rate, we’re going to switch gears and explore service times—the amount of time it takes drives to complete an I/O request. We’ll start with an overall mean service time before slicing and dicing the results.

Once again, the new VelociRaptor finds itself caught between two worlds. On one hand, its mean service time is comfortably shorter than any of the other mechanical drives. On the other, the SSDs lead by fair margins. The mean service times of the fastest solid-state drives are less than one fourth those of the new VelociRaptor.

Let’s see what we can learn by splitting service times between read and write requests.

The SSDs clean up with reads, but their write performance isn’t as impressive. The Crucial m4 has a longer mean write service time than a number of the mechanical drives, including the VelociRaptor 1TB, which nearly catches the Corsair Force Series 3.

There are millions of I/O requests in this trace, so we can’t easily graph service times to look at the variance. However, our analysis tools do report the standard deviation, which can give us a sense of how much service times vary from the mean.

Well, there’s your problem, Crucial. The m4 has much more variance in its write service times than any other drive, including the slowest mechanical model. The other SSDs offer much more consistent write service times.

The old VelociRaptor actually has less variance in its write service times than the old model, but the difference is quite small. The new model makes it up with reads, although the service times of the mechanical drives are much more variable as a whole.

If I haven’t already scared you off with too many graphs and statistics, this next pair will do it. We’re going to close out our DriveBench analysis with a look at the distribution of service times. I’ve split the tally between I/O requests that complete in 0-1 milliseconds, 1-100 ms, and those that take longer than 100 ms to complete.

Thanks to its solid-state cache, the Momentus XT 750GB is able to hang with the VelociRaptor 1TB in the read component of this metric. The latest VelociRaptor has more sub-millisecond read service times than its mechanical peers. The SSDs are in another tier, though. They also have fewer extremely long service times over 100 milliseconds, but the percentages for that category are quite low overall.

The new VelociRaptor again proves potent when we consider write performance. It’s the only mechanical drive to break into the SSD pack, completing more write requests in under a millisecond than the Crucial m4.

IOMeter

Our IOMeter workloads feature a ramping number of concurrent I/O requests. Most desktop systems will only have a few requests in flight at any given time (87% of DriveBench 2.0 requests have a queue depth of four or less). We’ve extended our scaling up to 32 concurrent requests to reach the depth of the Native Command Queuing pipeline associated with the Serial ATA specification. Ramping up the number of requests also gives us a sense of how the drives might perform in more demanding enterprise environments.

We run our IOMeter tests using the fully randomized data pattern, which is a worst-case scenario for SSDs, not that it matters. They’re so much faster in these IOMeter tests that including the results would completely throw off the scale of the graphs, making the mechanical drives impossible to distinguish from one another. If you want to see what I’m talking about, check out how the SSDs stack up on this page of our Vertex 4 review.

Our IOMeter workloads benefit from quick access times, and the 10k-RPM drives outclass their 7,200-RPM rivals as a result. The VelociRaptors offer much higher transaction rates from the lightest to the heaviest load. The new model extends its lead over the VR200M as the number of concurrent I/O requests ramps up in the web server, database, and workstation tests. However, the reverse is true in the file server test, where the difference between the two is particularly prominent. I suspect that’s because the file server test is the only one to contain 512-byte writes, which we’ve already seen pose problems for the new VelociRaptor.

Boot duration

Before timing a couple of real-world applications, we first have to load the OS. We can measure how long that takes by checking the Windows 7 boot duration using the operating system’s performance-monitoring tools. This is actually the first time we’re booting Windows 7 off each drive; up until this point, our testing has been hosted by an OS housed on a separate system drive.

The Momentus XT 750GB has some caching mojo dedicated to accelerating the Windows 7 boot process, and the new VelociRaptor 1TB can’t keep up. It’s a few seconds shy of the SSDs, too, but loads the OS quicker than all of the purely mechanical drives.

Level load times

Modern games lack built-in timing tests to measure level loads, so we busted out a stopwatch with a couple of reasonably recent titles.

The SSDs have a few seconds on the VelociRaptor 1TB here, as well. The latest Momentus XT hybrid enjoys quicker Duke Nukem Forever load times than the WD drive, but its solid-state cache isn’t as effective at speeding entry into one of Portal 2‘s test chambers. As we’ve seen throughout our tests, the new VelociRaptor offers better performance than purely mechanical drives, including its 10k-RPM sibling.

Power consumption

We tested power consumption under load with IOMeter’s workstation access pattern chewing through 32 concurrent I/O requests. Idle power consumption was probed one minute after processing Windows 7’s idle tasks on an empty desktop.

The new VelociRaptor’s power consumption is very close to that of the old model, which is surely a requirement of its underlying enterprise ambitions. While they can’t match the low power draw of the solid-state or notebook drives, the VelociRaptors do pull fewer watts than the desktop models.

Noise levels

We’re a little OCD here at TR, so we’ve constructed a Box ‘o Silence to test the noise emitted by mechanical hard drives. This 18″ x 20″ anechoic chamber is lined with acoustic foam, and we suspend hard drives inside it, exactly 4″ away from the tip of our TES-52 digital sound level meter. You can read more about the setup here.

To ensure the lowest possible ambient noise levels, we swapped the test system’s graphics card for a passively-cooled Gigabyte model and unplugged one of the Frio’s dual fans. Noise levels were measured after one minute of idling at the Windows desktop and during an HD Tune seek test.

The VelociRaptor 1TB is a bit of a chatterbox when seeking; it’s much louder than any of the other drives, including the old VelociRaptor. I wasn’t annoyed by the drive’s seek noise when it was running on my open test rack, which sits about five feet away, but the buzzing was more noticeable than the duller whir of the other mechanical drives.

At idle, the new VelociRaptor is less obtrusive, matching the noise level of the Caviar Black 1TB. It’s quieter than the VR200M, too, albeit by fewer decibels than the difference in seek noise between the two VelociRaptors.

The value perspective

Welcome to our famous value analysis, which adds capacity and pricing to the performance data we’ve explored over the preceding pages. We used Newegg prices to even the playing field, and we didn’t take mail-in rebates into account when performing our calculations. Since the VelociRaptor 1TB isn’t widely available online just yet, we’ve had to use WD’s suggested retail price.

First, we’ll look at the all-important cost per gigabyte, which we’ve obtained using the amount of storage capacity accessible to users in Windows.

Although the SSDs approach the arbitrarily magical dollar-per-gigabyte mark, the mechanical drives are cheaper, at least per gigabyte. The VelociRaptor 1TB’s $320 asking price makes it more expensive per gig than all the mechanical drives except the VR200M. The Momentus XT hybrids also command a premium, while the Barracuda 3TB represents the best value for your storage dollar—based on capacity alone.

Our remaining value calculations use a single performance score that we’ve derived by comparing how each drive stacks up against a common baseline provided by the Momentus 5400.4, a 2.5″ notebook drive with a painfully slow 5,400-RPM spindle speed. This index uses a subset of our performance data described on this page of our last SSD round-up.

That pretty much says it all. The VelociRaptor 1TB is clearly the fastest mechanical drive, and it offers substantially higher performance than the old model. However, SSDs represent a huge step up overall.

Now for the real magic. We can plot this overall score on one axis and each drive’s cost per gigabyte on the other to create a scatter plot of performance per dollar per gigabyte. Because the SSD performance scores are so much higher, I’ve busted out two scatter plots: one with labels and just the mechanical drives, and another that maps the relationship between those drives and the solid-state alternatives. The latter gets a little too crowded for labels, so you’ll have to make do with the color-coding.

The VelociRaptor 1TB is in a much more attractive region of our first scatter plot than the VR200M, which is both slower and more expensive per gig. While the other mechanical drives sit to the left of the new VelociRaptor on the dollar-per-gigabyte axis, they’re much lower on the performance scale.

