review seagates momentus xt 750gb hybrid hard drive

Seagate’s Momentus XT 750GB hybrid hard drive

For most folks, hybrids are probably the future of PC storage. Flash memory’s near-instantaneous access times are a performance benefit too good to pass up. At the same time, NAND flash’s relatively high cost per gigabyte, combined with our ever-growing thirst for storage capacity, ensures mechanical platters will remain in a supporting role for the foreseeable future. Music and movie archives aren’t likely to benefit from the wicked-fast access times of solid-state storage, anyway.

On the desktop, cobbling together a hybrid solution is as easy as putting a hard drive and an SSD in the same system. Users are free to manage the distribution of data between the two or to rely on third-party software to treat the SSD as a cache for the hard drive. Those approaches work just as well in the mobile world, but only in notebooks that support dual drives, which are relatively rare outside hulking desktop replacements.

For much of the notebook market, the only way to enjoy a hybrid storage configuration is with a single drive that combines flash and mechanical components inside the same 2.5″ form factor. Seagate has been making such drives for several generations now, and it claims to have sold more than a million of ’em over the years. Most of those hybrids hailed from the original Momentus XT family, which debuted last summer.

Unlike Seagate’s previous hybrid efforts, which relied on the ReadyDrive caching mechanism built into Windows Vista, the Momentus XT took matters into its own hands with an OS-independent caching scheme run entirely within the drive itself. The end result melded 500GB of mechanical storage with 4GB of flash memory dedicated to caching frequently accessed data. Today, that first-generation Momentus XT is being replaced by a new model that combines higher-capacity platters with faster flash memory—and 8GB of it. We’ve taken the new hybrid for a spin to see if it can hang with not only its mechanical competition, but also the droves of solid-state drives that have flooded the market over the past year.

Momentus XT, take two
Like its predecessor, the new Momentus XT uses its flash memory solely as a read cache. Incoming writes from the host bypass the flash completely. The only time data is written to the flash is when the drive’s caching scheme, dubbed Adaptive Memory, is populating it with data from the mechanical platters. Third-party caching solutions like Intel’s Smart Response Technology allow flash storage to cache both reads and writes, and Seagate indicates it may move in that direction with its own hybrids. For now, though, the Momentus XT’s flash component will only be capable of speeding up the drive’s read performance.

Seagate’s Adaptive Memory caching tech targets frequently accessed data at the block level rather than the file level. To better take advantage of the new drive’s larger allocation of flash memory, the caching scheme has received a number of tweaks that Segate has wrapped up in a marketing name: FAST Factor. The most interesting of these tweaks may be the changes to how the caching system treats Windows files associated with the boot process.

Dubbed FAST Factor Boot, this new approach aims to accelerate the Windows boot process whether you’re firing up the OS for the first time or rebooting after months of uninterrupted activity. With the old Momentus XT, Adaptive Memory needed a few boots to learn which OS-related data to cache. Using the drive for long enough without rebooting had the potential to push that data out of the cache, reversing any previous boot-time benefits. With the new XT, a segment of the NAND has been reserved solely for OS data related to the boot process. That section of the flash is populated as Windows is installed to the drive, which should speed things up starting from the first boot. If Windows is being installed via a drive image, the XT will need a couple of boots to determine what to put in the roped-off section of its cache. As with clean installs, that data won’t be kicked out of the cache between boots.

On our sample drive, the cache is an 8GB SLC NAND chip made by Micron. Seagate says it will be sourcing flash from multiple suppliers and that the chips could be configured with either 4KB or 8KB flash pages. The original XT uses smaller 2KB block sizes, but like the newer drive, it’s also based on SLC-type flash. SLC memory typically has 10 times the write-erase endurance of the MLC NAND chips found in consumer-grade SSDs, making it ideal for caching applications. The 311 Series SSD Intel designed for Smart Response Technology also uses SLC NAND, albeit 20GB of it.

The Momentus XT’s 8GB cache is still double the amount of flash available in the original drive. The NAND is faster, too. Seagate claims the new XT can read from its cache at 180MB/s, a 40MB/s improvement over the first model. Both hybrids can purportedly transfer data from the platters to the flash at 100MB/s.

Those speeds might not sound all that impressive in the context of modern SSDs, but keep in mind that the Momentus XT has only a single flash chip. Seagate says it’s using a custom bridge design to extract as much performance as possible from that solitary chip. However, the Momentus doesn’t have anywhere near the parallelism of modern SSDs, which use multi-channel controllers to address flash arrays typically made up of 8 to 16 NAND chips.

The only other memory on the Momentus XT is a 32MB Samsung DRAM chip that’s used as a traditional drive cache. The DRAM chip isn’t any bigger than the one used on the original Momentus XT, but Seagate says the new drive’s move to a 6Gbps Serial ATA interface will give the cache a little more room to stretch its legs. The XT’s flash component isn’t quick enough to benefit from the faster SATA pipe.

Momentus XT G2 Momentus XT G1
SATA interface 6Gbps 3Gbps
Spindle speed 7,200 RPM 7,200 RPM
DRAM cache size 32MB 32MB
Flash cache size 8GB 4GB
Available capacities 750GB 500, 320, 250GB
Platter capacity 375GB 250GB
Areal density 541 Gb/in² 394 Gb/in²
Warranty length Five years Five years

We’ve spent a lot of time talking about the Momentus XT’s flash component, but there’s a hard drive lurking under the hood, too. The Momentus XT has dual 375GB platters spinning at 7,200 RPM. Each platter has an areal density of 541GB/in², which is a 37% increase over the old model. Higher areal densities can improve performance by putting more data under the drive head with each revolution of the platter, so the new XT should offer a nice step up in performance even when its flash memory is going unused.

