Review of First 10K IDE Drive

Sivar writes "StorageReview has a review of the first 10,000 RPM IDE hard drive. Despite the speed that other technologies are improving, this is the first rotational speed increase in almost six years for standard IDE drives." The review is pretty thorough, but also warns to keep in mind that the reviewed unit is only beta hardware.

Comments

Finally...

[Anonymous Coward]

Nice to know they are finally starting to speed up the slowest part of the computer again.

Re:Finally...

[Twirlip of the Mists]

Nice to know they are finally starting to speed up the slowest part of the computer again.

You mean the user?

Re:Finally...

[Jeff DeMaagd]

Maybe a tie between the user and the floppy drive.

But seriously, ATA hard drives have still been increasing in speed even when "stuck" at 7200 RPM because the data density skyrocketed.

Re:Finally...

[sweetooth]

While the user may be the slowest part I will be damn happy when the current incarnation of the PCI bus goes away.

Re:Finally...

[ehiris]

Nice to know they are finally starting to speed up the slowest part of the computer again.

I wish I'd have network card speed close to the speed of my hard drive.

Re:Finally...

[wideBlueSkies]

You mean Windows?

Re:Finally...

[Sivar]

Yeah, but how long till it fries itself?

I'd rather have something slow that I can trust, rather than something that goes out in a brillant ball of fire--even though it was really fast.

The reviewed drive has a 5 year warranty. How long is the warranty on your slower drive?

The Seagate Cheetah X15.3 is the world's fastest hard drive (until the Maxtor Atlas 15K is released). It is one of the most reliable drives you can buy, with an extremely high rating in StorageReview.com's reliability survey, and an excellent history in IBM, Dell, etc's enterprise servers.

"Slower is more reliable" doesn't hold water anymore, though it is true that early 7200 RPM IDE drives were less reliable than the slower 5400 RPM drives.

Re:Finally...

[cbreaker]

I've never had a problem with PCI. Sure, it's becoming outdated in terms of speed, but overall PCI has served the PC industry very well.

The move from ISA to PCI as the general PC slot was a very good step forward. Gone were the hair-pulling configuration issues, jumper settings, and "ISA Plug'n'Play" that sometimes worked.

The next "PCI" for the PC will most likely be something like 3GIO, which was recently renamed to "PCI Express." It's a new bus, but it's software-compatible with PCI. Since PCI Express is a new hardware interface (new slots) it's not just for compatibility; it's because PCI works and there's no reason to change what you don't need to change.

At any rate, this topic is IDE drives. 10K SCSI drives tend to be pretty loud and run quite hot. I think that the 10k IDE drives will probably imploy some sort of technologies to keep them quiet and cooler, since IDE drives generally live on the desktop.

Re:Finally...

[JeanBaptiste]

you must be new. How happy we were to use PCI over ISA and/or EISA and/or MCA. But maybe Im just old school. Remember the 5MB seagate hard drive? It is now a doorstop. I still have some 8 inch floppies. My first job involved loading the tape to tape reels.

and you complain about PCI? kids these days...

Re:Finally...

[Anonymous Coward]

I've found you can significantly speed up users by replacing them with scripts.

Re:Finally...

[Penguin Follower]

0K SCSI drives tend to be pretty loud and run quite hot. I think that the 10k IDE drives will probably imploy some sort of technologies to keep them quiet and cooler, since IDE drives generally live on the desktop.

While that is generally true, I have a late model 10k rpm Cheetah in my file server that is quieter than my Maxtor IDE drives in this desktop machine. My other cheetah however, is an early 10K rpm scsi that is VERY noisy.

Re:Finally...

[Trolling4Dollars]

I have a 0K SCSI drive. I'm not too impressed with it.

Re:Finally...

[Sivar]

This has been brougt up numerous times in the StorageReview.com forums. There are several companies making solid-state drives, including PCI-SDRAM drives with battery backup, and FLASH drives which need no such backup.
They're great I'm sure, but the price per GB is sky high.
Take a look at the cost/GB of Compact Flash, multiply that by "server part for server budget" marketing. You get the idea.
I'd love to provide links, but I can't remember any of the manufacturer's names, and Google doesn't bring them up. You might try a quick search in SR's forums once the server recovers for "solid state" if you're interested though--I seem to remember a 4GB SCSI FLASH drive being almost within the realm of the average geek's budget.

Re:Finally...

[Anonymous Coward]

But it's pretty damn quiet.

Re:Finally...

