Seagate’s Barracuda 7200.9 hard drives

Manufacturer Seagate
Model Barracuda 7200.9
Price (500GB)

Availability Now

IN THE HARD DRIVE WORLD, capacity is king. Capacity actually comes in two flavors. The first and most obvious is the storage capacity of a hard drive. The second, and potentially more important, is the capacity of individual platters within a drive. This platter capacity is known as the areal density, and it can do more than just increase the amount of data a drive can hold. Higher areal densities allow drives to offer more storage capacity with fewer platters, potentially lowering noise levels, cutting power consumption, and reducing the risk of a catastrophic head crash. As if that weren’t enough, higher areal densities can also improve performance by allowing the drive head to access the same amount of data over a shorter physical distance.

Seagate’s new Barracuda 7200.9 family of Serial ATA hard drives packs storage capacity on both fronts, with one model weighing in at a hefty half-terabyte and another packing a single 160GB platter whose areal density is 25% higher than its closest competitor. We’ve rounded up both models and run them through a brutal gauntlet of storage tests against earlier Barracudas and drives from Hitachi, Maxtor, and Western Digital. Read on to see how the 7200.9s compare.

Drive specs
Although the Serial ATA Revision 2.5 spec has yet to be ratified by the Serial ATA International Organization (SATA-IO), all of the draft spec’s standard and optional features are implemented across the Barracuda 7200.9 line. Those features include, but are not limited to, provisions for staggered spin-up, hot plugging, auto-negotiated backwards compatibility, ClickConnect connectors, and (most importantly) Native Command Queuing (NCQ) and 300MB/s host transfer rates.

  Barracuda 7200.9 Barracuda 7200.8
Maximum external transfer rate 300MB/s 150MB/s
Average seek time 11ms
Average rotational latency 4.16ms
Spindle speed 7,200-RPM
Available capacities 80, 120, 160, 200, 250, 300, 400, 500GB 200, 250, 300, 400GB
Cache size 8MB (80-250GB)
16MB (300-500GB)
Platter size 100GB (200, 300GB)
120GB (120GB)
125GB (250, 500GB)
133GB (400GB)
160GB (80, 160GB)
Idle acoustics 2.5-2.8 bels 2.8 bels
Idle power consumption 6.9W 7.2W
Seek  power consumption 8.1W 12.4W
Service life Five years
Warranty length Five years

300MB/s Serial ATA transfer rates are all the rage, but not even the fastest 15K-RPM SCSI drives can sustain transfer rates that saturate a 150MB/s Serial ATA interface, giving 7,200-RPM drives little hope of benefiting from anything faster. However, 300MB/s host transfer rates can allow for faster burst transfers from a hard drive’s cache. With drive cache sizes growing, it may be unwise to write off 300MB/s Serial ATA transfer rates as a gimmick just yet.

Speaking of cache, drives in the 7200.9 family are equipped with either 8MB or 16MB, depending on their capacities. Higher capacity drives get more cache, with the 300GB, 400GB, and 500GB models packing 16MB each.

The 7200.9 line is also segmented when it comes to platter density. There are quite a few variations, with five different platters spanning eight drive capacities. Today we’ll be focusing our attention on the 500GB and 160GB capacities, which feature 125GB and 160GB platters, respectively. 125GB platters are hardly unique in the hard drive world, but they allow Seagate to build a 500GB drive with only four platters. The only other 500GB drive on the market, Hitachi’s 7K500, uses five 100GB platters to reach the half-terabyte mark.

The Barracuda 7200.9 500GB (left) and 160GB (right)

While 125GB platters aren’t all that special, no one can match Seagate’s ultra-dense 160GB platters. Only Seagate’s own 133GB platters come close, and even then, the 160GB platters offer a 25% jump in areal density. That allows the 160GB Barracuda 7200.9 to be built using just a single platter, with all of the attendant advantages of fewer platters. The 160GB model’s denser platter should also provide a nice performance boost. You won’t find 160GB platters in drives larger than 160GB, though, and these platters aren’t available in drives with more than 8MB of cache, either.

Fortunately, all members of the Barracuda 7200.9 family are covered by Seagate’s fantastic five-year warranty. This longer warranty term is usually reserved for SCSI drives and enterprise-class products, while desktop hard drives from other manufacturers are generally only covered for three years. Seagate, however, warrants its entire internal hard drive line for five years.

Barracudas over easy

Test notes
We’ll be comparing the 160GB and 500GB flavors of the Barracuda 7200.9 against an array of competitors. Most of the drives we’re testing are desktop parts, but the Western Digital Caviar RE2 and Raptor WD740GD are technically enterprise-class products. We won’t tell anyone if you don’t.