In our second plot, the solid-state drives stick to the top-right corner, with extremely high performance that costs quite a lot more per gigabyte. The mechanical drives occupy the lower left quadrant, offering less performance at a lower per-gig cost.

To look at the value perspective from a slightly different angle, we’ve divided our overall performance score by the sum of our test system’s components. Those parts total around $800 before we add the cost of the SSDs.

As the most expensive mechanical drive of the bunch, the VelociRaptor 1TB sits the farthest to the right on our first scatter plot. Its price premium at least comes with higher performance than the other mechanical drives.

The second plot is really interesting because the new VelociRaptor costs about as much as our 240-256GB SSDs. They offer about a quarter of the capacity, of course, but also several times the performance.

Conclusions

The VelociRaptor 1TB is easily the fastest mechanical hard drive we’ve tested. In addition to offering the highest sequential transfer rates, it delivers best-in-class access times. The VelociRaptor’s 10k-RPM spindle speed deserves much of the credit, and the move to higher-density platters allows the new model to improve upon the old one’s performance while also offering an extra 400GB of storage capacity.

As we saw throughout our test suite, the VelociRaptor’s performance falls between SSDs and slower-rotating mechanical drives. So does its cost per gigabyte. I’m just not sure how many desktop users are looking for storage in that middle ground. Most of them would probably be better served by the combination of a solid-state boot drive backed by secondary mechanical storage. Indeed, one could pair WD’s own Caviar Black 1TB with a 128GB SSD for less than the new VelociRaptor’s $320 asking price.

Capacity is the only advantage the VelociRaptor has over the growing field of solid-state drives. Modern SSDs are faster across the board. They’re pricier, of course, but only per gigabyte. Our favorite SSD, the Samsung 830 Series 256GB, has been selling for around $300, or $20 less than the new VelociRaptor’s suggested retail price. The VelociRaptor offers four times the storage capacity, though.

I’ve held off calling the VelociRaptor a dinosaur until now, but that’s the inescapable reality. The fact is the VelociRaptor hails from an era when 10k-RPM drives were the only way to get substantially quicker access times in a desktop PC. SSDs have redefined the market for high-performance desktop storage, relegating the VelociRaptor to a much smaller niche. It seems unlikely that enthusiasts will plunk down $320 for a terabyte of mechanical storage unless they’re building an ultra-high-end system already equipped with an SSD. The VelociRaptor seems more likely to appeal to workstation users and content-creation professionals working with data sets too large for affordable SSDs.

Let’s not forget the fact that WD is surely grooming a version of the VelociRaptor targeted squarely at servers. That drive will also face challenges from solid-state storage, but its capacity could prove more valuable for enterprise applications that deal with loads of data. And no one will notice the VelociRaptor’s seek chatter when it’s running inside a crowded server room.

Server versions of the VelociRaptor have always seemed to be Western Digital’s primary focus. The desktop flavors have felt like a sort of bonus for enthusiasts, an admittedly small niche in the larger PC market going back as far as the original Raptor. To remain relevant to that niche, the next generation will need to do something drastic—likely with solid-state storage. Whether the next ‘Raptor sheds its mechanical platters completely or become some sort of robot-dinosaur hybrid remains to be seen. I guess we’ll let you know in a couple of years.

Comments closed
    • LoneWolf15
    • 7 years ago

    I’m still using a VRaptor 600GB boot drive in my desktop system and very happy with it; I bought it on release around two years ago.

    SSDs that have the capacity I’d be interested in (512GB) for my desktop have unrealistic prices, especially ones that have the reliability I can count on. My desktop is on almost 24/7, so boot time isn’t that important.

    I have an SSD in my laptop (which also fits a regular HDD) and one in my wife’s (she doesn’t need high capacity), where they make sense for us, but not in the desktops. And for my desktop, the `Raptor has been a great boot drive, with no issues.

    • rjseo1
    • 7 years ago
    • rrr
    • 8 years ago

    Meh, it still doesn’t provide SSD like random read/write performance. And fat chance any mechanical disk would.

    • rjseo
    • 8 years ago
      • boing
      • 8 years ago

      I’ve seen this before, that time it was a claim of a single mother earning $9000 or some per week. What caught my interest that time was that the single mother supposedly lived within the same small town as I did at the time, so I tried finding her to see if the claim was actually real.

      Turned out, after checking census records, that no person with that name (or similar) actually exists! The single mother on the picture didn’t exist and probably not the money either.

    • ptsant
    • 8 years ago

    SSD are too small, too expensive. RAM is cheap and with 32 Gb, stuff that is not in the cache is already too large to be on an SSD (my photos and games wouldn’t fit on an SSD). The operating system and major applications are already in memory, unless just after a reboot, which happens once a week. I can find usage patterns for an SSD (a database, a boot drive) but I don’t fit into any of those…

      • Airmantharp
      • 8 years ago

      Given that SSDs are relatively ‘small’, you wouldn’t be putting data (photos, videos, documents) on them. You also wouldn’t be loading your entire Steam/Origin/whatever collection on one either. Good thing that many games don’t really benefit from SSDs enough to warrant being moved over.

      Thing is, unless you’re using a RAM drive (that must be loaded from non-volatile storage itself), you’re not going to get anything better than an SSD. Further, SSDs are definitely big enough for the OS, major applications, and what games you use often that also benefit from them to make using SSDs worthwhile. For me, just having BF3 and SWTOR on my SSD is more than enough; and when I use the machine for work, not having to stop for a spinning disk is also very gratifying.

    • Compton
    • 8 years ago

    I was once the very proud owner of the original SATA 1.5gbps 36GB 10K Raptor, which along with a rather sweet Intel mainboard (which even did RAID 0 and 1 with it’s two SATA ports!) made me quite happy for some time.

    But that might as well be a thousand years ago, not ten. The Raptor is pretty much extinct, and I can find no reason to get one when for the same money you could buy a 1TB HDD and an SSD. There is little point, and I’m not sure who is chomping at the bit for an expensive VR 1TB for anything other than the sake of nostalgia.

      • clone
      • 8 years ago

      I had 2 36gb Raptors in a RAID array for speed, they were never impressive but they did sound like SCSI drives.

      I had 2 74gb Raptors in a RAID array for speed and they did impress noticeably.

      I had 2 150gb Raptors last year scooped em up for $74 a piece which apparently was a temporary and good price at the time… anyway put them into an Array and while interesting I also had an SSD and was using them mainly for storage so while I was happy with my comp they didn’t shine as much as the SSD.

      broke the array and sold them in separate computers to family members, when I’m at their house I do notice the speed difference and kind of miss them.

      I just bought my 2nd SSD an Intel 520 series 120gb which will be going into my main rig while my older Corsair Force 120 is going into my slower rig an Intel C2D 6300.

      I don’t use a lot of space dump my data once I’m done with it…. currently using a 40gb Seagate sata drive while I wait for the new parts…. superfetch hides hdd’s shortcomings quite well once you get past post.

    • MargaretNKinnley
    • 8 years ago
    • merryjohn1
    • 8 years ago
      • ImSpartacus
      • 8 years ago

      Your best friend’s sister too? No way!

    • merryjohn1
    • 8 years ago
    • Chrispy_
    • 8 years ago

    Dinosaur indeed.

    I want an affordable RAID controller to provide simultaneous mixed raiding. Intel’s ICH10R’s don’t let you take two drives and use half of each disk for RAID1 and half of each disk for RAID0.

    A typical PC I build these days has a 128GB SSD and two 2TB drives for storage. 128GB is quite limiting, but the 2TB is slow in RAID1.

    If I could use just two disks to have a 512MB RAID0 volume for games and things than need good sequential performance, AND a 1.75GB RAID1, That would be great. At the moment you’d need to buy an extra RAID card to do that, and you might as well just get two more disks instead.