The original Momentus XT was available in multiple capacities, but Seagate is limiting the incoming drive to a single 750GB flavor. The old XT’s lower capacity points simply didn’t sell well, the company says. The original 500GB model will persist alongside the new hybrid, but there’s quite a difference in price between them. Just hours before its official product announcement, Seagate changed the price of the 750GB XT from $189 to $245. The old 500GB model currently sells for around $150 online.

Both drives are covered by a five-year warranty, which is considerably longer than the three years of coverage commonly attached to consumer storage products, SSDs included. Only Western Digital’s Black series of hard drives matches the XT’s five-year warranty.

The twins get a cooling upgrade
Our intrepid team of storage test systems received a rather major upgrade to Sandy Bridge hardware a few months back. More recently, we swapped their Thermaltake SpinQ coolers for a pair of much beefier Frio air towers.

These dual-fan monsters situate a pair of 120-mm spinners on either side of a heatpipe-infused radiator. The three-pin fans hang on rubber mounts to dampen vibration, and their speeds can be tweaked using a pair of in-line rheostats that offer fine-grained control. With both fans spinning at their slowest pace, the Frio is reasonably quiet. Crank them up to full tilt, and Thermaltake says the cooler will dissipate up to 220W—more than double the TDP of the fastest Sandy Bridge CPU.

Our testing methods
When briefing me on the Momentus XT, Seagate was quick to extol the drive’s virtues for desktop applications, citing Dell’s use of the old model in some of its business-oriented Optiplex systems. This isn’t just a drive targeted at mainstream notebooks, so we have no qualms about pitting it against a wide variety of competitors that include a 3.5″ Caviar Black desktop drive, a huge selection of the latest and greatest SSDs, and a handful of more direct 2.5″ mobile competitors. The chart below summarizes all of the particulars of the drives involved. I’ve greyed out the SSDs here and in the graphs on the following pages to make things more readable—and to clearly separate a different class of products.

Interface Cache Spindle speed Areal density Flash controller NAND
Corsair Force Series 3 120GB 6GBps NA NA NA SandForce SF-2281 25-nm Micron async MLC
Corsair Force Series 3 240GB 6Gbps NA NA NA SandForce SF-2281 25-nm Micron async MLC
Corsair Force Series GT 120GB 6GBps NA NA NA SandForce SF-2281 25-nm Intel sync MLC
Corsair Force Series GT 240GB 6GBps NA NA NA SandForce SF-2281 25-nm Intel sync MLC
Corsair Performance 3 Series 128GB 6GBps 128MB NA NA Marvell 88SS9174 34-nm Toshiba MLC
Crucial m4 128GB 6GBps 128MB NA NA Marvell 88SS9174 25-nm Micron sync MLC
Crucial m4 256GB 6Gbps 256MB NA NA Marvell 88SS9174 25-nm Micron sync MLC
Intel 320 Series 120GB 3GBps 64MB NA NA Intel PC29AS21BA0 25-nm Intel MLC
Intel 320 Series 300GB 3Gbps 64MB NA NA Intel PC29AS21BA0 25-nm Intel MLC
Intel 510 Series 120GB 6GBps 128MB NA NA Marvell 88SS9174 34-nm Intel MLC
Intel 510 Series 250GB 6Gbps 128MB NA NA Marvell 88SS9174 34-nm Intel MLC
Kingston HyperX 120GB 6GBps NA NA NA SandForce SF-2281 25-nm Intel sync MLC
OCZ Agility 3 120GB 6GBps NA NA NA SandForce SF-2281 25-nm Micron async MLC
OCZ Vertex 3 120GB 6GBps NA NA NA SandForce SF-2281 25-nm Intel sync MLC
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 Scorpio Black 750GB 3Gbps 16MB 7,200 RPM 520 Gb/in² NA NA

The Momentus XT’s most direct competition comes from Western Digital’s Scorpio Black 750GB, another 2.5″ notebook drive with the same total capacity and 7,200-RPM spindle speed. There’s no flash memory inside the Scorpio, though, and it’s an older drive with less DRAM cache and a slower SATA interface. The Momentus XT 750GB’s hybrid competition will come from the original 500GB version of the Momentus XT.

Although its results are shown in all the performance graphs, we’re not too concerned with how the new XT fares against the Momentus 5400.4, a notebook drive that dates back to 2007. The 5400.4 will serve as our performance baseline for the value calculations that appear later in the review. It’s the slowest Serial ATA drive I could find in the Benchmarking Sweatshop.

We used the following system configuration for testing:

Processor Intel Core i7-2500K 3.3GHz
Motherboard Asus P8P67 Deluxe
Bios revision 1850
Platform hub Intel P67 Express
Platform drivers INF update
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 Series 3 120GB with 1.3 firmware
Corsair Force 3 Series 240GB with 1.3.2 firmware
Corsair Force Series GT 120GB with 1.3 firmware
Corsair Force Series GT 240GB with 1.3.2 firmware
Crucial m4 128GB with 0009 firmware
Corsair m4 256GB with 0009 firmware
Intel 320 Series 120GB with 4PC10362 firmware
Intel 320 Series 300GB with 4PC10362 firmware
Intel 510 Series 120GB with PPG4 firmware
Intel 510 Series 250GB with PWG2 firmware
Kingston HyperX 120GB with 320ABBF0 firmware
Corsair Performance 3 Series 128GB with 1.1 firmware
OCZ Agility 3 120GB with 2.15 firmware
OCZ Vertex 3 120GB with 2.15 firmware
WD Caviar Black 1TB with 05.01D05 firmware
Seagate Momentus 5400.4 250GB with 3.AAB firmware
Seagate Momentus XT 500GB with SD22 firmware
WD Scorpio Black 750GB with 01.01A01 firmware
Seagate Momentus XT 750GB with SM12 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. The Hybrid drives have also been subjected to five runs, but only in tests that show their performance improving after the first one.