The drives we’ll be looking at differ when it comes to external transfer rates, spindle speeds, cache sizes, platter densities, NCQ support, and capacity, all of which can have an impact on performance. Keep in mind the following differences as we move through our benchmarks:

  Barracuda 7200.7 NCQ Barracuda 7200.8 Barracuda 7200.9 (160GB) Barracuda 7200.9 (500GB) Caviar SE16 Caviar RE2 Deskstar 7K500 DiamondMax 10 Raptor WD740GD
Max external transfer rate 150MB/s 150MB/s 300MB/s 300MB/s 300MB/s 150MB/s 300MB/s 150MB/s 150MB/s
Spindle speed 7,200RPM 7,200RPM 7,200RPM 7,200RPM 7,200RPM 7,200RPM 7,200RPM 7,200RPM 10,000RPM
Cache size 8MB 8MB 8MB 16MB 16MB 16MB 16MB 16MB 8MB
Platter density 80GB 133GB 160GB 125GB 83GB 100GB 100GB 100GB 37GB
Capacity 160GB 400GB 160GB 500GB 250GB 400GB 500GB 300GB 74GB
Native Command Queuing? Yes Yes Yes Yes No Yes Yes Yes No*

Note that the Western Digital Caviar SE16 and Raptor WD740GD lack support for Native Command Queuing. The Raptor does support a form of command queuing known as Tagged Command Queuing (TCQ), but host controller and chipset support for TCQ is pretty thin. Our Intel 955X-based test platform doesn’t support TCQ.

Since Seagate makes versions of the 7200.7 with and without NCQ support, the 7200.7 in our tests appears as the “Barracuda 7200.7 NCQ” to clarify that it’s the NCQ version of the drive. The Deskstar T7K250, DiamondMax 10, 7200.8, and 7200.9 aren’t explicitly labeled as NCQ drives because they’re not available without NCQ support.

Our testing methods
All tests were run three times, and their results were averaged, using the following test systems.

Processor Pentium 4 Extreme Edition 3.4GHz
System bus 800MHz (200MHz quad-pumped)
Motherboard Asus P5WD2 Premium
Bios revision 0422
North bridge Intel 955X MCH
South bridge Intel ICH7R
Chipset drivers Chipset
Memory size 1GB (2 DIMMs)
Memory type Micron DDR2 SDRAM at 533MHz
CAS latency (CL) 3
RAS to CAS delay (tRCD) 3
RAS precharge (tRP) 3
Cycle time (tRAS) 8
Audio codec ALC882D
Graphics Radeon X700 Pro 256MB with CATALYST 5.7 drivers
Hard drives Seagate Barracuda 7200.8 400GB SATA
Seagate Barracuda 7200.7 NCQ 160GB SATA
Maxtor DiamondMax 10 300GB SATA
Western Digital Caviar SE16 250GB SATA
Western Digital Raptor WD740GD 74GB SATA
Hitachi 7K500 500GB SATA
Western Digital Caviar RE2 400GB SATA
Seagate Barracuda 7200.9 160GB SATA
Seagate Barracuda 7200.9 500GB SATA
OS Windows XP Professional
OS updates Service Pack 2

Our test system was powered by OCZ PowerStream power supply units. The PowerStream was one of our Editor’s Choice winners in our last PSU round-up.

We used the following versions of our test applications:

The test systems’ Windows desktop was set at 1280×1024 in 32-bit color at an 85Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.

All 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.


WorldBench overall performance
WorldBench uses scripting to step through a series of tasks in common Windows applications. It then produces an overall score. WorldBench also spits out individual results for its component application tests, allowing us to compare performance in each. We’ll look at the overall score, and then we’ll show individual application results alongside the results from some of our own application tests.

WorldBench’s overall results are pretty close, with most of our drives tied for second place behind the Raptor. The 7200.9s do score marginally higher than older Barracudas, though. Let’s bust WorldBench’s overall score into individual application tests for a closer look.

Multimedia editing and encoding

MusicMatch Jukebox

Windows Media Encoder

Adobe Premiere

VideoWave Movie Creator

There’s little variance in scores across WorldBench’s multimedia editing and encoding tests, although the 160GB 7200.9 pulls up a little short in Premiere.


Image processing

Adobe Photoshop

ACDSee PowerPack

Photoshop doesn’t seem to prefer one drive over the others, but ACDSee spreads the field a little. There, the 7200.9s are generally slower than their competition, but not by much.

Multitasking and office applications

Microsoft Office


Mozilla and Windows Media Encoder

The 7200.9s tie the rest of the field across WorldBench’s multitasking office tests.

Other applications



The new Barracudas shine in WinZip and Nero. They’re faster than everything short of the Raptor in WinZip, and take both top spots in Nero. Interestingly, the 500GB 7200.9 comes out just ahead of the 160GB drive in both instances, suggesting that cache size may be more important than platter density in these tests.


Boot and load times
To test system boot and game level load times, we busted out our trusty stopwatch.

The new Barracudas don’t fare so well here, finishing at or near the bottom of the pack in two of three tests.


File Copy Test
File Copy Test is a pseudo-real-world benchmark that times how long it takes to create, read, and copy files in various test patterns. File copying is tested twice: once with the source and target on the same partition, and once with the target on a separate partition. Scores are presented in MB/s.

Although the 160GB Barracuda 7200.9 fares rather well in FC-Test’s file creation tests, its 500GB sibling looks a little slow compared to Hitachi’s 500GB 7K500. Note the consistent gap in performance between the 160GB and 500GB flavors of the 7200.9.