      • Waco
      • 8 years ago

      Umm…last I checked you could do precisely what you’re thinking of using the Intel controller. It’s called mixed RAID mode if I remember correctly.

    • swiffer
    • 8 years ago

    [quote<]... or become some sort of robot-dinosaur hybrid ...[/quote<] I'm pretty sure there's hilarious video card box art somewhere with precisely this motif.

    • Coyote_ar
    • 8 years ago

    At this time, the only thing i use my mechanical drive for, its for storing media.

    I really see no reason to have a fast mechanical drive, the only thing i need from my drive is high sequential reads/writes. And you can have that with high density plates.

    I would like to see some uber sized mechanical drives … i would like having a 10 or more Tb, 5.25″ HDD, Quantum Bigfoot style. As long as its able to keep up with decent sequential speeds, i wouldnt mind slow seek time.

    For speed … i have an SSD, and i really dont need more than 200gb for that.

      • willmore
      • 8 years ago

      Agreed, bring back the bigfoot!

      • axeman
      • 8 years ago

      I wouldn’t, 2-3TB on one drive is enough data to lose at once.

        • Malphas
        • 8 years ago

        That’s what RAID and backups are for. It’s always safer to use the same amount of larger capacity drives than smaller ones, so I have no idea why people perpetuate this bizarre argument.

        e.g. You could have 4 x 250GB drives so you only lose 250GB of data if one of them dies, or you could have 4 x 1TB drives in two RAID 1 arrays, using one of the arrays as periodic backup, and be able to withstand up to 3 of the drives dying before you risk losing any data.

          • axeman
          • 7 years ago

          * It’s always safer to use the same amount of larger capacity drives than smaller ones.

          uh, what? I though the implied argument is it is better to use more drives to achieve the same usable space. Sorry, I’d rather have 4x1tb drives than 2x2tb drives. Arguing that bigger drives are better than smaller ones because you can have more copies of the data is pretty obvious. I’m pretty sure no sane person thinks spreading data across multiple smaller drives is a substitute for RAID or backups.

          *4 x 1TB drives in two RAID 1 arrays, using one of the arrays as periodic backup,

          you really think that most people who buy 4 x 1TB drives are going to use a total usable capacity of 1TB? In the enterprise, sure, but consumers? Most people are going to run a 2 drive RAID 1, maximum. It is *extremely* common for a drive to fail during a rebuild of a failed RAID, or before a spare is sourced (especially since in the consumer space, drives usually will be the same model, and even from the same production run), so if I’m going to lose the second drive while trying to rebuild the array, I’d rather lose 1TB than 3TB at once. Even if it’s backed up, it’s a smaller pain to restore a smaller dataset. I’m also sure that many people, like me, have large amounts of data that aren’t valuable enough to justify the cost of several extra TB of backup space. Not to mention, it seems the larger the drives are getting, the less reliable they are getting.

      • indeego
      • 8 years ago

      The thing that really bugs me about slow mechanicals (versus fast mechanicals) are backup operations. They just take way too long, especially for large 1 TB+ drives and that first backup. They absolutely lock out realistic persistent use of the drive during the backup, as well.

        • Coyote_ar
        • 8 years ago

        well, it depends on the type of backup. you could always use a raid 1 and have a backup on the fly.

        plus on a bigfoot style drive, you have a slow seek time, not necesarily a slow transfer rate. as long as your fragmentation is low, this type of drives should be pretty good at sustained read/writes.

          • indeego
          • 8 years ago

          [quote<]you could always use a raid 1 and have a backup on the fly.[/quote<] No, you couldn't.

            • Coyote_ar
            • 8 years ago

            why not? it is a backup against hardware failure. it wont be a backup against human errors, but if you use this kind of disks for media storage, you should be fine with a raid1 + a decently sized recycle bin.

            • Firestarter
            • 8 years ago

            Backup is a misnomer, RAID provides redunancy, as in the R of RAID.

            RAID doesn’t protect you against a virus or file system failure like a proper backup would. RAID doesn’t protect your data against your house burning down, or your computer frying itself and its components. If you value your data, you back it up.

            • Kurotetsu
            • 8 years ago

            Well, in his defense, he specifically said “backup against hardware failure”, which I took to mean the hard drive itself taking a crap. Though I suppose your house burning down is a kind of hardware failure….

            • travbrad
            • 8 years ago

            Even in the case of your house burning down your “real” backup needs to be off site for that to be any better than RAID. A lot of people don’t do off-site backups for their home PC (if they make backups at all)

        • Bensam123
        • 7 years ago

        Incremental backup in the background with a low priority sounds like a winner.

          • indeego
          • 7 years ago

          Except low priority for mechanicals still has a large impact on drive performance.

    • BIF
    • 8 years ago

    I loved my old 74 GB Raptor, especially for virtual instruments.

    But I didn’t realize how loud it was until much later when I water cooled the CPU and removed a bunch of noisy fans. Now the Raptor sounds like ice in a blender…awww, what a disappointment!

    It’s nice to see that WD is … somewhat keeping up.

    Unfortunately for them, I no longer need 10,000 RPM drives these days because Windows 7 64 bit allows so much storage. After the initial load of my project containing VST instruments, a whole orchestra can fit in my 8 GB of RAM now. Well, okay, maybe only a string quartet will fit; depends on the samples… 😉

    I’m not sure I could even justify a couple or five of these raided together in a NAS, unless it was a mission-critical database involved in life or death decisions….

      • willmore
      • 8 years ago

      “ice in a blender” I like that! I used to say they sounded like hot air popcorn poppers.

    • Bensam123
    • 8 years ago

    Curious, why are SSDs absent from the overall performance $/GB? This would be one of the most important graphs when comparing and contrasting cheap storage vs really fast storage as it gives the widest picture. It still needs a best fit line too.

    This most definitely is a dinosaur, it’s a product without a purpose. A monolith of pre-SSD days, when WD (in a very clever manner) devised a device for PC enthusiasts that wanted faster storage and something that could make other geeks gawk at.

    I’m surprised SAS or 15k drives aren’t even mentioned in the article. Matter of a fact this article seems to have a bias (which is unusual for TR) that makes it seem like they don’t even exist and 10k is top of the game for mechanical hard disks.

    They are the most direct competitor to this line of drives. The fact that these drives don’t even support SAS shows how little WD knows about the market they’re targeting. PC enthusiasts aren’t going to buy these drives and chances are if you own a workstation you have on board SAS or own a SAS hardware raid card, they aren’t all that much more expensive them SATA raid cards, especially if you buy them off eBay. You can get a Perc 5/6 for around $100-150. The performance is a lot better then chipset raid too.

    SAS isn’t voodoo magic like SCSI. SAS controllers in and of themselves are backwards compatible with SATA. Meaning you can plug a SATA drive into a SAS controller. Unfortunately you can’t plug a SAS drive into a SATA controller though.

    That said, I don’t understand why WD dropped down to a 2.5″ form factor. It was stated in the article that 2.5″ allows for higher storage densities, but I don’t see how that’s true as a 3.5″ drive has more storage on it to begin with. You’re just making the device smaller to fit into different devices, but since this drive is too large to fit in laptops or other storage devices it makes it a rather lackluster argument. Most server enclosures are designed almost exclusively for 3.5″ drives too.

    Going a bit further, why is WD still sticking with a 10k spindle speed? There have been 15k drives out for years and while this had a particular niche for PC enthusiasts, where they could sorta half ass it as they were the only game in town, they can’t do that anymore.

    Honestly, if I were to recommend mechanical storage to someone who wants a fast disk subsystem, but doesn’t want to go the SSD route, I would recommend plucking a couple last generation 15k Cheetahs off eBay and a Perc 6i. You can get the last generation drives for practically nothing. I bought 3x146Gb Cheetah 15k.6s for ~$80 a pop brand new with a 5 year warranty and a Perc 6i for ~$150 three years ago. In my raid 0 array they still give SSDs a run for their money in sequential tests and they’re cheaper too, especially when I first bought them. SAS is a whole other animal when it comes to competing with SSDs, it’s a shame TR doesn’t add that perspective.