  • 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 a line graph.

To make the line graph more readable, we’ve excluded the SSDs. The solid-state drives have also been greyed out in the bar graphs to focus our attention—and to keep the graphs from looking like a technicolor mess. The rest of the drives have been color-coded by manufacturer, with the Momentus XT 750GB highlighted in a different shade to set it apart from the other Seagate models.

The Scorpio Black has slightly higher sequential read speeds than the Momentus XT 750GB. Still, the new hybrid is a vast improvement over the old XT. Read speeds on the old XT fall off more dramatically than they do on the other mechanical drives.

Of course, neither the mechanical drives nor the hybrids are within striking distance of the SSDs. Even the slowest member of the solid-state pack has double the average read speed of the Caviar Black 1TB. The fastest SSD reads data at a whopping four times the rate of the Caviar.

Switching to HD Tune’s write speed test doesn’t change the dynamic between our 750GB notebook drives; the Momentus XT can’t quite catch the Scorpio Black. The gap between the Momentus hybrids is much narrower here than it was with reads. So is the distance to the closest SSD, which is only barely faster than the best the mechanical field has to offer.

Although SSDs tend to write slower than they read, the best of the current crop is still well ahead of the Momentus XT. The Crucial m4 128GB offers more than double the average write speed of the XT, and the fastest solid-state drives achieve more than triple the write rate.

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

Although the Momentus XT 750GB can’t sustain transfer rates that exceed the speed of a first-generation Serial ATA interface, short burst transfers will make use of the drive’s 6Gbps SATA link, if only just. The XT’s burst speeds are greater than those of its platter-bound compatriots—and even a few SSDs. This performance also represents quite a comeback versus the Scorpio Black, which is much slower with burst reads and writes.

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. We’ve also busted out the 4KB and 1MB transfers sizes into bar graphs that should be easier to read. Once again, we’ve dropped the SSDs from the line graphs for reasons that should become obvious when you look at the results.

At transfer sizes up to 64KB, the Momentus XTs are clearly serving HD Tune’s random reads out of their respective flash caches. The new model is a little bit quicker, but both are in the same ballpark as the SSDs and orders of magnitude faster than the purely mechanical drives.

The 1MB transfer size proves too large for Adaptive Memory to handle, bringing the access times of the hybrids in line with those of their mechanical cousins. The Momentus XT 750GB is only a millisecond ahead of the Scorpio Black here, but the new XT has a good 4.5 ms on the old one.

Interestingly, the 750GB Momentus XT and the Scorpio Black both have slower access times with 512-byte random reads. Those drives use Advanced Format, which segments the platters in 4KB rather than 512-byte sectors. We’re using the latest version of HD Tune, which is optimized for 4KB sectors.

Through the 64KB transfer size, the Momentus XT 750GB’s access times are a little quicker than those of the Scorpio Black. That situation is reversed at the 1MB transfer size.

The largest transfer size drags the SSDs out of sub-millisecond territory, but they’re still substantially faster than the hybrids and mechanical drives. Let’s see how that advantage translates to more real-world tests.

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:

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

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.

The Momentus XT 750GB lags behind the strictly solid-state drives in FileBench, but it’s consistently at the front of the non-SSD field. The XT is particularly proficient with the smaller files that make up the Mozilla and TR file sets, but I’m hesitant to credit Adaptive Memory. We tested the hybrids with five runs each, and their copy speeds didn’t improve by more than 1MB/s from the first run. Also, the original Momentus XT doesn’t fare that well with the Mozilla and TR file sets.

Whatever the reason, the XT delivers substantially faster copy speeds than the Scorpio Black. Even the 3.5″ Caviar Black desktop drive is put on the ropes by the hybrid.

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.

SSDs have always fared well in DriveBench, and they have a huge lead on the Momentus XT. Seagate’s newest hybrid remains comfortably ahead of the Caviar Black and the other mechanical drives, though. Let’s break down the individual workloads to see how things stack up.

With the exception of the copy workload, where the Momentus XT 750GB struggles mightily, the hybrid actually fares rather well. Each workload was run five times to let Adaptive Memory cache accordingly, and performance did improve through the first few runs. Even with those speedups, though, the Momentus XT is the slowest of the lot with the copy workload.

To be fair, the Momentus XT isn’t the only drive to suffer with one of the workloads. The Scorpio Black 750GB stumbles when we add video transcoding to our multitasking cocktail.

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 that it takes drives to complete an I/O request. We’ll start with an overall mean service time before slicing and dicing the results.

The Momentus XT 750GB makes a good first impression in DriveBench 2.0. While the solid-state drives remain firmly ahead, the hybrid has a healthy edge over everything else that has spinning platters.

Looking at the data split between reads and writes, it seems the Momentus XT 750GB’s flash cache is indeed getting a workout thanks to our two-week trace. The new hybrid outclasses its mechanical counterparts in reads, but it’s sandwiched between the Caviar and Scorpio Black in writes, which the flash can’t accelerate. Note the SSDs also fare better overall with reads than with writes.

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.

We’re looking at consistency rather than raw performance here, but the Momentus XT 750GB is stuck in the same old pattern. The drive is in a better competitive position with reads than it is with writes. The Seagate drive does offer more consistent access times than the Scorpio Black on both fronts, though.

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.

Here we have more evidence of the Momentus XT’s dual nature. The 750GB hybrid completes more reads in less than a millisecond than any other mechanical drive. For writes, however, it’s trumped by both WD drives.

At least the new Momentus XT is a clear upgrade over the old model. On the original Momentus XT, an awfully high percentage of both reads and writes take more than a millisecond to complete.

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, for which the Momentus XT is certainly not designed.

The randomized access patterns of our IOMeter workloads play right into the strengths of solid-state drives, which are orders of magnitude faster than their mechanical counterparts—hybrids included. That performance discrepancy makes the graphs unreadable with the SSD results included, so we’ve again excluded them.