The performance gap between the 160GB and 500GB Barracudas shrinks in FC-Test’s read tests, but the drives still hang around the middle of the pack.

Results in the copy tests are more mixed, as the 500GB 7200.9 edges out the 160GB model with a number of test patterns. Overall, the new Barracudas look better here than they have with FC-Test’s file creation and read tests.

The Barracuda 7200.9s also perform comparably well in FC-Test’s partition-to-partition copy tests, with the 160GB and 500GB models trading blows across all five test patterns.


iPEAK multitasking
We recently developed a series of disk-intensive multitasking tests to highlight the impact of command queuing on hard drive performance. You can get the low-down on these iPEAK-based tests here. The mean service time of each drive is reported in milliseconds, with lower values representing better performance.

Results are mixed in our first round of iPEAK multitasking tests, but a couple of patterns do emerge. The Barracuda 7200.9s seem to fare better with multitasking loads that include file copy operations, while they struggle with the VirtualDub import. Cache size also appears to be more important than platter density in these tests, with the 500GB 7200.9 beating the 160GB model across the board.


iPEAK multitasking – con’t

The same patterns persist throughout the remainder of our iPEAK tests, with the 7200.9 faring comparably better with multitasking loads that include file copy operations. Again, we see the 500GB drive coming out ahead of the 160GB model, likely because of the former’s larger cache.


IOMeter – Transaction rate
IOMeter presents a best-case scenario for command queuing, so the NCQ-less Western Digital drives should have a slight disadvantage here under higher loads.

The 7200.9s don’t fare nearly as well in our multi-user IOMeter tests as enterprise-class Serial ATA drives like the Raptor and Caviar RE2. Still, it’s worth noting that the 160GB 7200.9 performs better than most of the other desktop drives, including its 500GB sibling. Also note that the 500GB 7200.9 is much faster than the 500GB Deskstar with three of four test patterns.


IOMeter – Response time

The 500GB Seagate drive continues to provide better IOMeter performance than its direct competition, but overall, the 7200.9s seem poorly suited to multi-user loads.


IOMeter – CPU utilization

CPU utilization is low pretty much across the board.


HD Tach
We tested HD Tach with the benchmark’s full variable zone size setting.

The 160GB Barracuda 7200.9’s HD Tach transfer rates make a compelling case for higher platter densities. Despite its huge spindle speed disadvantage, the Seagate drive almost matches the performance of the 10K-RPM Raptor. However, higher areal density doesn’t help the 500GB Barracuda distance itself from the Deskstar 7K500.

The 7200.9s both come out on top in HD Tach’s read bust speed test, although their performance is still well shy of the 300MB/s theoretical peak bandwidth of our system’s Serial ATA interface.

HD Tach’s random access time test proves a little problematic for the 500GB 7200.9, but the 160GB drive’s performance is much better.

CPU utilization results are well within HD Tach’s +/- 2% margin of error in this test.


Noise levels
Noise levels were measured with an Extech 407727 Digital Sound Level meter 1″ from the side of the drives at idle and under an HD Tach seek load. Drives were run with the PCB facing up.

At idle, the 7200.9s are nearly silent. Results are split under a seek load, though. The 160GB drive proves quieter than any other we tested, but the 500GB model is among the loudest. Extra capacity alone isn’t the culprit, since the Deskstar 7K500’s seek noise levels are a full two decibels lower than the 500GB 7200.9.

Power consumption
Power consumption was measured for the entire system, sans monitor, at the outlet. We used the same idle and load environments as the noise level tests.

The 160GB 7200.9 proves the most frugal with power, perhaps in part because its mechanical motor only needs to spin the weight of one platter.


First, we should applaud Seagate for integrating all of the Serial ATA 2.5 draft spec’s standard and optional features across the entire Barracuda 7200.9 line. This move ensures that users will enjoy a common set of features regardless of which drive capacity they choose, and these days, every new SATA drive should support Native Command Queuing, 300MB/s transfer rates, hot plugging, and the like. Unfortunately, the Barracuda 7200.9 family doesn’t standardize on a common cache or platter sizes, resulting in a lineup littered with seven combinations of the two. Each unique combination will offer slightly different performance characteristics, so we’ll have to limit our analysis to the 160GB and 500GB models we’ve tested.

Fortunately, we tested the hell out of them. The results of our tests are rather enlightening, too. For example, the 160GB 7200.9’s high density platter clearly offers lower random access times and better performance with sustained transfers and multi-user loads. The drive is also much quieter than the 500GB model, and its power consumption is a little lower. However, the 500GB drive’s extra cache gives it an edge in desktop applications, some file transfers, and in our disk-intensive multitasking tests. The 500GB drive does cost close to four times more than the 160GB model, but you get more than three times the storage capacity in the same 3.5″ form factor.

Then there’s the not-so-small matter of the half-dozen or so competing drives from other manufacturers, some of which narrow the 7200.9’s appeal. Seagate’s new Barracudas are clearly outmatched by Western Digital’s enterprise-class Serial ATA drives in multi-user environments, making it hard to recommend the ‘cudas for servers or even workstations with heavy I/O demands. The 7200.9s are a considerably more attractive option for single-user desktops, but they aren’t consistently faster than the competition, so performance alone won’t make them a favorite. Seagate’s five-year warranty just might, though. For many users, an extra two years of warranty coverage may be worth overlooking the 7200.9 drives’ poorer performances in some of our tests. The 500GB 7200.9 is currently the most affordable half-terabyte drive on the market, too.