[matguy]

You want loud, I used to have 4 5.25" full height (that's twice as tall as a standard cd-rom drive) SCSI drives running in an old server. The drives were 600mb each and consumed 23watts each. I had to stagger out the spin up times so they didn't all spin up simultaniously and overload the power supply.

I took apart the 600mb drives for the big voice coil magnets that are strong enough to be very hard to remove from metal surfaces. They make decent floppy erasers.

I still have an old 5.25" full height, 2 gig SCSI drive hanging around, but not being used as well as a 3.5" double height (double the height of a 3.5" floppy, which might be "full height," but I'm not sure) 2 gig SCSI drive, but it's not loud and I'm not sure what speed it is. I kind of have a thing for old hardware that might be somewhat usefull sometime...

Re:Finally...

[pmz]

10K SCSI drives tend to be pretty loud and run quite hot.

Not any more. The latest 10K SCSI drives purr like a sleeping kitten (aww...). I just bought a Ultra160 Seagate Cheetah, and I can only hear it by putting my ear up to the computer's case. The fans are way louder.

Stand back and watch for now..

[SystematicPsycho]

Until new drives seem reliable and we don't hear of any issues with them there is nothing wrong with what I've currentlty got. Hardware also is hideously expensive when it first hits the shelves.

Re:Stand back and watch for now..

[grimt007]

Luckily new tech unveilings have a wonderful habit of driving current prices down, maybe we'll see 7200 RPM's at consistently less than $1 a gig!

then look out cause RAID here i come.

Re:Stand back and watch for now..

[Sivar]

Hardware also is hideously expensive when it first hits the shelves.
Of course you are correct, but this drive is expected to be priced at $160USD, which isn't really all that bad, all things considered.

As far as reliability, the WD Raptor is targeted at servers and has a 5-yr warranty. Western Digital has experience designing SCSI drives, and I suspect that the Raptor is essentially a 10,000RPM SCSI assembly with a serial ATA--instead of SCSI--interface (as well as a few other tweaks). Certainly the mechanical characteristics appear to strongly resemble common 10,000RPM SCSI drives, such as the sub-6 millisecond access times.

Re:Stand back and watch for now..

[Sivar]

There's no margin on a $60.00 drive. It seems to be that way, since once drives hit about the $75 mark they tend to be phased out.
I find it extremely impressive that they can get that cheap at all.
MaxtorSCSI, a SCSI engineer at Maxtor (funny, that), and a forum user on StorageReview.com, stated once that hard drives are the highest precision mechanical devices, by far, in the average person's home.
The platter has to be so flat that, spinning at thousands of RPM, the heads must float above the platter at less than 1/50 the width of a human hair, or slightly more distance than the width of an average smoke particle. And they have to survive being bumped, because if those heads touch the platters, all hell (and the heads) breaks loose.

Re:Stand back and watch for now..

[Sivar]

My main concern is its actual capabilities when being used to store/delete/etc large, numerous files and how long until the hard drive finally crashes and dies out.

The Raptor has a 5-yr warranty (5 times as long as most desktop hard drives) and is targetted for the server market. Unless WD seriously screwed up, I am willing to be that it is about as reliable as other enterprise 10K drives (all of which are SCSI)--that is to say, incredibly reliable.

A 10k IDE drive is bound to have a ton of hard drive space

Actually, the faster the platter spins, the lower density each platter must be in order for the heads to keep up. For example, the Western Digital Raptor is a 36GB drive with a single 36GB platter (that's 18GB/side). This is the same size of platter as on the largest of 10KRPM SCSI drives.
To contrast, the largest platter size on a 7200RPM drive is 80GB/platter (or 40GB/side), and Weste3rn Digital is about to release a 250GB drive which will have three 83.3GB platters.

Higher platter density improves speed as well, but generally speaking (VERY generally speaking), increasing rotational speed improves drive performance more than having a somewhat higher density platter. Those of course varies based on what you are doing with the drive, whether it involves lots of random accesses (mail/webserver) or lots of linear accesses (video editing) or something in between.

In general, expect higher RPM drives to trail behind lower RPM drives in platter density, and therefore in maximum available disk space.

ObPrediction

[commodoresloat]

7200 RPM should be fast enough for anybody.

Re:ObPrediction

[Sivar]

Virtually memory was a short term hack when memory was critically limited: Nowadays you could literally disable it in most configurations.