Whatever prevents the old Momentus hybrid’s transaction rates from scaling up in IOMeter hasn’t infected the new XT. The 750GB hybrid largely shadows the Scorpio Black, with both notebook drives turning in slower transaction rates than the desktop Caviar. Surprisingly, our performance baseline, the Momentus 5400.4, doesn’t fare as poorly as one might expect.

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.

Even with our test system running off a disk image rather than a fresh install, the Momentus XT 750GB had no problem figuring out how to speed up the Windows 7 boot process. The drive’s first boot took nearly 15 seconds, but five of those seconds had been shaved off by the third iteration. We saw a similar improvement in boot times with the 500GB hybrid, which is still quite a bit slower overall. The old hybrid is quicker to load the OS than the Scorpio Black, though.

Impressively, the 750GB XT’s boot time brings it to within just two seconds of the fastest SSD. Let’s see if the hybrid can stay close when loading game levels.

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.

Adaptive Memory worked its magic in Duke Nukem Forever, but it didn’t have much of an impact in Portal 2. The latter loaded a second or two faster on the hybrids after the first run, while Duke load times were cut by 5-8 seconds.

Even without its flash cache hastening Portal 2 load times, Seagate’s latest hybrid is still faster than the Scorpio Black. As one might expect, the XT is more competitive with the solid-state field in Duke Nukem.

Noise levels
While some folks spent Black Friday stampeding through retail stores in search of deep discounts, I was holed up in the garage trying to spruce up our noise testing for hard drive reviews. We don’t have the space for anything elaborate, but I was able to whip up an 18″ x 20″ anechoic chamber on the cheap using 3/4″ MDF lined with acoustic foam. Say hello to the Box ‘o Silence, which is meant to isolate drives from the background noise of the Benchmarking Sweatshop.

Our TES-52 digital sound level meter slides into a hole drilled into one end of the box, and SATA power and data cables have been routed through another hole at the opposite end. Both holes are sealed tight with additional foam to prevent external sounds from seeping into the enclosure. To keep the lid nice and tight, elastic cord hooks onto anchor points at each corner.

The cord is leftover material from the suspension system designed to hold hard drives in place. Drives are suspended exactly 4″ from the sound meter’s foam tip and about the same distance from the box’s floor, ceiling, and side walls. 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 Momentus XT 750GB has impressively low noise levels at idle and under load. So does the Scorpio Black, which is only a fraction of a decibel louder. Props to Seagate for lowering the Momentus XT’s noise levels by several decibels; the old model is noticeably chatty in comparison.

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 Momentus XT 750GB has to power two 7,200-RPM platters in addition to 8GB of flash memory, so I can understand why it consumes more juice than the Scorpio Black. Rather than tasking Adaptive Memory with speculatively caching data that would allow the drive’s platters to be spun down more often or for longer, data is cached only if it will improve performance.

Note that the SSDs consume close to the same amount of power as the hybrids and mechanical drives. Switching to a solid-state drive is unlikely to deliver a substantial improvement in notebook battery life.

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 for all the drives, and we didn’t take mail-in rebates into account when performing our calculations.

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.

Even with the massive flooding in Thailand causing hard drive prices to skyrocket, SSDs remain an expensive proposition on a cost-per-gigabyte basis. The Momentus XT 750GB costs just 33 cents per gigabyte, while you’ll pay more than a dollar per gig for the cheapest of the solid-state options we tested.

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. This index uses a subset of our performance data described on this page of our last SSD round-up. Some of the drives were actually slower than our baseline in a couple of the included tests, so we’ve fudged the numbers a little to prevent those results from messing up the overall picture.

Our overall performance index rates the Momentus XT 750GB higher than not only the Scorpio Black, but also its desktop Caviar cousin. The new hybrid also scores higher than the old Momentus XT, which sits 25 percentage points adrift. As you can see, though, the SSDs are in another class entirely.

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. The plot gets a bit messy with all the SSD names included, so the solid-state drives are cloaked in gerbilesque anonymity. I’ve included the data points to illustrate the overall trend.

With better overall performance than its mechanical rivals, the Momentus XT 750GB sits atop the other inhabitants of the lower-left corner of our scatter plot. The drive’s relatively high asking price hurts its value proposition somewhat, but the premium doesn’t seem unjustified in light of the performance advantage the hybrid enjoys over the Scorpio Black. The SSDs deliver substantially better performance at a much higher cost per gigabyte, spreading the solid-state drives through the upper-right quadrant of the plot.

In general, there’s a bigger difference between the performance of SSDs and drives with mechanical underpinnings than there is between the individual members of each camp. That said, the plot clearly illustrates that performance tends to vary much more between different SSD models than it does with mechanical drives and hybrids.

Although this analysis is helpful when evaluating drives on their own, what happens when we consider their cost in the context of a complete system? To find out, we’ve divided our overall performance score by the sum of our test system’s components. Those parts total around $800, which also happens to be a reasonable price for a modern notebook.

With a suggested retail price of $245, the Momentus XT 750GB costs more than almost all of the 120-128GB SSDs we’ve tested. Solid-state drives offer a lot less storage capacity, of course, but this look at the numbers doesn’t take gigabytes into account. Although they might be in the minority among notebook users, some folks simply won’t need more than the 120GB of storage capacity provided by SSDs that cost less than the XT’s asking price.

The Momentus XT 750GB is a more refined hybrid solution than its 500GB predecessor. The flash memory still only works as a read cache, but it’s bigger, faster, and being used to greater effect. Seagate has upgraded the rest of the drive, too, giving it higher-density platters and a DRAM cache that’s just fast enough to take advantage of the 6Gbps SATA interface.