Looking back on the Barracuda 7200.9 line, I can’t help but feel a little teased. Those 160GB platters are oh-so tempting, but it seems cruel to pair them with only 8MB of cache. I’m not crazy about being limited to 160GB of capacity per drive, either. Let’s hope Seagate is able to squeeze a couple of those 160GB platters into a drive with 16MB of cache. I’d take two. 

Comments closed
    • Grigory
    • 14 years ago

    I don’t care for speed. Can I have my 1 TB drive now? Thanks.

      • Yahoolian
      • 14 years ago


    • Ryu Connor
    • 14 years ago


    • albundy
    • 14 years ago

    oh, god! not another 7200rpm drive. This is just as bad as more x86 cpu news…or even worse. How long can they possibly milk a slow rpm drive?

      • babybalrog
      • 14 years ago

      They are milking storage not speed, sadly.

      • Krogoth
      • 14 years ago

      It’s very difficult to improve a technology that has already met some phyiscal limitations. 10K and 15K drives need more robust motors and ultra-light plattes for any good long-term reliablity. The only improvement in the HDD industry for the last couple of years is increasing areal denstiy= more GBs and better STR performance. The next big thing is perpendicular read and write which will result in even higher areal densites.

      Soild-State Drives are faster, but are too darn expensive per GB for 99% of the market. 3d Optical media and MRAM are too experimental and will likely not see the mainstream marketplace until the decade.

    • swaaye
    • 14 years ago

    I have a hard time tangibly noticing the difference between the 8MB drives produced in the last few years. I have WDSEs and a few Seagates. The known-to-be slow 7200.7 drives are quite fast enough for me. Only when I go back to 7200’s from ’00 or so, like my Quantum Fireball Plus KX rocks-in-a-blender, or maybe my 75GXP (yup, still works!), do I feel it being noticeably slower.

    Actually, by FAR my primary concern when picking out a drive is whether or not it’s basically silent. I don’t want to hear HDDs anymore.

    When you’ve progressed from a 286 w/ a 20MB hard card to a system today with 200GB storage, the short-term speed increases of mechanical hard drives feels a bit trivial. The first hard drive I was excited to own was a Quantum Fireball 1280A in ’95 or so. Loyd Case gave that sucker a glowing review, and sure enough, it was a ton faster than my 210MB Maxtor. $220 for a 1.2GB drive with 128kb cache. 🙂

    Hard drives aren’t fast in any definition in my book. The difference in speed between system RAM and that wonderful spinning mechanical unreliability is sickening. I want something new.

      • indeego
      • 14 years ago

      This I agree. Other than the Raptors, there really hasn’t been much out there that is revolutionizing this segment. Silence is golden: period. There’s no chance I could install a system for a friend or at work where the drive is loud and churning– it’s a sign of something wrong, as far as the customer is concernedg{<.<}g

    • jim100flower
    • 14 years ago

    Thank you for including platter density in your analysis. This is the review I had hoped to read at last month.

    • Krogoth
    • 14 years ago

    #10, you forgot that the 74GB Raptors have been out for two years. It has taken 7200RPM drives with far greater areal density that long to reach the Raptor in STR performance. The Raptors still rock in random access speed department due to their higher spindle speed.

    #12 Higher-spindle drives are lighter on capacity due to the fact their plattes are less massive. A few grams less makes a difference in long-term motor reliablity at 10K and 15K RPM.

    • Freon
    • 14 years ago

    I’m still not sold on NCQ. TR seems to be stuck on it. Sure it probably doesn’t hurt, but I don’t see it as something to really even be shopping for.

    The SE16 only looks particularly bad in a few of the iPeak tests, and it’s on the slow half in IO meter (which is really a server oriented test, IMHO), but hardly awful. The Barracudas being reviewed actually seem to look worse on average in the iPeak tests. The only load time test where it is poor is Doom 3, but its only about 1 second out of ~30 behind the average, and that was tested with a stopwatch. Looking at this review, I can’t say I’d necessarily choose the 7200.9 with NCQ over the SE16. I’d buy whichever was cheaper and had the best warranty.

    I just don’t see a strong positive trend and never have. Get a Raptor drive (NCQ-less) for your main system drive, and with a 10k spindle it pretty much cleans house because spindle speed means very real and consistent performance gain. The rest of your drives? Probably doesn’t matter. It’s a wash. Harddrives are a commodity. Buy a brand you trust not to take a crap.

    I just can’t justify even mentioning NCQ to any of my friends or telling anyone to actually shop based on that feature.

    NCQ is just not worth a mention. It’s not worth my time to worry about. The difference between two random drives with NCQ is no more or less than a drive with compared to a drive without. I just don’t get it.

    • FireGryphon
    • 14 years ago

    Excellent review, as always.