Actually, virtual memory is still used today and has nothing to do with hard drives, though this is widely believed. Virtual memory is the ability for an operating system to tell all programs that they can address the full addressable range of the processor, that is, with 32-bit CPUs that each program has access to 4GB of RAM. It happens that many operating systems use hard disk space as a substitute for RAM when there isn't enough physical memory, but the use of hard drive space as memory is not the definition of virtual memory.
I was under the same impression myself not too long ago. ;)

Re:Can they produce these with a serial ATA interf

[cheezedawg]

They sure can- and they do. I have been playing around with a 10k RPM SATA drive from Western Digital at work this week.

About your other question- there are a lot of factors that contribute to drive performance, but rotational speed is one of the biggest.

It would be nice

[T5]

if the manufacturers of these 10K SATA drives would offer two different sets of firmware - one optimized for locality access for desktops and another for the more scatter/gather usage patterns seen on servers. How WD et.al. will position this drive for production remains to be seen.

Re:It would be nice

[Fulcrum of Evil]

two different sets of firmware - one optimized for locality access for desktops and another for the more scatter/gather usage patterns seen on servers.

How about making it configurable with a jumper or a utility. They already do this for a speed/noise tradeoff.

Re:Can they produce these with a serial ATA interf

[CanSpice]

You mean exactly like the one they reviewed?

Did you even read the article?

Big deal.

[jericho4.0]

So what? An increase in heat and wear and tear on components, for what theroy says is ~25% speed increase. This drive doesn't even come close to that.
I would think that for most apps that need this, a SCSI or RAID (or both) solution would be better.

Oh well, faster is pretty marketable, I guess.

Re:Big deal.

[Sokie]

Well this WD drive does sport a 1.2 million hour MTBF and 5 year warantee. It's pretty much built with reliability in mind since they are targetting entry- and mid-level servers.

-Sokie

Re:Big deal.

[smallpaul]

So what? An increase in heat and wear and tear on components, for what theroy says is ~25% speed increase. This drive doesn't even come close to that. I would think that for most apps that need this, a SCSI or RAID (or both) solution would be better.

Why would SCSI be less prone to heat and wear than IDE?

Re:Big deal.

[dhovis]

hy would SCSI be less prone to heat and wear than IDE?

I think the point was just that SCSI provides better performance, even with 7200RPM. Much of that comes from the fact that SCSI drives are "smart" and require almost no CPU time, whereas IDE drives are "dumb", and require the CPU to handle much of the work.

The price differential, OTOH, is substantial.

Re:Big deal.

[silverhalide]

Since SCSI drives are obviously not targetted at the commodity/home market, manufacturers spend a few extra bucks on the mechanical parts, including fluid bearings, better motors, and more testing time to build a drive more suited for a server market. With a better mechanical build, less heat is generated and higher MTBF figures are posted. Mechanically, there is little required difference between a SCSI and IDE drive, but electronically and in the market, there is. At the same time, these cooler, better drives do cost more for a reason!

Re:Big deal.

[steveha]

IDE drives are "dumb", and require the CPU to handle much of the work.

No longer really true. Ever since UltraDMA/33 mode, the CPU has not had much work to do with an IDE drive. SCSI drives still have a few tricks such as tagged queuing, but those features have been filtering down to IDE drives as well.

SCSI drives intended for servers cost more, and generally are better made, than IDE drives. They also come with much longer warranties (makes sense since they are made better).

steveha

Re:Big deal.

[Sokie]

Hey man, I don't make the news, I just report it.

Perhaps a little googling would have enlightened you as to what exactly an MTBF is. It's not quite as simple as it sounds:

(Thus spoketh the web page:)

It is generally accepted among reliability specialists (and you, therefore, must not question it) that a thing's failure rate isn't constant, but generally goes through three phases over a thing's lifetime. In the first phase the failure rate is relatively high, but decreases over time -- this is called the "infant mortality" phase (sensitive guys these reliability specialists). In the second phase the failure rate is low and essentially constant -- this is (imaginatively) called the "constant failure rate" phase. In the third phase the failure rate begins increasing again, often quite rapidly, -- this is called the "wearout" phase. The reliability specialists noticed that when plotted as a function of time the failure rate resembled a familiar bathroom appliance -- but they called it a "bathtub" curve anyway. The units of failure rate are failures per unit of "thing-time"; e.g. failures per machine-hour or failures per system-year.