Our exhaustive suite of tests has exposed a couple of weaknesses in Seagate’s latest hybrid. The drive’s sequential transfer rates are a little behind those of the Scorpio Black 750GB in targeted benchmarks, and the XT’s poor performance in the copy component of our first batch of DriveBench workloads is cause for concern. Don’t be too concerned, though. The Momentus XT offers excellent real-world copy speeds, and it’s especially good at moving around large batches of small files.

While the Momentus XT boasts much better all-around performance than its Scorpio competition, Seagate must contend with a wave of solid-state drives that are faster across the board—in some cases by huge margins. The hybrid has one discrepancy on its side, though. With 750GB under the hood, the Momentus XT serves up hundreds of gigabytes more than the SSDs in its price range.

If you can squeeze everything you need into 120GB (or 250GB), then by all means spend a little bit less (or a little bit more) on solid-state storage. For everyone else, which probably includes most of the folks using notebooks as their primary PCs, the Momentus XT offers an excellent compromise between solid-state and mechanical storage. It would be nice if the drive were equipped with more flash and the ability to cache incoming writes, but the Momentus XT is still without peers. The only way to get anything comparable is to run dual drives with separate caching software, and that’s just not possible with a huge swath of notebooks.

I was completely sold on the Momentus XT 750GB when it was set to cost $189. With the drive now priced at $245, I’m not as enthusiastic. This is still a great alternative to standard mechanical disks for single-drive notebooks, and it’s TR Recommended for that segment of the market. However, this next-gen hybrid costs amost as much as a mechanical drive plus a small SSD, and I’d take the dual-drive combo in any system that could accommodate it.

0 responses to “Seagate’s Momentus XT 750GB hybrid hard drive

  1. If the HD would be conscious of the size of files they could move small ones into the flash. This would be a great OS RAID feature.
    The latency has been a constant but the number of files has increased.
    The HD is the bottleneck for normal PC use.

  2. They could make sense if the SS part would be in a similar volume region as the mechanical volume.
    One could couple 128GB Flash with a 384GB platter and interleave them.
    Any read bigger than 12kB would involve some Flash and reading it could fill the time until the head gets there. It would take a 4.8MB strip to fill all the time.
    The gain for a 4.8MB strip is -20% in delay. For big reads it is +33% rate.

  3. That’s because of lack of TLER I think. One thing to question is whether the firmware can recognize that it is in RAID configuration and deliver good hybrid performance with different RAID configurations.

  4. Sorry to resurrect a dead thread but it’s worth mentioning that this drive has come all the way down to $180. At that price, I’m not sure why you would buy anything else for a laptop.

  5. I’ve been using 2 of the original model in a RAID 0 setup for the past year. They have been rock solid.

  6. Makes sense, except I’ve seen both SSD’s and Mechanical actually say on the marketing materials “not for use in RAID”. The WD Green drives are like that, I think (or perhaps the Advanced Format).

    Given the odd setup in this case, I would really like to see if it would actually work. Any way TR can run some tests? Come on, we are relying on you guys before making some serious purchasing decisions – in the 10-100k ranges 🙂

  7. I don’t see anything that should stop them from working in RAID. The caching scheme is completely OS and driver independent, so the XT looks like a standard hard drive to the OS and, presumably, whatever RAID controller you might want to use.

  8. But having to ‘relearn’ the data would put it right back to the same as a mechanical drive with cache. All you’re really talking about is drive cache, I suppose HDD makers could continue to increase cache size but cache size above a certain amount is largely a marketing trick – we know from past HDD reviews that cache size (again, above a certain amount) isn’t a primary factor in HDD performance – the slow spinny platters are the bottlenecks and higher areal density gives much greater performance increase. HDD firmware tuning also makes a difference and is probably more significant than cache size.

    In your second paragraph you’re basically talking about replicating Windows prefetch. Important for pre-Vista and non-MS OSes I suppose if they don’t have comparable technology, but it would be redundant and perhaps less efficient to have the HDD do this. System RAM is also faster than a SATA connection anyway, by orders of magnitude.

    This drive is more like a self-contained version of Intel’s Smart Response which uses flash as a buffer to mitigate the slowness of mechanical drives.

  9. Yeah, it’s a problem if it loses it’s entire cache and has to repopulate it. It isn’t a critical lose that would be associated with a complete hardware wipe though. It would relearn the data. The circumstances it would be under for that to take place wouldn’t be a day to day scenario (depending on how long a capacitor could keep the memory charged). That is if the computer isn’t shut down properly.

    If it’s shut down properly, the drive could write a map of what files are cached on the disk and shut down (or update it depending on the data). When the drive powers back up it could check and see if there is data loss, if there is it could simply halt the learning process on the memory and restore the files when it’s idle.

    I was lamenting at the lack of hard drive cache upgrades in general. Technology like this could’ve been implemented awhile ago. It’s been a long time since hard drive caches have stagnated. It’s only been recently that flash prices have become affordable, let alone in the sizes appropriate for a hybrid drive.

  10. But every time you lose data you lose precious performance. The drive has to re-learn what to cache in the buffer. You don’t see that as a problem?

    You are right in that prices of components are much lower than end products. A quick search on [url=<]DRAMeXchange[/url<] tells me 2Gb chip costs around $0.70. A 4Gb chip, which is apparently the highest capacity currently available (not 8Gb as I falsely assumed before) will be at least double that, most likely more, but let's say they're $1.6. Multiply that by 16, and you get approximately $25. You are also right that SLC is more expensive than MLC - two to three times more, I think (read it somewhere, no idea where). So let's say we have price parity for capacity. But then you have to place 16 DRAM modules on that teeny-tiny PCB. And that's still not going to happen. And power consumption - it is [b<]very[/b<] different. [url=<]This document from Micron[/url<] says that a 1GB DDR3-1066 module populated by 8 1Gb devices uses about 7W of power during activity (65% of max bandwidth). You can guess how that figure changes if you raise capacity 8-fold - it will not be going down. Compare that with an entire SSD consuming around 2W during activity. A single SLC chip will consume a lot less than that. SDRAM as a replacement for the flash on [u<]this[/u<] drive is simply not workable. Maybe in a desktop drive, but power consumption will still grow significantly.