    Perhaps they don’t beat the competition every time, but the 7200.9’s look like solid drives with impressive feature lists that will probably find their way into my next system.

    • IntelMole
    • 14 years ago

    In other news, 20 gigs of storage is suddenly much more important that using three 160 gig platters.

    C’mon Seagate, is it really that much that your customers will bitch at you for it? :-S,

      • spiralscratch
      • 14 years ago


    • droopy1592
    • 14 years ago

    If size is not important, I’d go Samsung for speed and silence any day before Seagate.

      • adisor19
      • 14 years ago

      If warranty is important, there is only seagate and some select WD drives..


        • droopy1592
        • 14 years ago

        Samsung warranty is 3 years, Seagate is 5. I’d bet willing to be your main drive would be replaced before a 3 year period. I rarely keep a drive more than 2.5 years because I want a tad more speed and more capacity. I put my old drives in external enclosures. The idle seek noise on the Seagate 7200.8 drives me nuts.

          • adisor19
          • 14 years ago

          No way, why would i want to scrap a nice 250G drive for no reason ?? It’s not like i have money to throw away for a little 5% performance boost.. and if the drive dies in 4 years, i would defiently prefer a free replacement then to have to spend for a new one.


            • Chrispy_
            • 14 years ago

            I have use for any hard drive that still works fast enough to use, that includes a 10GB drive from 7 years ago.

            5 years and 3 years have a fight, 5 wins.

      • fatpipes
      • 14 years ago

      Oops, wrong thread

    • Bensam123
    • 14 years ago

    Almost every serious gamer I know has a setup with Raptors in it or a raid 0 array. Why haven’t companies produced more 10,000 RPM or 15,000 RPM Sata drives?

    In the article the new Cudas come close to the Raptors max transfer rates and that says something from a drive that is 2,800 RPMs slower. Thats like comparing a 5,400 drive to a 7,200. If memory serves me right it didn’t take nearly this long to step up from 5,400’s to 7,200’s.

    Where is the gamers hard drive? Gamers do have a budget and can’t afford jamming a scsi-320 controller in their system and a array of 15k rpm drives but they will pay for that extra edge as long as it’s reasonable.

    Wonder which HD company will jump on it first. I thought WD was doing that but they truely now seem to be targeting enterprises with their Raptor series.

    Native Sata 2.5, 15k RPMs, 32mb Cache (why not, memorys not expensive is it?), 32 or 74 gigs (you only need the space for your OS and game installs), clear case (oooh spinning drives), a upper 100 to lower 200 dollar price range and you’d have yourself a nice niche.

    Why hasn’t it been done? Why are hard drives pretty much the only thing in the computer industry that hasn’t changed over the past few years? Is it that hard to change magnetic storage? Do I just not understand some fundamental concept that everyone else knows? By the time they actually make some decent advance that trickles its way down to the halfway insanely affordable range, it will be cheap enough to just outfit your system with RAM (forgot the acronym for the kind that remembers everything after its powered off). It is almost to that point already.

    Want a radical idea? Design a 7,200 RPM hard drive with built in Raid 0. Off the top of my head but just make a drive that would fit in a 5.25 bay or 3.5 if possible with built in hardware raid or just two seperate drives in the same enclosure that have two plugs. Almost all chipsets have built in raid controllers now. What would happen if you raid a already hardware raided array? Is it even possible? You can see how this just stems from the approach of two things working together if it’s not feasible to just make one thing go faster.

    Sorry, my brain was about to explode so it dumped and BSOD’ed.

      • wierdo
      • 14 years ago

      perhaps the “serious gamer” crowd is a small niche that doesn’t attract enough competition at this point 😛

      • Shintai
      • 14 years ago

      Higher RPM = shorter lifespan and less resistance to shock etc. People sometimes kick or hit their chassis. Servers in racks don’t.

      Also NAND drives gonna takeover in the foreseable future. Mechnical drives are on the deathmarch.

      • IntelMole
      • 14 years ago

      The RAID 0 in a black box config was about a while back. It died on it’s ass after a while according to reports from StorageReview.

        • Delphis
        • 14 years ago

        Exactly. Reliability, or lack thereof. That’s the real problem with RAID-0 .. It’s really not worth halving your MTBF so you can load stuff faster.

        RAID 0+1 makes sense for the few applications that benefit from faster disk access, like video editing.

        But from Gamers? What, you can’t wait another second for Doom3 to load? Stupid.

          • Bensam123
          • 14 years ago

          I can’t explain how shaving seconds off your time will benefit a gamer if you don’t understand that already or how people don’t like waiting.

          If a array exists in a single enclosure it is no longer considered two drives. That is one drive covered by a single warranty, if it fails then it gets replaced. That is no different then having a normal drive fail.

          You wouldn’t be putting crucial files on it, thats what they make high density uber half a terabyte drives of doom for. I wasn’t talking about databases or anything of that like (thats a entirely different market). Games and windows can be re-installed.

          There is also a comparision of a engine. A car engine has how many moving parts? I don’t know but its alot more then two hard drive motors and two seek heads. They also all move around 8x faster and last longer then five years. Of course some do break but for the most part as long as you don’t put a car through hell the engine will remain in working order for quite a few years. Whats stopping a hard drive for lasting the same?