What, you may ask, does all this have to do with MTBF? MTBF is the inverse of the failure rate in the constant failure rate phase. Nothing more and nothing less. The units of MTBF are (or, should be) units of "thing-time" pre failure; e.g. machine-hours per failure or system-years per failure but the "thing" part and the "per failure" part are almost always omitted to enhance the mystique and confusion and to make MTBF appear to have the units of "time" which it doesn't. We will bow to the convention of speaking of MTBF in hours or years -- but we all know what we really mean.

What does MTBF have to do with lifetime? Nothing at all! It is not at all unusual for things to have MTBF's which significantly exceed their lifetime as defined by wearout -- in fact, you know many such things. A "thirty-something" American (well within his constant failure rate phase) has a failure (death) rate of about 1.1 deaths per 1000 person-years and, therefore, has an MTBF of 900 years (of course its really 900 person-years per death). Even the best ones, however, wear out long before that.

This example points out one other important characteristic of MTBF -- it is an ensemble characteristic which applies to populations (i.e. "lots") of things; not a sample characteristic which applies to one specific thing. In the good old days when failure rates were relatively high (and, therefore, MTBF relatively low) this characteristic of MTBF was a curiosity which created lively (?) debate at conventions of reliability specialists (them) but otherwise didn't unduly bother right-thinking people (us). Things, however, have changed. For many systems of interest today the required failure rates are so low that the MTBF substantially exceeds the lifetime (obviously nature had this right a long time ago). In these cases MTBF's are not only "not necessarily" sample characteristics, but are "necessarily not" sample characteristics. In the terms of the reliability cognoscenti, failure processes are not ergodic (i.e. you can't blithely trade population statistics for time statistics). The key implication of this essential characteristic of MTBF is that it can only be determined from populations and it should only be applied to populations.

MTBF is, therefore an excellent characteristic for determining how many spare hard drives are needed to support 1000 PC's, but a poor characteristic for guiding you on when you should change your hard drive to avoid a crash.

(An excerpt from this page.)

-Sokie

10K hard drive?!

[EverStoned]

Did we go back in time to 1975?!

Does that really help?

[Anonymous Coward]

But hasn't there been several articles around that show hard drive RPM to be a minimal factor in the performance of HDDs?

5400 -> 7200 wasn't that advantagous, but will 7200 -> 10000 be that much better?

Don't we get better performance improvements from tweaks to the file system and how it writes and spaces out its blocks and cylinders?? Or are we at those limits already?

An important paragraph...

[La Temperanza]

When Western Digital raised the bar nearly 1.5 years ago, we repeatedly pointed out that the Special Edition (JB series) Caviar was what readers really wanted when they speculated over 10,000 RPM ATA drives.

Equipped with an 8-megabyte buffer and accompanying firmware aggressively tuned for single-user scenarios, the WD1000JB easily matched and even exceeded the performance that the best 10k RPM SCSI drives of the era delivered when it came to desktop performance.

While SCSI drives feature superior mechanics, their server orientation forces them to trade away firmware optimized for highly-localized patterns in favor of strategies that maximize returns in random access scenarios. In the Raptor, WD faces much of the same quandary.

Re:An important paragraph...

[edmudama]

"While SCSI drives feature superior mechanics, their server orientation forces them to trade away firmware optimized for highly-localized patterns in favor of strategies that maximize returns in random access scenarios. In the Raptor, WD faces much of the same quandary."

There is no cache optimization for random access scenario, since you're guaranteed to almost never get a read cache hit.

Maximizing random performance = mechanics.
Maximizing local performance = scheduling.

Reliability is more important to me

[Compact Dick]

Two of my friends purchased Seagate's 40GB 7200RPM Barracuda drives. In the space of eight weeks, both began sprouting bad sectors all over the place. This is totally unacceptable, especially when you consider that the standard HDD warranty is now 1 year [from 3.]

Focus on improving reliability, not increasing rotation speeds. Or just bring on those cool holographic drives - that should fix things up :-)

Cheers,
CD

Re:Reliability is more important to me

[Anonymous Freak]

Um... Do you have any empirical evidence for your claim that WD isn't reliable? Anything other than years-old anecdotes?

Check out StorageReview's reliability database, and you'll see that WD drives are just as (in some cases more) reliable than those from Maxtor and others. (About the only drive company that has had reliability problems recently is IBM, who has now gotten out of that market entirely.)

Re:Reliability is more important to me

[BrookHarty]

Ive seen almost every HD, Motherboard or hardware go bad. Other than IBM and the bad batch of drives awhile ago, most seem pretty good. Just save your money and buy the cheapest oem you can get. (IMHO).