    I still don’t trust SSDs for my desktop and probably wont think about adopting them for another three years or so.

      • willmore
      • 8 years ago

      I agree with your post. I just want to answer the 2.5″ form factor question. I’d suggest that it’s because it alters the spindle density of storage solutions. And if seeks are an issue or RAID device granulatiry, that can matter. It’s no game changing advantage, but more, smaller drives vs fewer, larger drives is a meaningful tradeoff for different storage workloads.

        • Bensam123
        • 8 years ago

        3.5″ drives can have the same area density, I don’t know why that’d be different. They’re easier to make because the platters are thicker too, so there is higher tolerance (as far as warping goes).

          • willmore
          • 8 years ago

          No, I think you’re misunderstanding ‘spindle density’. That’s the number of separate spindles you can get per unit of volume. A 2.5″ drive is a spindle. A 3.5″ drive is a spindle. You can pack more 2.5″ drives into a given volume than you can 3.5″ drives. So, 2.5″ drives have a higher ‘spindle density’.

          I’m not talking about ‘storage density’ nor storage per spindle.

            • Bensam123
            • 8 years ago

            Ah, you confused me with how you worded that. So essentially you’re saying smaller drives can be useful because you can make bigger raid arrays (drive wise) as there are more drives for the same amount of space?

            I could see that, but like you said, that’s not really a game changer, especially for the pros and cons of using a mechanical over a SSD.

      • StashTheVampede
      • 8 years ago

      SAS is good, but it’s definitely not for everyone (Intel’s PCIe SSD uses SAS, fyi).

      15k drives are definitely fast, but the additional noise, heat and controller aren’t what most consumers are looking for. Once you’re splashing 200+ for a true SAS card and drives, that’s right along SSD territory for less power/faster random reads/writes, lower noise and probably a higher throughput than your current 3x 15.6x cheetah’s in RAID0.

      Most consumers don’t give a crap what the storage is. They want to buy a laptop at X price and don’t really care too much if there is a choice between SSD and spinning platters. At some point, spinning platters will be niche because for the same dollar, more sellers would prefer to put SSD into their builds instead.

      SSDs will have some heavy changes in the next 12-24 months. NVM_Express will be a big deal and it will eliminate all of the SATA bottleneck that these drives don’t need.

        • Krogoth
        • 8 years ago

        NVM_Express is just a specialized form of PCIe and it will be enterprise-only.

          • StashTheVampede
          • 8 years ago

          NVM_Express has both enterprise and consumer targets for adoption. NVM_Express also isn’t designed only for PCIe cards, it’s a spec that will provide non-volatile ram as a “storage” device to the OS over a PCIe bus.

          This is the true successor for SSD storage, it’ll work in laptops, desktops and enterprise servers. All current solutions are a hack into either SAS, SATA and they all have limitations that could be avoided.

        • Bensam123
        • 8 years ago

        I think people who say SAS drives are noisy haven’t been around them in quite awhile. I have three sitting in my computer right now and they aren’t any louder then a 7.2k drive (on seek they are, but they’re pretty much twice as fast at that). Older SAS, particularly back before SAS – SCSI drives, were ridiculously noisy. Perc 5/6i cards are true SAS cards… I don’t really know what defines a ‘true’ card though. :l

        Bottom paragraph details how much I spent. That’s for 438GB three years ago too.

        I wouldn’t recommend SAS drives for consumers though, nor would I recommend a Raptor after SSDs came out… which leaves Raptors in corporate territory. Only reason I pointed out the snippet about my SAS setup was because it was remarkably cheap for the performance $/GB if someone wants to go out of their way to find an alternative to SSDs. Even including SSDs it’s probably up there, but TR isn’t testing such a setup nor has results from it.

        I agree that SSDs are going to be bumping against the cap for SATA3, but I’m pretty sure 400MB/s isn’t it.

      • Dissonance
      • 8 years ago

      [i<]Curious, why are SSDs absent from the overall performance $/GB?[/i<] They're not. From the value section: "Because the SSD performance scores are so much higher, I've busted out two scatter plots: one with labels and just the mechanical drives, and another that maps the relationship between those drives and the solid-state alternatives. The latter gets a little too crowded for labels, so you'll have to make do with the color-coding."

        • Bensam123
        • 8 years ago

        Weird, I don’t know how I missed that… I think one of the graphs didn’t finish loading when I was on Wifi this afternoon. My bad.

      • Vaughn
      • 8 years ago

      Just wanted to quote this point Bensam123.

      “I’m surprised SAS or 15k drives aren’t even mentioned in the article.”

      I think there reason there is no comparison is SAS and 15k drives aren’t consumer products they are considered enterprise. This drive is a consumer drive and not in the same league.

      As mentioned in the article WD will release a SAS version which would then make a valid comparison.

        • Bensam123
        • 8 years ago

        Lines like this have been tripped over before. I don’t think product segmentation or marketing should be a deciding factor of what you think you can and can’t use in your computer. There is nothing stopping you from using SAS drives in your desktop or SATA drives in your SAN cluster… matter of a fact I’m sure there are a lot of companies that do that latter.

        Do you believe that tacking on a SAS interface to this drive will change it measurably?

      • Krogoth
      • 8 years ago

      SAS is just SATA engineered for enterprise needs (more devices on the controller and longer cable lengths).

      SAS doesn’t make sense unless your system is driving dozens of HDDs.

        • Bensam123
        • 8 years ago

        They don’t sell 15k SATA drives and you can usually find SAS controllers on eBay for cheap, because people don’t realize they work with SATA drives.

        I wasn’t implying people should run out and buy a SAS controller just cause they can. Rather the drives that are attached to SAS controllers are in a completely different league… territory this drive starts stepping into based on cost and where WD implies them being used.

          • Krogoth
          • 8 years ago

          Right.

          However, the SAS controller reverts the port that SATA device is tied to into SATA mode along with its restrictions (2M cable length, 1-4 devices per port etc.)

            • Bensam123
            • 8 years ago

            Yeah… That doesn’t change the SAS controller, which is almost always tailored for enterprise needs as you said so yourself. That entails higher levels of performance, it doesn’t matter if the cable itself doesn’t yield anything tangible in that realm. Both SATA and SAS drives connecting to it will benefit from it.

    • sircharles32
    • 8 years ago

    The 500 GB model looks quite compelling. It will do well with transferring multiple files to multiple systems, at the same time (compared to my current WD4000KD). It could very well find a home in my upcoming workstation rebuild, as a temporary holding drive. $210 asking price is a tad steep though.

    • tootercomputer
    • 8 years ago

    I bought my first 74g raptor for a new build in 2004. So much faster than 72rpm drives (which blew me away when I bought one to replace a 5400rpm drive). I ended up buying two more 74g raptors, then the 300g velociraptor, all used in different systems, and every single one of them still works perfectly. These are drives that do not die, and they are of the highest quality build.

    I’ve started using SSDs now, my first was a BSOD nightmare, the other two work great. So to a certain degree, the raptor is a bit of a dinosaur relative to the SSDs. But hats off to WD for making the raptors available. They significantly improved the performance of systems for many years.

    • jdaven
    • 8 years ago

    What would be cool is to run the Gigabyte GC-RAMDISK PCI card along with all the other hard/SSD drives in the test suite. This would show how close SSD drives are getting to the speed of DDR RAM and provide an upper limit to how far non-volatile storage limit has to go to get to volatile RAM speeds.

      • willmore
      • 8 years ago

      You’d have to do some work in windows to get it to stop caching. Considering the largest these devices could be configured was 4GiB. Plus, they only had SATA-I or SATA-II connectors, that’s going to give modern SSDs with SATA-III interconnects will have a bit of an advantage.