          If something like this has been made in the past perhaps it came before its time. The gaming industry has grown substantially compared to last year or the year before. Motherboards are targeting gamers along with graphics cards, processers, memory, sound cards and pretty much every single component of your system in one form or another except the hard drive.

          Gamers look for something fast and treat it like gold. Unless they have money to burn, most are extremely careful with their system.

            • Palek
            • 14 years ago


            • Buub
            • 14 years ago

            And don’t forget that an engine costs significantly more than a hard drive. Pay many thousands of dollars for a hard drive and I’m sure you could make it live for decades, too!

            • Bensam123
            • 14 years ago

            Ooops… misplaced the zeros. Either way, a engine has infinitely more moving parts and while it can be worked on, most engines (the engine itself) don’t need repairs. It needs a oil change, a air filter change, new spark plugs every few thousand miles but other then that it needs relatively little maintenance for how long it last before it needs to be rebuilt or has a catastrophic failure of some kind. Two moving parts should last five years especially when they aren’t put through the same conditions a car is.

            • Palek
            • 14 years ago

            You mean you would LIKE them to last longer. Simply saying that they should last 5 years is pretty much meaningless unless you are intimately familiar with the mechanics, materials used, degree of wear-and-tear etc. and of course the economics involved. It’s all about striking a good balance of acceptable performance and durability at a reasonable price.

            Also, while oil change may not seem that significant, without proper lubrication, ie periodically changed oil the fast-moving parts inside engines would be shredded to bits very quickly. Even my bicycle screams for grease/oil after just a few months of regular use. I realise that hard drives are not exposed to the same kind of rough handling as a combustion engine or bicycle gears, but I still think that it’s quite a feat to produce a hard drive that keeps chugging away at 7200 rpm 8 hours a day, 365 days a year, for 5 years, without needing any sort of maintenance whatsoever.

            Then of course there is also the fact that hard drive manufacturers would not like to be driven out of business by their own products that outlast the users. 🙂

            • Bensam123
            • 14 years ago

            All about the money, not what a consumer wants.^

            If you appease the consumer, you make more money even if you loose a bit in the short term. You also make patrons which are always a plus, especially if you’re the first company to do something users really like.

            I wasn’t pointing what I wrote at economics though. If you wanted to make lots of money, you wouldn’t be targeting a niche market.

            • Palek
            • 14 years ago

            I understand what you are saying, the problem is that there isn’t enough money in this particular niche market and therefore none of the hard drive manufacturers are interested. It takes substantial financial, technical and engineering resources to develop, manufacture and market a different product line, and the returns on the investment are just not there because it is a niche market without fat corporate wallets. Those resources are used elsewhere instead (SCSI, SAS etc.), where profit margins are better.

            While pleasing the consumer is of course very important, it is just a necessary step in parting them with their money. It is, after all, ultimately about the money.

            PS. Just to be clear, I would also love to see the “enthusiast” hard drives that you outlined, I just don’t see it happening.

            (Edited for clarity)

            • Bensam123
            • 14 years ago

            I dunno… The niche for the hardcore gamer seems to be big enough for motherboards, processers, memory, video cards, and sound cards to be produced specifically for it…

            • fatpipes
            • 14 years ago

            Companies like Google burn through more hard drives in a day then you may buy in your entire life. Tell me which market you would rather dedicate factory time and space?

            • Buub
            • 14 years ago

            I read your response to be something similar to “make customers happy at all costs”.

            I’m sorry, but this isn’t the most cost-effective approach. A study was done some years ago that shows the most cost-effective business model was to keep customers “just happy enough”. That way you minimize the extra expense while keeping them happy enough to come back and buy more. You lose a few, you keep a lot more.

            It goes along with a study Wendys did. They did a study that determined the highest profit seating to put in their restaurants was the most uncomfortable seating. The customer will come into the restaurant because of the food. But once there, the uncomfortable seating will get them out without excessive lingering.

            • Bensam123
            • 14 years ago

            I never was speaking from the economic standpoint, nor was I talking about giving hard drives away for free.

            High 100 dollar price range to low 200 dollar price range is alot of money for a 74 or 36 gig hard drive. I wasn’t talking about SCSI’s either if you read the last part of my post.

            What you’re describing isn’t the best market, it’s the one that has always worked in the past and is tried and true. You don’t make impressive results with what everyone else knows. I never said sacrafice there enterprise drive line, there should always be room for some out of the ordinary R&D in a company. A study doesn’t qualify as making a entirely new market nor does it suggest what could be the best given unknowns. It takes into account what is already on hand.

            What I’m suggesting would be similar to Wendys selling a hamburger that would only match some peoples tastes (which there already are) and most people wouldn’t buy but there are enough people that would buy it to justify making it.

            While there might not be a huge market as with the classic cheese burger, they would be the only place selling them and that in itself attracts patrons who just decide to eat there even if they don’t want the special burger.

            BTW that Wendys study group should do a long term study on the seating and see if people stop going there because they like to sit down and have a decent meal. It may not effect the patrons but it would deter someone that just wants a nice lunch.