BTW, I waw a good deal on pricewatch, 200gig 7200RPM 8M WD's for 240 at newegg.

Ahem... not true.

[Geminus]

I have a Maxtor 14 GB IDE-HD that is 10,000 RPM. Sure it sounds like a jet engine... but I've had it almost three years now, with no problems.

Re:Ahem... not true.

[Anonymous Freak]

Maxtor IDE 10k RPM??? What is it's model number? I don't think ANYONE had heard of any 10k RPM ATA drive before this one.

I know of plenty of 10k RPM SCSI drives, but not ATA. And, yes, the early 10k SCSI drives screamed like a jet...

almost slashdotted... (non karma whore post)

[Anonymous Coward]

March 5, 2003 Author: Eugene Ra

Western Digital Raptor Available Capacities
Model Number

Capacity
WD360GD

36 GB

Estimated Price: $160 (36 GB)
Manufacturer Specifications
Beta unit provided by Hypermicro.com
Remember, mention StorageReview in your HyperMicro.com order and receive free UPS ground shipping!

Introduction

StorageReview.com readers have been speculating for the better part of three years on when the industry would ratchet up the spindle speed of ATA hard drives. When would it happen? Which company would start the trend? Speculation finally gave way to a real announcement on February 10th when Western Digital officially announced its Raptor Serial ATA drive.

Western Digital is in many ways the perfect company to lead ATA to a next-generation spindle speed. Ever since it introduced the Caviar WD400BB, WD has consistently led the field when it came to ATA performance. That's a 2.5-year run at the top- very impressive in the competitive computer hardware field. More importantly, however, the firm has no SCSI business to protect. The last thing that established SCSI powerhouses such as Seagate, IBM, and Maxtor want to see is the erosion of the relatively cushy margins associated with SCSI drives. Now that WD has opened this veritable Pandora's Box, the competition is sure to follow.

According to WD, the key factor holding back higher spindle speeds was parallel ATA's lack of specification-level hot swap functionality. To be successful (initially, at least), any 10k RPM ATA drive must gun for the enterprise market. And in the enterprise, a sector that views outages as unacceptable, the ability to swap out a failed drive for another unit with minimal downtime is crucial. Serial ATA provides for such hot-swap functionality. Now that SATA is trickling into the channel, WD believes 10k RPM ATA's time has come.

The Raptor comes in just a single configuration- a single 36-gigabyte platter. WD specifies the drive's seek time at just 5.2 milliseconds, solidly within SCSI territory. An 8-megabyte buffer accompanies the drive. Some folks may be disappointed with the drive's relatively paltry capacity- after all, today's SCSI drives deliver 147 GB of storage in a low-profile chassis. Much like its namesake made popular by 1993's Jurassic Park, however, WD envisions Raptors in multiple-drive configurations running off of relatively inexpensive SATA RAID controllers. Reflecting its enterprise orientation, the Raptor claims a 1.2 million hour MTBF spec and features a five-year warranty.

It is important to note that the market for the Raptor is primarily the entry- and mid-level server markets and not the enthusiast desktop sector. When Western Digital raised the bar nearly 1.5 years ago, we repeatedly pointed out that the Special Edition (JB series) Caviar was what readers really wanted when they speculated over 10,000 RPM ATA drives. Equipped with an 8-megabyte buffer and accompanying firmware aggressively tuned for single-user scenarios, the WD1000JB easily matched and even exceeded the performance that the best 10k RPM SCSI drives of the era delivered when it came to desktop performance. While SCSI drives feature superior mechanics, their server orientation forces them to trade away firmware optimized for highly-localized patterns in favor of strategies that maximize returns in random access scenarios. In the Raptor, WD faces much of the same quandary. With its enterprise-class warranty and seek time, however, its clear that server performance is WD's first priority for the Raptor.

The drive tested for this review is a beta unit provided by longtime SR sponsor HyperMicro.com rather than Western Digital itself. With a handful of exceptions, SR generally has not published performance figures for products this early in the development cycle. Please remember the final Raptor product may deliver results substantially different from those that follow.

Keeping that in mind, let's see what kind of performance this beta sample delivers!

ow-Level Results

IPEAK SPT's AnalyzeDisk assesses many low-level characteristics of hard drives. Two tests, Read Service Time and Write Service Time, each respectively conduct 25,000 random single-sector reads and writes across the entire breadth of the drive. The result is perfectly equivalent to an access time test. Results come both as an average and as a graphic that plots the percentage of accesses across the amount of time they each take to complete. For more information, please click here.