      • Bensam123
      • 8 years ago

      Yeah… Gigabytes ramdisk was a cool idea, but they never took it far enough… No one really has taken that concept any further. Sorta disappointing. They were never polished enough to be competitive.

        • Waco
        • 8 years ago

        It’s cheaper and easier these days to simply outfit your machine with a ton of RAM and have a RAMdisk that gets flushed to disk when you shut down.

          • willmore
          • 8 years ago

          Does windows have anything like tempfs? (which is just what you descirbe, but a completely automated.

    • Ushio01
    • 8 years ago

    After reading the Anandtech review I was really tempted but they never mention how loud it is. So I come to good old reliable Techreport and find out it’s a whooping 51 decibles when seeking so it’s off my buy list instantly.

    Changed my mind again just out of curiosity I looked up my current HDD techreport review.

    [url<]https://techreport.com/articles.x/15363/13[/url<] 52 decibels when idle 55 at seek. Maybe the new raptor isn't so bad afterall.

      • swaaye
      • 8 years ago

      Velociraptor supports AAM so you can make them surprisingly quiet and the performance change isn’t very noticeable. I used to run a VR150 in the quietest AAM setting.

    • axeman
    • 8 years ago

    Maybe because I’m an old fart I want one. Not going to buy it, of course, but I want it. Probably ’cause I always wanted a system with a bunch of 10K rpm drives clacking away. Maybe I should just get some old Cheetah drives and build it, just to have one of the things I wanted when I was a youngster.

      • willmore
      • 8 years ago

      I’ve got a pair in the basement.

      (oh, man, I just realized that’s walking into a “that’s what she said”)

      They’re not as exciting in real life as they were in my desires.

      • Bensam123
      • 8 years ago

      I don’t know what model Willmore has, but I have a 3x146Gb raid 0 array of the 15k.6 models and they’re faster sequentially then the fastest SSDs on the market (not PCI-E). Random seeks aren’t even close, but they’re still pretty close to twice as fast as a standard mechanical.

        • willmore
        • 8 years ago

        Some crusty old made for Sun drives that Seagate won’t cop to having made. They look like a half capacity version of the 34G 10K drives.

        They’re SCSI-U2W. Each drive can do 55M/s on the outside and the two of them together can saturage the bus even at the end of the drives. I’m sure they were darn fast a decade or more ago when they were new. They’re still not too bad–if very loud.

        I agree that a bargin big array of fast old enterprise class drives can keep an SSD against the ropes for sequential performance. Get enough spindles and enough load and they can beat on on seeks–but at that point, they’re probably losing on price though winning on capacity. Also, they’d take up a lot more space, power, etc.

        [Edited for spelling, derp.]

      • Chrispy_
      • 8 years ago

      I ran a couple of 15K cheetahs in RAID0 on an Adaptec controller a few years back. It was underwhelming against even the 1st gen, slightly-flaky SSD’s 🙁

      More annoying though wasn’t the clacking noise, it was the incredibly subtle, but ever-present 15KRPM whine.

        • willmore
        • 8 years ago

        That’s one of the reasons they’re enterprise drives–they’re meant to be used in noisy machine rooms where the people are either deaf or wearing ear protection. 🙂

    • yokem55
    • 8 years ago

    These things will do well in the low end vps server market. They need the high rpm’s to keep I/O performance up and the cost per gigabyte down so they can hit low monthly charges for their customers ….

      • thanatos355
      • 8 years ago

      I don’t know who thumb-downed you, but I upped to negate its effects, since this was almost my exact thinking while reading this article.

        • axeman
        • 8 years ago

        I can see this as well, often what limits server performance in virtual machine loads is I/O performance, and the cost of SSDs per GB is too high, and 7200rpm drives really struggle in that environment. Go 15krpm SAS drives even, and the total price of the server is going WAY up.

    • holophrastic
    • 8 years ago

    Surprisingly, I think there’s a missing comparison. The thing is, you can’t currently build most systems on SSDs alone. Given a typical desktop usage scenario, you need at least 1TB of data storage, usually 2TB. Doing that on SSDs alone is a big fat waste of money. At the same time, much like you won’t get full SATA speed from a mechanic drive, you don’t use the full speed scenarios of SSD drives all the time either.

    So what’s missing is the comparison of actual usage. Given a typical workstation or a typical gaming rig. How often is the increased storage space used vs the short burst speeds?

    And when I consider building a machine, with one SSD and one HDD at a minimum, which of each is the question. I may choose to boot and swap from an SSD, but my long-term data storage just can’t ever fit into one.

      • thanatos355
      • 8 years ago

      Preach, brother!

      I have six mechanical hard drives for a total of five terabytes of storage and I am STILL constantly on the cusp of exhausting my available space.

      Adding in a (comparatively) puny 256GB SSD for an (comparatively) insane amount of money for twice as fast load times in DN?

      I really would love to see some real world data on these things. I, for one, find the performance degradation cause by a “used state” extremely interesting. It makes me very curious as to how these drives would actually perform in an in-use desktop. Not just some rich boy’s toy.

        • Ushio01
        • 8 years ago

        I’ve been using a 64GB SSD since july 08 and it’s been problem free, well except when ever I have to use a PC with a HDD then it’s like pulling teeth even my older rig with the 1st gen VelociRaptor.

      • swaaye
      • 8 years ago

      The typical desktop scenario I run into is drives with 20-30GB of space consumed. If the individual is an iTunes user, then add 20GB or so of storage. Most people only use a web browser and some office apps.

      Of course, people here are a little different than the norm. 😉

      ~120GB SSDs can’t become cheap (and reliable) fast enough.

    • NeelyCam
    • 8 years ago

    I’m happy to finally see ‘System performance/dollar’ plots; they are more meaningful than looking at component prices independently.

    I hope to see this in CPU reviews as well – for a long time I’ve argued that it’s worth paying 50% extra for a CPU for a 30% increase in performance because it’s only 15% extra in the system price.

      • odizzido
      • 8 years ago

      They are both useful. Most of the time I purchase HDs independently so isolating the component is better for me.

    • Scrotos
    • 8 years ago

    Why are the areal densities for the VelociRaptors listed as NA? They still got platters, don’t they?

      • Scrotos
      • 8 years ago

      And from page 2, Samsung 830 Series 256GB has 256A of cache? MB?

        • bcronce
        • 8 years ago

        I found it interesting that the burst rate of the 830 was lower than the sustained rate by quite a bit. This means it is not fully isolating the cache.

      • Dissonance
      • 8 years ago

      On the very same page as that table…

      “We’ve never been able to coax Western Digital into revealing the exact areal density of the platters in its VelociRaptor drives. However, the company did confirm that the new model uses a trio of 333GB platters.”

        • Goty
        • 8 years ago

        Hooray for reading the article!

    • jwilliams
    • 8 years ago

    Platters usually use both sides, but they can use one side and save one read/write head. Normally, one would expect all capacities to be a multiple of the capacity of one platter side (which would be half of the platter capacity), which would be 166.6GB if these drives have 333GB platters.

    So, the 500GB model would use 3 sides (one and a half platters). Possibly it could use 2 full platters (4 sides) but not the full surface of the platters — but that would make it faster than the 1TB model, which seems an unlikely design choice.

    But a 250GB models sounds odd. It could be a mistake — maybe it will really be a 333GB model? But if there actually is a 250GB model, most likely it is using single platters that did not pass muster over the entire surface, so they have regions of the platter that are marked as “do not use” in the firmware. That might be testable in HD Tune — I would expect to see discontinuous “jumps” in the speed trace. Unless they only use platters with defects at the beginning or end. But that should also be noticeable in HD Tune, with the “fast” or “slow” portions of the trace looking shorter.