            • fatpipes
            • 14 years ago

            I would check your facts before you start saying stuff like that. No, drive arrays do not equate to a single warranty. Ask anybody who has SANs and big fat NAS servers.

            If, by enclosure, you mean a home storage appliance, then you are correct. But a RAID array is still an “array of independent disks.”

      • Freon
      • 14 years ago

      So, like how do you swap out one failed harddrive in a two-drive one-piece “blackbox” array?

      The only way that’s going to work is if you make a device that works with standard drives so you can replace failed drives easily and cheaply.

      They already have these types of devices. They generally fit in PCI slots, or sometimes integrated onto the motherboard… *cough*

        • Bensam123
        • 14 years ago

        One drive, one warranty. It would no longer be seen as more then one hard drive, its not different then a single drive failing.

      • Freon
      • 14 years ago

      Over the years, the only thing that has ever seemed to really matter for consistent performance gain is spindle speed (and maybe to a lesser extent cache memory). Nothing else has come along that seems to make a definitive difference. Drives have tended to get slightly faster as time passes even without spindle speed and cache upgrades, but by no measure worth mentioning. 1-2% a year maybe? SATA was nice, but didn’t actually improve the performance of the drive itself. I tend to think it helps just to keep you from having two drives on the same cable. The move from 2MB to 8MB seemed to come with a small, but notable gain. At least you could reasonably assume an 8MB 7200rpm drive would outperform even another manufacturer’s 2MB 7200rpm without checking specific benchmarks.

      I honestly don’t know why no one else is jumping to port their 10k SCSI drives to SATA. WD has to be making a killing, because for a consumer workstation, the Raptors just completely rule the world. Well, as much as a 5-15% gain in HD performance can rule the world. Maybe without wide OEM support they aren’t doing as well as I think?

      Buy a Raptor for your system drive, then just use whatever is cheap and won’t die on you for the rest of your system. Because the Raptor matters, and nothing much else does.

        • Anomymous Gerbil
        • 14 years ago

        Are Raptors noisy (I’ve never had one or heard one)?

          • Bensam123
          • 14 years ago

          I have one and my fans in my system over-power the noise it makes.

          • Koly
          • 14 years ago

          The new ones with fluid dynamic bearing motors are very quiet at idle.. But to say they are not noisy in seek is the same as to say artillery is peaceful. AAM doesn’t help very much. Mounted in effective suspension it gets more acceptable.

      • Chrispy_
      • 14 years ago

      You know the answer: HDD companies just don’t care about niche markets, they ship millions of units a year and the small-scale stuff just doesn’t bother them in the slightest.

      I want a Microsoft Natural Keybard Elite, with the gimpy little arrowkeys and “horrific” bend in it. But I don’t want it in beige, I want it in black. How hard is it for MS to put different coloured plastic in a mould? But no, it’s never going to happen because I am just one voice.

      If you can get 100,000 buddies to email all the HDD companies to say “Yeah, we’d buy <one of these>” maybe they’d change their mind.

      • babybalrog
      • 14 years ago



        • Krogoth
        • 14 years ago

        A single Raptor is noticably faster then my two Seagate SATA 7200RPM drives. My OS feels a bit more responsive and having a dual core CPU is just icing on the cake. UT2K4, D3 and FEAR load up a couple seconds faster as well.

          • babybalrog
          • 14 years ago

          Well I can’t argue about OS responsivnes, which is mainly do to your pagefile, plater location, etc. but as far as games go you jsut proved my point ” a few seconds” that’s the differnace between what? 34 and 30 secodns? so liek a 5-6% increase where are they actuall bandwidth changed from 50-60 to around a 100 so like 40-50% increase. Not to mention some of that overhead can be from having your system on multiple disk.

          So what would a 160 GB 15K RPM 32MB RAPORT give you?
          well about 120-140 MB/s Raw transfer and a burst up near the max of the bus, assuming it’s SATA 2.5 compatable thats 300MB/s

          So sure if your doing…
          Video editing
          runnign a server

          but not for,
          gaming (maybe laod time but not FPS)

          And for the same cost you can jsut buy more RAM. So untill you have the maax of yoru syste, 4GB in most cases, and are still majorly hitting teh HDD then by all means get a RAID-0 Raptor,

            • Bensam123
            • 14 years ago

            You’re thinking too much in numbers and not enough in feel.

            I do have a 74gig Raptor and am quite sad that there aren’t any new or faster ones arriving.

            Games do load faster, that translates to grabing a vehicle first, caping a waypoint on a map before anyone else and giving your team the lead. It may seem like nothing when you look at it from the numbers standpoint but that can determine a game.

            That doesn’t even include when your game reads from your page file and kills you either which can really piss you off.

            I agree my OS does feel more responsive with a Raptor and that in itself would make me get another one for any main system I would use.

            I don’t understand how platter deformation plays into making a single solution raid 0 HD. If you basically have two hard drives upside down ontop of eachother how is it different from two seperate hard drives besides it being one unit and having one interface port?