Note: Scores on top are better.
Service Time Graphs (in milliseconds)
Average Read Service Time
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 7.6 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 8.0 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 8.7 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 12.9 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 13.8 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 14.8 |
|
WD360GD (BETA) Average Read Service Time

The beta Raptor delivers a measured average access time of 8.7 milliseconds. Subtracting 3.0 ms to account for the rotational latency of a 10k RPM spindle speed yields a measured seek time of 5.7 ms. While excellent for an ATA drive, the score is a bit off of the manufacturer's 5.2 ms claim as well as a bit higher than what we've come to expect from 10k SCSI drives.

The use of an external controller (the Promise SATA150 TX4) and its associated driver unfortunately makes it more difficult to consistently disable write caching which unfortunately precludes us from presenting average write access times.

eTesting Lab's WinBench 99 v2.0 features a test that measures a drive's read sequential transfer rates across the entire drive. The benchmark reports results both in quantitative numbers as well as in a graphic that plots the transfer rate across the capacity of the drive.

Note: Scores on top are better.
Transfer Rate Graphs (in megabytes per second)
Transfer Rate - Begin
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 70.9 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 69.0 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 59.2 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 57.6 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 56.5 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 56.2 |
|

Transfer Rate - End
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 44.1 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 40.4 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 37.6 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 33.7 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 32.8 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 30.7 |
|
WD360GD (BETA) Transfer Rate

Despite its higher spindle speed, the Raptor's outer-zone transfer rates aren't much better than today's top 7200 RPM units. Its score of 57.6 MB/sec narrowly beats the Caviar WD2000JB and slightly trails the DiamondMax Plus 9. Thanks to its smaller platter diameter, the Raptor exhibits a bit less decay as it moves towards its inner zones. Its minimum score of 37.6 MB/sec tops other ATA drives yet still fails to reach the levels of a Cheetah or Atlas.

Desktop Performance...

Formulated utilizing IPEAK SPT's WinTrace32 and RankDisk, the StorageReview.com Desktop DriveMarks exactingly reproduce pre-recorded, contemporary access patterns on tested hard drives. For more information, please click here.

Note: Scores on top are better.
Desktop Performance Graphs (in I/Os per second)
SR Office DriveMark 2002
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 503 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 450 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 431 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 418 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 418 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 391 |
|

SR High-End DriveMark 2002
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 444 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 427 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 415 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 388 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 382 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 300 |
|
SR Bootup DriveMark 2002
Western Digital Raptor WD360GD BETA (36 GB SATA) - 455 |
|
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 422 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 391 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 386 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 348 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 307 |
|

SR Gaming DriveMark 2002
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 649 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 548 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 546 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 531 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 528 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 508 |
|

The beta Raptor turns in a StorageReview.com Office DriveMark 2002 of 418 I/Os per second. While such a score places it among the top ATA drives, the Raptor fails to match WD's own Caviar WD2000JB. A top-level 10k SCSI unit such as Maxtor's Atlas 10k IV substantially outpaces the Raptor.

Differences become more glaring in the High-End DriveMark. At just 300 I/Os per second, the Raptor places in the middle of a pack of 7200 RPM drives equipped with 2-megabyte buffers. Here the WD2000JB outscores the WD360GD by a substantial 43% margin.

The Windows XP bootup process recorded in the SR Bootup DriveMark 2002 features an unusually high average queue depth for a desktop scenario. In this test, the Raptor stretches its legs, easily besting all comparable ATA and SCSI disks.

Finally, in the SR Gaming DriveMark 2002, the Raptor delivers 531 I/Os per second, a figure comparable to a top-end ATA drive yet trailing the Atlas 10k IV by a significant margin.

To be fair, we should point out that the 36-gigabyte Raptor faces flagship drives of much greater capacity in our tests. The margins between the Raptor and smaller ATA or SCSI drives would likely not be as pronounced since the competition would then be forced to work across a greater percentage of its platter zones.

Server Performance...

Server Performance

The StorageReview.com Server DriveMarks consist of IOMeter trials using predefined patterns supplied by Intel across varying load depths. The reported scores represent a normalized average of results from 1 to 64 outstanding IO/s. For more information click here.