  11. For as often as the data loss would occur, it doesn’t matter. There is no permanent data loss associated with it and it is recoverable. It depends completely on how long the charge lasts.

    Check out SO-DIMMS. Memory prices for major distributors that are using the chips is completely different from the price of memory you buy on the market. The particular chip they use is SLC based and has it’s own premium associated with it. Chips in general are complex, I don’t think wiring complexity weighs into this at all. I never said this pertains exclusively to this drive, just hybrids in general. I don’t know how much power a single DDR chip uses and I’m guessing you don’t either. I’m pretty sure its on par with the flash equivalent though (this relates to my point about this not pertaining exclusively to this drive though).

  12. It helps in normal desktop use, look at the windows boot time. The cache doesn’t help in random read/write benchmarks, it helps in loading your previously opened office documents, or games. There is a benefit here, but I think the drive would be better served with more cache/storage and a lower price.

  13. Waitwaitwait…so you’re saying that you’re not impressed by this as an upgrade option?? I am shocked, just SHOCKED I tell you.

  14. I haven’t had time for extended testing, but it definitely cuts outside noise. The ambient reading drops by a couple of decibels just from putting on the lid. I’m also noticing less variability in the readings than I was used to seeing with our old method. I’m not sure I trust the box enough to do my noise testing during the day with other test systems running (the box has to sit right next to my test rack for the cabling to reach), but it should give us cleaner results than testing outside the, er, box.

  15. [quote<]Curiously, why does a hybrid drive need a non-volatile read cache? Volatile memory is oodles faster and quite a bit more reliable. They could simply use a small capacitor or battery to keep the cell charged while the system is powered down. If the data is lost, it really doesn't matter.[/quote<] Several reasons. I'll work under the assumption that you are suggesting replacing 8GB of flash with 8GB of RAM. 1. Size. Look at the PCB of that hard drive. Now imagine trying to squeeze 8GB worth of SDRAM chips on it. Even if you go with high-density 8Gb DDR3 chips you are looking at finding room for 8 of them. Not gonna work. 2. Circuit complexity. 8GB or SDRAM requires a 64-bit DDR3 bus at least. That's a LOT of signals to route. 3. Price. 8GB of DDR3 SDRAM can be had for about $80+. An 8GB flash chip costs more like $8 USD. 4. Power. Do you know how much power 8GB worth of DDR3 SDRAM uses? You can bet that it's a lot more than NAND flash. Not exactly ideal for a drive aimed at portables. 5. You are totally missing the point if you think it doesn't matter that data is lost between sessions. The SSD-like boot-up and application startup is one of the major attractions of this drive.

  16. Yeah, that’s why you could use a small capacitor to keep the memory charged for quite some time. If the capacitor is drained and the memory is lost, it could simply repopulate it without worrying about data loss.

  17. You state that the advantages are there:
    [quote<]It is certainly faster then a normal 7200RPM HDD as long as the data doesn't exceed the buffer,[/quote<] Then go on to stay something completely redundant like: [quote<]but it still falls short in front of a dedicated SSD unit.[/quote<] You have a real knack for stating the bleeding obvious. Of course it falls short of an SSD unit. The only way for a hybrid drive to catch up with an SSD is to integrate a full-on SSD. [quote<]It is more of a less, a run of the hull 7200RPM HDD with a very large buffer slapped on.[/quote<] Yes, and Sandy Bridge is just a Pentium III with a few tweaks.

  18. Short lifespan? You need to read more about SSD’s.

    I think Anand mentioned with the Intel 320 review that you could write [i<]staggeringly unrealistic[/i<] quantities of data, [b<]every single day for FIVE YEARS[/b<] and still be nowhere near having a dead SSD. Not only that, once your flash blocks reach the end of their life the disk isn't suddenly totalled, it'll just mark bad blocks off one at a time and your drive will ever-so-slowly lose capacity over time. Your 300GB worn out SSD may be only 280GB another couple of years down the line, but I sure as hell prefer that to the "OMG SUDDEN DEATH" that always accompanies a head crash or mechanical failure with traditional disks.

  19. It is certainly faster then a normal 7200RPM HDD as long as the data doesn’t exceed the buffer, but it still falls short in front of a dedicated SSD unit. Once the data size exceeds the buffer, it is no faster then the a run of the mill 7200RPM HDD.

    It is more of a less, a run of the hull 7200RPM HDD with a very large buffer slapped on.

  20. Average joe doesn’t care for nor need a SSD.

    Let alone a hybrid drive.

    Any of the current HDDs will satisfy their I/O needs.

  21. Geoff, how good is this pocket anechoic chamber? Have you done any tests to see how much it attenuates various sounds from the outside? Sure it is a great idea.

  22. I think it’s pretty clear he didn’t read the article at all.

    Once the price drops down it’ll be a nice option for someone who needs space AND speed in a single-bay laptop.

  23. Curiously, why does a hybrid drive need a non-volatile read cache? Volatile memory is oodles faster and quite a bit more reliable. They could simply use a small capacitor or battery to keep the cell charged while the system is powered down. If the data is lost, it really doesn’t matter.

    This really seems like tech that should be implemented already in current caches on the hard drive. :l

  24. If it only uses the ssd part as read mem buffer, than that could be interesting.
    The short life spam of full ssd is the real culprit, more than the price.
    Prices tend to go down anyway, but durability is often neglected in ict. I hate that.