      • willmore
      • 8 years ago

      I don’t know what they’ve done with their plater usage on the reduced capacity models, so I don’t comment on that.

      WRT your comments on it being testable by looking at whole drive transfer rates, You may be able to detect if they are using a reduced # of surfaces vs using less of the same number of platters.

      The former would look just like the existing curve (maybe a few MB/s lower due to slightly higher # of seeks–this factor will depends a lot on the firmware and how their secondary actuators work.

      The latter would look like the first part of the curve of the full size drive.

    • FranzVonPapen
    • 8 years ago

    Consumer-grade SSDs continue to be a source of huge uncertainty. Is your SSD firmware going to BSOD your PC? Eat your data? Will your 25nm flash wear out in 18 months?

    I have a lot more faith in mechanical hard drives, particularly the Raptor series. The fact that I get a lot of fast storage for less $$$ is a bonus. I don’t mind it being a little slower than a SSD.

      • Vaughn
      • 8 years ago

      My Intel G2 SSD that has been rock solid since dec 2009 disagree’s with your statement.

      “I have a lot more faith in mechanical hard drives, particularly the Raptor series. The fact that I get a lot of fast storage for less $$$ is a bonus. I don’t mind it being a little slower than a SSD.”

        • axeman
        • 8 years ago

        Almost 2 and half years! Based on that one instance, you have proven OP’s opinion completely unfounded.

          • FranzVonPapen
          • 8 years ago

          Behold the power of anecdotes!

            • NeelyCam
            • 8 years ago

            Out of the five I’ve had, only one has died, and it wasn’t a flash wearout or BSOD – it was the power input melting for who-knows-why. A larger portion of my HDDs have failed.

            You’re letting your opinion be influenced by OCZ a bit too much, methinks

            EDIT: mine are1-2.5years old.

            • FranzVonPapen
            • 8 years ago

            Fine. Let’s anecdote. Out of the 50-60 hard drives used in PCs I’ve personally owned in the last 20 years, exactly one has died. All drives were used for at least 2 years, with many operating 3-4 years before being set aside during an upgrade.

            From my perspective, mechanical hard drives are doing [i<]just fine[/i<].

            • jensend
            • 8 years ago

            Yeah, mechanical drives are fine if you upgrade them every two years. But if you extend the timetable the picture becomes pretty bad pretty fast. Quoting [url=http://www.pdl.cmu.edu/PDL-FTP/Failure/CMU-PDL-06-111.pdf<]one of the better papers[/url<]:[quote<]• For less than 5 year old drives, field failure rates are by a factor of 2–12 larger than what the datasheet MTTF suggests. For 5-8 year old drives, field failure rates can be as much as a factor of 30 higher than what the datasheet MTTF suggests. • Changes in failure rates during the first 5 years of the lifecycle are more dramatic than often assumed. While failure rates are often expected to be in steady state in year 2-5 of operation (bottom of the “bathtub curve”), we observe a continuous increase in failure rates, starting as early as in the second year of operation.[/quote<]For SSDs, on the other hand, failure rates seem to be more constant, so though there are more early failures you may well be able to use a SSD for 4x as long as a hard drive.

            • flip-mode
            • 8 years ago

            So you have an HDD research but no SSD research, so we are no closer to understanding SSD failure rates.

            My anecdote: I have had one single hard drive fail at my company in the last 6 years and it failed at 8 years of age. That’s out of a sample of some 50 hard drives. We had one computer we shut down after 14 years of continuous operation – running the original hard drive the whole time.

            • NeelyCam
            • 8 years ago

            If that also means you haven’t had any HDDs DOA, then you’re one lucky ****.

            BTW, how many SSDs have you personally owned? How many have failed?

            • BobbinThreadbare
            • 8 years ago

            I feel like the best time for a HD to die is on arrival.

            • axeman
            • 8 years ago

            What brand of SSDs are they? Just curious, I really want to know more about good experiences, I’m tired of the bad ones, OCZ.

            • DancinJack
            • 8 years ago

            I haven’t had any issues with the two OCZ drives I have used. Obviously, others have had tons of issues. I’ve also not had a single issue with any of the Intel drives I’ve used. I think those consisted of 5-6 G2 drives and a couple 320 series drives. The two Crucial drives I used were also great.

            I guess I’m one of the lucky ones. I don’t think I’ve ever had an issue with a single SSD that I’ve either had in my machine, or installed in a friends/family system.

            • NeelyCam
            • 8 years ago

            I think it’s possible that those whose stuff fail are very vocal about it, while those without problems don’t necessarily feel compelled to make that known.

            • axeman
            • 8 years ago

            It’s human nature, we’re better at complaining than being happy.

            • travbrad
            • 8 years ago

            That’s true, but OCZ drives DID have some serious problems (especially the Vertex 2 line), otherwise there would have been a similar number of complaints about other SSDs. There’s a reason the Vertex 2 drives had 30%+ 1-egg ratings on newegg, while the Crucial m4 was around 5%. Even the Vertex 3 drives seem to be a lot less reliable (around 20% 1-egg ratings) than Samsung, Intel, Crucial, etc though.

            I’m sure a lot of people have OCZ drives that work perfectly fine (as you said people are more likely to post negative reviews than positive), but I’d rather save myself the potential headaches and get a drive with consistently good reliability reports.

            • NeelyCam
            • 8 years ago

            The one that died was Intel 80GB X18-G1 (got replaced with an 80GB Intel 320). I also have an 80GB X25-G2, 160GB 320 and 64GB OCZ Onyx. I also have an 80GB X18-G2 inside a USB3.0 enclosure as a tiny-yet-speedy external drive, but that doesn’t always get recognized by Win7 (might be the USB3-SATA bridge chip inside the enclosure..)

            The PC with the Onyx sometimes fails to boot, but it could be unrelated to the SSD – poweroff/poweron fixes it.

          • TO11MTM
          • 8 years ago

          [url=http://2.bp.blogspot.com/_vI6LGR352yE/TIRAbSmWwfI/AAAAAAAAACo/8iIV6gUDNRE/s1600/keep+trolling.jpeg<] Fred Durst Would Like a Word With You (NSFW)[/url<] Seriously though. Anecdotal evidence aside SSDs keep getting more reliable and affordable, especially as more reputable manufacturers are making the smaller sizes. A large number of SSD Failiures are attributed to much older models that typically had fairly decent reasons for their issues. The cheapie SSDs in netbooks were often very lowest bidder products and rather shite as a result. The early OCZ drives, all I can say is it's a new technology, Lots of people who have jumped on the early-tech bandwagon (I'm looking at you, Phenom 1, and Pre B3 Stepping Intel 6-series Chipsets) Have run into design issues. I wouldn't buy a modern Seagate Desktop drive because they seem to have design issues. Does that mean I have written off hard drives as a whole and gone trolling every HDD News/Review post? I don't think I have, anyway...

          • Vaughn
          • 8 years ago

          I’ll be happy to let you know how the drive is doing if I still have it for another 5 years.

          And i’m pretty sure it will still be running without issue.

          “Your sample size of 1 doesnt show much at all, and december 2009 is less than 2.5 years, i sure hope any storage device lasts longer than that.”

          My sample size is 2 since I own another one. If he wants to stay with HDD for the next 10 years all the power to him. I don’t have to prove anything to him that’s his business if he wants to live in the past 😛

        • Farting Bob
        • 8 years ago

        Your sample size of 1 doesnt show much at all, and december 2009 is less than 2.5 years, i sure hope any storage device lasts longer than that.

        • Malphas
        • 8 years ago

        Maybe so but anecdotal experience doesn’t override actual facts. SSD’s are still an uncertainty at present, due to firmware issues and the fact they simply don’t have the same track record as mechanical hard drives yet.