Note: Scores on top are better.
Server Performance Graphs (in I/Os per second)
SR File Server DriveMark 2002
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 271 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 258 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 177 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 131 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 129 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 116 |
|

SR Web Server DriveMark 2002
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 261 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 255 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 181 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 134 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 119 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 115 |
|

In the SCSI-stronghold of random, server-oriented performance, the Raptor, while delivering scores significantly better than traditional ATA drives, nonetheless falls behind contemporary SCSI drives by a significant margin. Even older drives such as the Seagate Cheetah 36ES (not represented; see the performance database to create custom comparisons) unquestionably trounce the WD360GD. The beta Raptor delivers the server performance that one would expect from a good 7200 RPM SCSI drive- definitely a cut above standard ATA, but not up to 10k RPM levels.

Legacy Performance

eTesting Lab's WinBench 99 Disk WinMark tests are benchmarks that attempt to measure desktop performance through a rather dated recording of high-level applications. Despite their age, the Disk WinMarks are somewhat of an industry standard. The following results serve only as a reference; SR does not factor them into final judgments.

Note: Scores on top are better.
Legacy Performance Graphs (in megabytes per second)
ZD Business Disk WinMark 99
Western Digital Caviar WD2000JB (200 GB ATA-100) - 16.4 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 16.1 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 15.9 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 15.7 |
|
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 12.1 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 11.7 |
|

ZD High-End Disk WinMark 99
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 44.9 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 39.2 |
|
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 38.0 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 36.9 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 33.3 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 25.0 |
|

Heat and Noise...

A Fluke thermometer and an Extech Type II SPL meter respectively deliver objective operating temperature and sound pressure measurements. Note that objective noise measurements are gathered only after subjective impressions have been penned. For more information, please click here.

Note: Scores on top are better.
Heat and Noise
Idle Noise (in dB/A @ 18mm)
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 40.1 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 40.4 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 41.0 |
|
Western Digital Caviar WD2000JB (200 GB ATA-100) - 45.5 |
|
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 47.7 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 48.5 |
|

Net Drive Temperature (in degrees celsius)
Western Digital Caviar WD2000JB (200 GB ATA-100) - 19.7 |
|
Maxtor DiamondMax Plus 9 [8MB, 80GB/plat] (160 GB ATA-133) - 19.7 |
|
Western Digital Raptor WD360GD BETA (36 GB SATA) - 20.6 |
|
IBM Deskstar 180GXP 8 MB (180 GB ATA-100) - 22.1 |
|
Seagate Cheetah 10K.6 (146 GB Ultra320 SCSI) - 24.4 |
|
Maxtor Atlas 10k IV (147 GB Ultra320 SCSI) - 30.0 |
|

Objectively speaking, the beta Raptor turns in impressively low noise floors, likely due to its single-platter design. A score of 40.4 approaches the noise floor delivered by the latest Barracuda ATA drives. Subjectively speaking, however, our sample emits an irritating high-pitched squeal reminiscent of early 10k RPM SCSI disks. The whine was audible even over the testbed's relatively loud drive cooler fans.

Seek noises land somewhere between today's louder ATA disks and a typical reviewed SCSI unit. While the Raptor features random seeks similar to that of 10k RPM SCSI, it features just a single platter contrasted with the four typically found in today's flagship units. The resultant actuator noise is quite unobtrusive.

The Raptor's single-platter configuration also yields a relatively low operating drive temperature. Our measurements reached 20.6 degrees Celsius above ambient room temperature- on the high side for an ATA drive but well below the typical SCSI disk.

Conclusion...

It's very difficult to draw firm conclusions on a drive that is obviously far from its final state. Firms manufacture pre-release units not for performance demonstrations but rather for system-integration purposes- resellers need to qualify the unit in their systems for extended periods of time before the drive hits general availability.

Many readers may be disappointed with the Raptor's relatively lackluster desktop performance. For various reasons, enthusiasts view an increased spindle speed as the largest factor in single-user performance. The reality, however, is that desktop usage predominately consists of highly-localized patterns and is affected more by caching strategies than marginal mechanical improvements. Western Digital's JB series may very well continue to stand as the premiere choice for those seeking the ultimate in single-user speed.

We're more concerned with the Raptor's server performance. While it is definitely a step above standard 7200 RPM ATA drives, the beta Raptor trails today's 10k RPM SCSI drives by substantial margins. If WD and SATA are to have a chance at cracking the enterprise market, the Raptor's multi-user performance must approach the levels delivered by Cheetahs and Atlases.

Again, all figures, analyses, and conclusions have been drawn from an early pre-production sample. It is likely that the performance delivered by the final product will differ significantly from what we've seen today. We wish WD the best, and eagerly await the opportunity to officially put the Raptor through its paces.