            • babybalrog
            • 14 years ago

            I seriously dounght that your system is relaodign the maps everytime you start a mission, they will already be in RAM,

            The deformation issue happens because the actautor can’t compensate for both platers at teh same time, ie can’t move in opposite directions at the same time.

            Still say $200 on raptor or $200 in RAM, take the RAM

            • Bensam123
            • 14 years ago

            Play BF2 or practically any other FPS. BF2:SF saturates my two gigs worth of memory.

            Thats why you have two actuators. Two drives ontop of eachother.

            • babybalrog
            • 14 years ago

            which is yes called two drives, You try and go fit two actuators into a single HDD. tell me how that turn out for you.

            Try 4 gigs of RAM 😉

            • Krogoth
            • 14 years ago

            The problem is that Windows just sucks at memory mangment. It still loads content onto the HDD regradless of you having 2GBs+ of system memory. The other problem is that data from the HDD has to be first loaded onto the memory to take advantage of it’s superior speed.

            • Bensam123
            • 14 years ago

            I wouldn’t see it being that much trouble… just make the drive longer and put one on each end…

            Until you can store games on memory and have it be cheaper it isn’t worth it. Thats no different then buying a high end scsi-320 setup for your system. Totally undermines price/performance.

    • Smooth Beaver
    • 14 years ago

    How about a new Raptor with a single 160gig plater, 16mb cache, SATA II spec and the same 10rpm?

      • albundy
      • 14 years ago

      10rpm? you mean 10k rpm right?

    • Forge
    • 14 years ago

    At least in the word of mouth market, Seagate has ‘quiet’ ‘good warranty’ and ‘cool running’. Haven’t heard anyone mention ‘OMF performance’ in at least a few years.

    I’m looking to upgrade my 4*250GB in the not-too-distant future, but 160GB is the wrong way from here, and 500GB is an awfully expensive step up. Looking in the 300-400GB range, Seagate! Let’s get that 200$ 320GB drive going!

      • BobbinThreadbare
      • 14 years ago

      Have you looked at Western Digital’s RE2 drives? They are “enterprise-class” drives so they get a five year warranty, and as you can see from this review they are plenty fast.

    • blitzy
    • 14 years ago

    the performance of the 7200.9 drives doesnt seem all that great, they were even beaten by the earlier 7200 drives in quite a few of the tests not to mention the competition. Not that Seagates drives are bad performers, they just seem to be a solid average overall. The main appealing points for chosing Seagate at the moment seems to be their 5 year warranty and their relatively cheaper pricing (at least over here they seem to be cheaper).

    • dragmor
    • 14 years ago

    The 160gb drive looks nice, its has the performance and most importantly the low noise. Still I’m going to wait for SPCR to pick one up and measure the sound levels before I buy one.

      • Koly
      • 14 years ago

      I am not holding my breath, read MikeC’s (SPCR reviewer) comments on the 160GB 7200.9 in the discussion to the 500GB version review:

      §[<<]§ Don't be fooled that the 7200.9 fares better in TR's noise measurements compared to high capacity multi-platter drives. Caviar SE measured nearly the same as 160GB 7200.9 in seek, but it's default seek noise is far away from anything that could be called quiet §[<<]§ On the other hand, the single platter version looks much better in seek than the terrible 7200.7 and 7200.8 series, so it might be an OK choice for those not too sensitive to (seek) noise. EDIT: Uhm, I meant it might be an OK choice in the last sentence, not that it might /[

        • dragmor
        • 14 years ago

        hmm not great news. I’m actually temped to put a 2.5″ 7200 disk in the PC. But I havent found a quiet ones of those either.

          • Koly
          • 14 years ago

          SPCR’s review of the 7200RPM 2.5” 7200.1 isn’t very promising. I have just built a small office PC using a 5400RPM 2.5” Samsung, it is as quiet as one can hope, but it definitely is slower.

          Maybe there is hope – I have finally found a suspension which is not made on a knee using laces, works wonderfully, fits in in a 5.25” bay, doesn’t hinder cooling too much and is cheap enough:

          §[<<]§ I have bought one as a test sample and ordered more. It makes wonders with a Raptor, my new 200GB Samsung P120 is almost on the noise level of a 2.5'' drive with it and even my older WD80JB's typical whine gets damped down. I am going to try it out on my all time seek noise champion, a Seagate SATA 7200.7, soon, I am curious how it will deal with it.

    • Xylker
    • 14 years ago

    Meh… I think I am so spoiled by the advances in other places in the PC, that the pace of improvement for HDDs is just too slow. I guess I could get in line for a couple of the 2x 160GB platter + 16 MB of cache drives Geoff is advocating, but it would have to be ~200 dollars or less.

    I would REALLY like to see a better RAID controller for the personal/workstation level type systems. A consistent 100MB/s out of a 3 drive array would be a nice place to start.

      • FireGryphon
      • 14 years ago

      You’ll be instantly pleased by progression in the hdd sector if you consider for a moment the progression in the sound card sector.

        • Chrispy_
        • 14 years ago

        Hehe, indeed.

        SBLive! Value from 1996/7 (whenever they came out):
        It’s crap, and it works fine for everthing I do so I love it and won’t change it until it blows up.

Pin It on Pinterest

Share This