  25. Windows is probably the only OS around these days that goes out of its way to stop you installing the OS (e.g. everything from C:\ except C:\users) on separate devices/partitions. It’s one of the reasons that “small” SSD’s are seen as unacceptable, since windows makes it hard for people to separate their programs from their data.

  26. Hybrid drives are a great idea….but I think users should have the ability to manually select files to be stored on flash.

  27. The drive uses SLC NAND. I highly doubt flash longevity would be an issue except in the most extreme of circumstances. However, I’d be wary of purchasing this drive for the promised write cache feature. I’d have to see it functioning well, both in terms of performance and data integrity, before getting too excited about it.

  28. I’m not entirely sure either of your points make sense. This particular hybrid drive has a slow mechanical component, if it was combined with your typical 3.5 that wouldn’t be the case, but it’s target would be different. The price tag associated with this drive is Seagate jacking the price, I’m sure it doesn’t cost almost 2x as much to make as the mechanical component without the hybrid addition.

    Now some of your argument could apply to this particular drive, but not hybrid drives as a whole and is quite conditional.

  29. I’m not sure what your angle is there other than possibly making fun of MS, but with Win 7 libraries the concept of an application and storage drive has been blurred by MS, and if they continue down that path it will be to the point where it’s transparent to the user.

  30. Maybe Geoff should rerun tests on the 500GB drive if firmware updates have made a signifcant difference?

  31. [quote<]The only caveat is that it requires two SATA ports, which doesn't exist on a laptop or a smaller form factor.[/quote<] Gee, maybe that's why Seagate only makes it in 2.5" form factor, where it actually does make sense?? (And some DTR laptops do have multiple 2.5" drive bays but they probably aren't very high volume.)

  32. I did the same thing for my older laptop and I agree. It was an extra $20 ($99 vs $79) versus comparable 500GB 2.5″ laptop drives and I was already spoiled because I had an SSD that it was replacing. It is noticeably faster than the lame 30GB drive I found in this laptop when I first got hold of it.

  33. [quote<]The Momentus XT is effectively a vanilla 7200RPM HDD with a very large write buffer.[/quote<] Did you bother reading the article? The XT uses the flash as a READ buffer, and uses it very well. OS startup times are biting at the heels of SSDs, and game load times also showed some improvement. It seems like a safe bet that other applications would get a boost, too. [quote<]The only caveat is that it requires two SATA ports, which doesn't exist on a laptop or a smaller form factor.[/quote<] And incredibly enough, that is exactly the segment Seagate is trying to address with this drive. The 2.5" form factor of the drive may have been a clue. [quote<]Hybrid SSD side's performance is still inferior to a pure SSD, even the budget units.[/quote<] Well, duh. The current MSRP makes the new XT drive a bad value proposition, but once prices come down to saner levels this will be a very attractive storage choice for laptops.

  34. I moved my 160GB Intel series 320 SSD from my old free laptop to my new llano laptop. That 160GB SSD is only $50 more than the $245 750GB Momentus XT. Then for my old laptop, for storing photos and a few apps that only run in XP; I bought the 500GB Momentus XT for $99.

  35. Try explaining the concept of an application drive and a storage drive to the Best Buy crowd.

  36. I wonder how well the dedicated boot flash space adapts to a dual boot setup, say a MacBook Pro with OS X and Windows in Boot Camp? Can the drive figure out which is your primary OS, do the OSs evict each other from the catch or does the dedicate boot flash space double to accommodate 2 OS starving caching of other files?

  37. This.

    In both hardware and software design there is well known concept that your speed is only as good as your most limiting step (in a process).

    Also, I think Geoff is being a little short sighted in thinking hard drives will continue playing a role for the foreseeable future. Mainstream computing systems (tablets, mac book air etc.) are already hard drive free.

    * “SSD Shipments to Achieve Compound Annual Growth Rate of 54% – IDC” – that was in 2010. At that rate, SSDs may completely displace hard drives in mainstream laptops and desktops by 2020…

  38. Yeah, I was going to ask what the margin of error was on your power measurements. I know the difference is not really worthy of further attention, but I keep looking at m4 128GB figures to justify an impending purchase and it just stuck out that only the m4 duo were in the “wrong” order. Thanks for taking the time to reply!

  39. Although caching incoming writes is quite feasible, it would severely limit the amount of write cycles the drive can sustain before it’s cache fails. these drives will probably still function if the cache fails because i’ts read only however a write back cache failing will corrupt data or make the drive unusable.

  40. Hybrid drives never made any sense.

    They try to combine the best of both worlds, but instead they simply inherit their weaknesses.

    The Momentus XT is effectively a vanilla 7200RPM HDD with a very large write buffer. Data transfers that exceed its write buffet size are immediately held by the mechanical aspect of Hybird. It costs more than a vanilla 7200RPM HDD on a GB/$$$$ basis and the speed of the “write buffer” doesn’t help in bulk data storage (multimedia, backups etc). Hybrid SSD side’s performance is still inferior to a pure SSD, even the budget units.

    You get far better value by investing in a dedicated SSD (boot, I/O demanding appliactions) and HDD (bulk data storage) combo. The only caveat is that it requires two SATA ports, which doesn’t exist on a laptop or a smaller form factor.

  41. Any idea if FAST Factor Boot is specific to Windows or if it works with other operating systems?

  42. The power numbers are correct according to my spreadsheet, but the difference there is quite slight. I actually have a 64GB m4 in the labs for an upcoming article, so we’ll see how that drive fares versus its larger siblings. Because SSDs perform garbage collection and other optimization routines during idle downtime, it’s possible that background processes were drawing more power with the 128GB drive, even though it was idling at the Windows desktop after the OS’ own background processes had completed.

  43. From a desktop perspective this new hybrid is a great option for SFF, home theatre, or all-in-one PCs. Speed, storage, low power, low noise, 2.5″.