        Look at the BSOD bug the Crucial M4 had, that essentially makes [i<]every[/i<] M4 a timebomb to failure unless you patch the new firmware. I was one of those that hit the 5200 hour mark before the problem was diagnosed and thus had to wait two weeks before I could flash the drive to became usable again. I was fine with because I'm somewhat of an early adopter and was prepared for that eventuality, but to most people that's unacceptable. Just because [i<]your[/i<] drive has been fine, doesn't mean SSDs are more or as reliable as mechanical hard drives on the whole. In fact the more I think about it, the more it occurs to me what a ludicrous thing that is to say.

      • Goty
      • 8 years ago

      25nm flash should last nearly ten years in most desktop usage cases. That’s more than can be said for products from a number of HDD manufacturers.

        • FranzVonPapen
        • 8 years ago

        By the numbers, it should. But I’d be more curious to look back in just [i<]five[/i<] years and see how well today's SSDs have fared under real circumstances for millions of users.

      • axeman
      • 8 years ago

      Mechanical drives still fail quite often, and it seems more often in recent years. So no matter what, you should have good backups. 🙂 But I agree with you, even with backups, outside of maybe Intel SSDs, the failure rate is too damn high. It’s anecdotal, but I worked with an organization with a sample size of around 10,000 desktops and laptops, and I can say the statistics do not seem to favor SSDs. Plenty of desktops and laptops have drives churning away after the better part of a decade. In theory, SSDs should last this long based on rated life, but it seems due to quality control issues this seems yet to be realized. Perhaps Intel SSDs are really *that good*, I guess time will tell.

      Honestly, what we need is for some big players to emerge who start to own the SSD marketshare due to performance and reliability, and push the wannabes into irrelevance. The hard disk market was really like this once upon a time, many were bought out or went bankrupt by the end of the 90s, and this is long before the fading importance of mechanical drives was driving the consolidation of the industry we have now.

      [url<]http://en.wikipedia.org/wiki/List_of_defunct_hard_disk_manufacturers[/url<]

        • Kurotetsu
        • 8 years ago

        [quote<]Mechanical drives still fail quite often[/quote<] While that is true, its worth pointing out that when they do fail it tends to happen pretty quickly. Its preferable to have a drive that craps out before you put any data on it versus a drive that craps out over time.

          • rrr
          • 8 years ago

          Obviously not true, as anyone who accidentally dropped a HDD full of data can attest to.

            • Kurotetsu
            • 8 years ago

            ….I can’t tell if you’re being serious or not.

        • squeeb
        • 8 years ago

        I still remember Quantum. I had a Big Foot 6GB~ drive in my IBM Aptiva back in like ’99.

      • Firestarter
      • 8 years ago

      My Samsung 830 keeps me up at night, I get nightmares of having to use my office PC!

      • rrr
      • 8 years ago

      HDD industry proved no better, see Seagate’s firmware fiasco. And “a little” slower? Funny rhetoric. Random reads/writes are >>orders of magnitude<< slower than in SSDs – see benchmarks like CrystalMark or IOMeter. I have noticed that very clearly after going for SSD. Difference in a comfort of work is enormous,

      And for “data eating”, I have backups, like, you know, most sane people in the world, so that isn’t much of a concern. Ergo: stop scaremongering and get real.

      • kamikaziechameleon
      • 8 years ago

      Um, while I understand your apprehension of the SSD quality we continue to see fluctuating before our eyes… I would point out backups will resolve most of your concerns. I think if you look at drives of all type of late you’ll notice quality has dropped off the face of the map. BACKUPS! 🙂

    • Krogoth
    • 8 years ago

    This is clearly a workstation/server grade HDD tied to a customer grade interface.

    It is useful in areas where SSDs are too unreliable, but mainstream HDDs are not fast enough.

    SAS flavor of this HDD will fare much better.

    I cannot recommand Raptor for a desktop system. If you need to go nuts with HDDs. Just stick with 2-4TB mainstream units for the same or less while getting almost the same performance expect random access speed.

      • Airmantharp
      • 8 years ago

      I will say that the Samsung F3 1TB that I put in a sibling’s school/photo/video editing machine was lightning fast with 7 Pro and applications installed; I’d think that one of these would make for an excellent computing experience.

      Still, the cost is just too prohibitive- a 128GB 830 and a larger ‘green’ drive for storage makes much more sense.

      I can only really see this drive being useful in workstation situations as you mention.

    • nerdrage
    • 8 years ago

    [quote<]Removing the 2.5" drive from this sled requires little more than loosening a few Torx screws. [b<]One of those screws is covered by a warranty voiding sticker, though; remove with caution.[/b<][/quote<] Well that sucks. So if you wanted to use this in a laptop, you're SOL in terms of warranty? Or is the sled [b<]required[/b<] for cooling purposes, and these are not intended for laptops or other SFF systems?

      • FranzVonPapen
      • 8 years ago

      RTFM. The drive is 15mm thick; even if you didn’t have power consumption/heat dissipation concerns, this won’t fit your laptop’s drive bay.

    • willmore
    • 8 years ago

    Were the noise tests on the VR drives done with them in their carriers or ‘bare’?

      • Dissonance
      • 8 years ago

      In their sleds.

        • willmore
        • 8 years ago

        Wow, then they really are chatty. I’ve got a pair of 10K Cheeta drives in a server in the basement and even burried in the case, you know *exactly* what they’re up to at all times. That’s over the whine of the stacked, high speed, 60mm fans.

        Contrary to what some ePeen folks joke(?) you don’t want to try to put one of these in your laptop. You’ll get kicked out of the conference room/library/classroom faster than you can boot windows.

    • [TR]
    • 8 years ago

    Why does unscrewing the drive from the cradle void your warranty? Is it crucial to cooling the drive?

      • Farting Bob
      • 8 years ago

      Highly doubtful, but companies just love ways of not paying for a new drive if your one dies.

    • derFunkenstein
    • 8 years ago

    Page 2 – test system setup – your CPU is listed as:

    Intel Core i7 2500K 3.3GHz

    The link is to an i5 2500K, so my guess is that the i7 model should be an i5?

    edit: I can’t type!

    • Xylker
    • 8 years ago

    Before you go all “dinosaur-relic” on the mechanical HDDs, read this [i<]lengthy[/i<] interview with the Seagate CEO. [url<]http://www.forbes.com/sites/ericsavitz/2012/04/12/seagate-ceo-luczo-on-drives-zettabytes-flash-and-his-tattoo/[/url<] He makes some very interesting points. So, while this may not be the ideal drive for the ideal enthusiast system, it certainly has its applications.

    • revcrisis
    • 8 years ago

    Why didn’t they use a single 1TB platter??? Instead of three platters? I feel like a single platter at 10,000 rpm with fast access times would be an amazing HDD and something I would spend money for.

      • MadManOriginal
      • 8 years ago

      You’re right! Those silly WD engineers probably never even thought of that…quick, send them an email!

        • pragma
        • 8 years ago

        WD guys are so 20th century! Today, smart engeneer take one 300 MB platter and overclock to 1TB. Viola!11

      • willmore
      • 8 years ago

      Because that one platter would have to be larger. Or, are you asking why they aren’t using higher aerial density platters? That one is driven by electronics. The electronics can only handle a certain data rate. Since linear density * RPM determins that limit and linear density = sqrt(aerial density), RPM and aerial density are linked for a given speed of electronics. And that’s what causes them to use three lower density platters–that’s the highest density plattery they can spin at 10K and have the electronics function well. And, it just happens to work out to three platters.

      Also, keep in mind to keep the seek speed up, they don’t use as much of each platters surface as a normal drive. You can tell that (even measure it) by looking at the drop off in speed as you move along the drive surface. Note that the 3T Seagate drive drops by more than the new VR drive. If you look at the first 1T of the two drives and compare them, the 3T Seagate actually beats the VR drive–it’s only in those two extra T that it slows down more than the VR does–because the VR is using less of the platter surface.