    The 600GB Velociraptor seems to be over $300 now ($0.55 per GB). While it would probably outperform the Momentus XT overall, it wouldn’t be a clean sweep IMO.

    From a long range view, I think the demonstration of this soundly beating the closest mechanical equivalent, the Scorpio, bodes well for future hybrids. This drive shows more potential for hybrids than the earlier 500GB Momentus XT did.

  44. Perfmon/htop/sysstat on your host to see why it is sluggish. Almost certainly is disk/memory I/O related, a frequent Virtualization bottleneck when multiple guests on same box.

    SSD’s certainly would help. Don’t go with consumer-drives like TR reviews, go with an Enterprise SLC-based drive that is [b<]validated[/b<] for whatever your RAID controller is, and validate the software running on top with the various vendors. Good UPS+shutdown scripts+proper cache settings are required. It's going to be pricey, though, SLC+Lots of room is very, very expensive. If your guests don't do a lot of writing you could get away with MLC likely.

  45. Do you have enough drives to test in a RAID configuration? We are have Virtualization servers that we typically have in a RAID “3+1” setup, the 3 being for performance and +1 being parity. Still ends up being a bit sluggish due to so many virtual OS’es sharing the same disk array.

    Would LOVE to know if a RAID with these drives would improve the situation drastically – or not.

    A little scared to commit to full-on SSD’s right now due to firmware issues etc.

  46. That’s why I said, ” from a desktop user’s perspective”.

    I figured it was a fair comment, do to the fact it was being compared against SSDs, which are commonly being used as boot drives in desktop systems.

    Besides, if it had shown promise, I would have looked to it for my “next” boot drive.

  47. You realize this is a 2.5″ drive, right? I don’t think its really meant to be used in a desktop. The Caviar Black was included as a point of comparison only.

  48. I’d rather go Z68 than RAID0 on any other chipset; the greater chance of failure just isn’t worth the performance boost. RAID0 rarely scales linearly.

  49. A nice evolution but honestly, nothing new. This is still a 1st gen hybrid drive

    If this were a 2nd gen hybrid, I’d expect at least 16-32GB of flash per terabyte, and it wouldn’t be read-only either.

  50. From a desktop perspective, it doesn’t seem to be that much (if any) better than the Caviar Black.
    And in that price range, I think it would do comparatively worse against one of the new Velociraptors.
    Even with the current “Flood” prices, it’s value is questionable (at best) against the Caviar Black (again, from a desktop users perspective).

  51. The first Momentus XT was pretty good but came with one to many caveats, this Gen2 seems pretty good though. The problem is pricing. It competes with 120GB SSDs which is imho large enough for laptops, but after the price will come down to a more manageable 150$ it should sell pretty well.

    Would it be possible to do some performance testing with 2 of them in RAID-0? 16GB of RAID-ed SLC seems close in specs to Intel’s 311, but without Z68’s strings attached.

    Great review

  52. I would rather consider intel 311 mSATA 20G SLC + WD scorpio blue 1TB (two plates) with intel RST.
    Performance-wise or value-wise, the new MXT is definitely a bad choice.

  53. Thanks Geoff, very informative as always. The hybrid performs surprisingly well for a Seagate. It seems they’re finally also getting a handle on drive noise. I remember that Seagate drives were outstandingly quiet back in the days of Barracuda IV and V, then they lost the quiet crown with 7200.7 and beyond.


    On page 3 there’s a typo: “color-colored” instead of color-coded.

    If we are to believe the graphs on page 10 the 256GB Crucial m4 has lower idle power consumption than the 128GB model. Did the figures get swapped or is this the correct data? All the other drive models represented by two different capacities have either identical figures or a lower number for the lower capacity.

  54. The retail price should end up lower than the recommended price,it always does for HDDs (and this is floods pricing).
    Curious to see the next gen,now that LSI has bought Sandforce, maybe they can make a much better HDD SoC for hybrids, ofc assuming Seagate keeps using LSI.

  55. They should totally just call it the Hybrid and at least subconsciously cash in on the whole ‘green’ angle even if they don’t get silly and try to market it as green (unlike the silly Taiwanese mobo and PSU makers who would have you believe that producing new items that save a few percent in power actually grows trees, lol)

  56. I think hybrids really are the way to go, at least for mechanical drives. If Seagate is serious about this, they should spread this technology throughout their entire product line, not just one drive for notebooks. Obviously hybrid solutions still have weaknesses. Higher density, higher internal transfer rates, and faster seeks on the mechanical side would help to aide this.

    Since they’ve already started on the groundwork R/D for such a solution, it seems like a shame to limit it to a certain niche part of the market. It reminds me of what Creative did with on board memory for their sound cards.

    These drives may also suffer from a product name as well. Unless they try and tell consumers it’s a hybrid drive in big bold letters or tell them how these drives benefit from being a hybrid people wont buy them. Momentus has always been a traditional mechanical laptop hard drive name.

    They could name the new line ‘Seagate Hybrid’ or something cachey like that to differentiate them from mechanical drives and introduce a bunch of new drives for both the 2.5 and 3.5 form factor. Like Geoff said, they do need to add a bigger cache though, I agree with that.

  57. Maybe the price hike is temporary because of the Thailand flooding.

    Anyway, looking over the performance, even though SSDs still generally beat it especially wehn IOPS count (duh) I think this drive should be the standard, or a no-cost alternative, recommended by Intel for ‘ultrabooks’. The capcaity is important in this day and age of digital media, and in some ways like boot time it is in the ballpark with SSDs. I would totally want a drive like this in a notebook I own until SSDs drop at least another 50% in $/GB to avoid the pesky alternative of additional or external mass storage.

  58. I think I’m happy with the 256GB Crucial M4 I just ordered, [url<][/url<]