      Doing that brings two benefits. One, it keeps the full drive read/write speed more consistant from beginning to end. Secondly, it means that there is less distance for the head to move between the beginning and end of the drive–so seeks will be faster due to the lessened head movement.

      It might be interesting to see some ‘short stroke’ tests added to this review. Compare the first 1T of the other drives with the full VR drive. The systemic tests will likely be unchanged–the filesystem isn’t using anything more than it needs to, anyway. But, some of the synthetic tests may make the other drive look better as compared to the VR drive.

        • cygnus1
        • 8 years ago

        So you’re saying they are short stroking the drive during manufacturing?

        [url<]http://en.wikipedia.org/wiki/Disk-drive_performance_characteristics#Short_stroking[/url<]

          • willmore
          • 8 years ago

          In the sense that they’re using less of the drive for stoarge than others. But, they do it for good, solid reasons. A 10K drive needs a larger hub due to the bearings needing to be larger and that they will dissipate more heat so the structure has to be adapted to handle that.

        • Bensam123
        • 8 years ago

        Reading up on this, it has more to do with physics then with how fast the heads can read and write. There are already drives in existence that spin faster then 10k RPMs with similar areal densities, such as Seagates Cheetah lineup.

        If they used the same platters they do in 7.2k drives in 10k or 15k drives they’d warp and distort. Platters for 7.2k drives are cheaper to produce and thinner as well.

          • willmore
          • 8 years ago

          Yes, there is rarely *one* limiting factor to a design decision.

      • FranzVonPapen
      • 8 years ago

      Why can’t I hold all these platters???

      • bcronce
      • 8 years ago

      The faster the RPM, the harder it is to read the data, which causes the 10k drives to have lower density than 7k and 5k drives.

      • Malphas
      • 8 years ago

      1 TB platters for 2.5″ form factor don’t exist yet…

    • kamikaziechameleon
    • 8 years ago

    So, Basically its a decent but poorly priced product. Its price makes it irrelevant. If they put it under the 200 dollar mark it could have uses otherwise its really a relic.

      • Airmantharp
      • 8 years ago

      One would hope that they drop them in price. A drive like this would make a good scratch/work drive for those that need space and speed, especially if they need more space and are willing to stripe a few together.

        • kamikaziechameleon
        • 8 years ago

        yeah, SSD’s need better RAID support ASAP.

          • Firestarter
          • 8 years ago

          Did you mean: RAID needs better SSD support?

    • yogibbear
    • 8 years ago

    Give me a 1TB SSD and I won’t even buy a mechanical HDD anymore.

    We must be only a few years off this being do-able. There’s a 1TB flash thumb drive out (only costs $3k mind you), so a 1TB SSD must be pretty close to reality, and then give it a couple of years to come down to a reasonable price and BAM no more Western Digital or Seagate. Sell your stocks now!

      • Corrado
      • 8 years ago

      Theres plenty of 512GB SSDs. I’m sure they COULD make a 1TB one, but in the end you could JBOD them into a single 1TB volume.

      • Krogoth
      • 8 years ago

      1TB SSDs do exist, but they cost 10-15 times more than 1TB Raptor……

      • rpsgc
      • 8 years ago

      OCZ has a few 1TB SSD.

      Consumer grade:

      OCZ Octane SATA-6Gbps 1TB
      OCZ RevoDrive X2 PCI-E 960GB

      Workstation / Enterprise (i.e. über expensive):

      OCZ RevoDrive 3 X2 PCI-E 960GB
      OCZ RevoDrive 3 X2 Max IOPS PCI-E 960GB
      OCZ VeloDrive PCI-E 960GB

      • Malphas
      • 8 years ago

      By which time 1 TB will be considered too small an amount for reasonable data storage. Mechanical hard drives are going to be around for a long time yet, just not for system drives or users with limited requirements. Most factors are pushing towards an ever increasing amount of storage, e.g. higher resolution cameras, faster Internet speeds, higher resolution video, larger videogame textures, timeshift recording, etc. The demand for storage is greater than ever.

      • ptsant
      • 8 years ago

      But do you really NEED to load everything in milliseconds? I mean why do you need to store your photos, videos and mp3s on an SSD? Once your system is up and running (OS, browser, office, a few files…) what do you need to process so quickly?

      And, by the way, even if you use SSDs for everything, the internet will not store itself… The big players (youtube, google, amazon etc) use HDs for maximum capacity and their storage requirements are enormous. So less HD consumption for typical users like you doesn’t mean that the overall demand for HDs is reduced.

        • Firestarter
        • 8 years ago

        Do you really NEED more than 1GB of RAM? Or 640kb for that matter?

        Hard drives have been the slowest component and a bottleneck for computing for DECADES. Since then, the pace with which other components have become faster has far outstripped that of hard drives. HDDs were slow back in 1992, and relatively speaking they’re even slower now.

        We’re so used to the limits of having to read and write data off a spinning disk, that what we do with a computer is in part limited by how slow that disk access used to be. SSDs will start to shine even more when they allow computers to do what they could never before with HDDs!

        For now though, it’s just a easy way to instantly transform any semi-decent PC into a very nice and smooth PC, and that’s ought to be reason enough.

        • ludi
        • 8 years ago

        Have you used a computer with an SSD? You really have no idea how much of your total computing experience is being defined by disk subsystem latency until you start removing it from the equation. Then, you don’t want to go back.

          • Coyote_ar
          • 8 years ago

          1Tb SSD??? really? for a workstation … sure depending on the use it may pay off. For a desktop that the most intensive usage it will see is gaming??? 120gb es enough, 240gb is a lot.

          I have a 120gb SSD, good for OS + regular use documents + 4-5 games i play. and still have some extra 30gb free for temp files etc.

          the day those 4-5 games exceed a 240gb SSD, then maybe its time for a 480gb one …

      • Bensam123
      • 8 years ago

      512GB SSD + 512GB SSD in Raid 0 array = 1024GB plus faster speeds!

    • Elsoze
    • 8 years ago

    I still don’t understand why the VelociRaptor hasn’t been made into an SSD hybrid

      • Corrado
      • 8 years ago

      It didn’t do much for those Momentus XT drives that were included here, so … seems kinda pointless.

    • Corrado
    • 8 years ago

    This review convinced me of one thing… I want to get one of those 3TB Barracudas. How has the reliability on those been, since I know Seagate hit a rough patch with their 1TB and 1.5TB drives not too long ago? The 3TB drive is $169 on Newegg right now. I need more storage to house my BD rips.

      • cygnus1
      • 8 years ago

      I don’t know about the Seagates, but I’ve had 5 3TB Caviar Greens in a Synology shoebox SAN server for the last ~6 months I’ve been over here in Afghanistan, in my tent… Seem to be pretty reliable so far. I bought a 6th one as a spare, but it’s been in a drawer this whole time.

    • Nutmeg
    • 8 years ago

    Like you said Geoff, the Raptor seems pretty pointless now when you can get a fairly decent sized SSD and a terabyte storage drive for the same price as it. This is probably the last one they’ll make, unless they keep the name for something else.

      • ShadowEyez
      • 8 years ago

      Well said. I (still) own the original 36 gig version and the 150 gig version, but now that SSD’s have come, it seems like… a dinosaur. Or that strange middle ground between the capacity and speed, with lots of noise made – my raptors always made lots of noise.

      Maybe they’ll rebrand the “Raptor” name for a new high end SSD…?

    • Ifalna
    • 8 years ago

    [quote<]Box 'o Silence[/quote<] Epic ^_^ Raptor looks good. Way too expensive of course, but I like it.

      • thanatos355
      • 8 years ago

      They should market a commercial model for discerning boyfriends everywhere.

      But how to convince the gf to climb inside while I’m gaming…hmm…

Pin It on Pinterest

Share This