Re:Can they produce these with a serial ATA interf

[BrookHarty]

If they can make these beauties with a serial-ATA interface, I AM SOLD!

[snip]
Speculation finally gave way to a real announcement on February 10th when Western Digital officially announced its Raptor Serial ATA drive.
[/snip]

Did I miss something, the article says its SATA.

Looks like

[Rooked_One]

even 10,000 rotations per minute isn't enough to keep up with the /. effect.

Re:What to be more shocked about

[TeknoHog]

Don't worry, soon CmdrTypo will repost this to confirm that the news is in fact about the 10k rpm drive.

Things To Keep In Mind

[MBCook]

There are a few things to keep in mind about these numbers. Most of them are mentioned in the article, but they're scattered around. Just think about these things:

• Seperate Card - Remember that the SATA controller is on a seperate card, it's not integrated into the chipset. So these number could (and probably will) change for the better when we see SATA built into the southbridge later this year (was it Grandale from Intel that will do this? I'm too lazy to look it up).
• Drive Size - The drive in the review is up to 1/6th the size of some of the other drives in the review. So if you're comparing this drive you have to remember that it would perform better if it was a 160 gig drive and didn't have to work all over it's platter.
• SATA - All the other drives in this review are either ATA or SCSI. So as SATA goes, this drive might be king of the hill by far.

Those said, I have a few other things I'd like to say. First of all, it's nice to see that the drive is quiet. Even many 5400 and 7200 RPM drives are quite loud today. It's nice to know that going to 10k isn't going to turn my PC into a jet engine. Also, they mention that the reason that we haven't seen 10k IDE drives before was that servers didn't want them since they couldn't be hotswapped like SCSI. SATA supports hotswap in theory, but can you hotswap today? I don't think Windows lets you, IIRC (or if it does the system is a bit unstable afterwards). Does Linux let you hotswap SATA drives? If all the drives are one one controller (say RAID 5, or something else redundant) and you swap a drive, does the OS even know it happened? I don't have any expirence with hotswapping hard drives.

Re:Things To Keep In Mind

[Anonymous Coward]

Uhhh, my motherboard already has integreated Serial ATA. ASUS A7N8X, nForce2 based. Also does Serial-ATA raid. And the reason they say it's quiet is likely due to the fact it only has one platter (as opposed to 2 or 3 in many of the others).

Re:Things To Keep In Mind

[DeathB]

That's one reason why you haven't seen them in servers but there are others.

• Testing - The test patterns a server drive and a desktop IDE drive go through are very different. IDE drives aren't run for more than a single pass or two, while SCSI drives can be subjected to a day or more of continuous testing
  
• Firmware - SCSI firmware is made to do a better job at trying to save your data. In most cases it's 4-5x loc of IDE firmware. Much of this is different techniques of reading a block which contains an error.
  
• Queueing - Now you can argue that SATA also brings this to the ATA platform, but it's only sort of true. SCSI drives get advantages from queueing for two reasons, greater queue depth leads to shorter overall write time, and the ability to reorder queues. If a drive can write the closest block to the head, first, it is going to perform better. Now it would be possible to do this on a SATA drive, but not at IDE costs. One of the biggest differences in chipsets is integration, usually SCSI drives will offload things like servo control to a separate processor. Unfortunatly on IDE, its one processor is at 80-90% load trying just to do servo control. It doesn't have time to reorder your queue. (This is of course fixable, but people would have to want to pay for the extra processor power)
  
• Rotational error - SCSI drives are designed to handle the types of errors which come from putting several drives in the same case. That is each drive tries to transfer some of its rotational momentum into the case. The intersection of these forces is a case which torques in the direction of drive spin. In IDE drives this can cause drastic reductions in throughput as more and more retries are necessary. (There are some IDE RAID cases good enough to fix this completely, but most only do partially). I've actually seen conflicting research on this last pont, but only in how good the case has to be to prevent these effects.
  

I do have some experience hotswapping drives. Linux sort of handles it. echo `scsi remove-single-device 0 0 1 0` > /proc/scsi/scsi and then
echo `scsi add-single-device 0 0 1 0` > /proc/scsi/scsi will in theory hot swap target 1. However, I've had about a 50-75% success rate with not crashing the machine doing it. Hotswappable IDE is even possible, but your controller has to support it.

Adam

WTF?

[Jeffrey Baker]

Great! Now I can get a second-rate, first generation 10KRPM hard drive with bad server performance and almost no capacity, from a company that disavowed the high end years ago by bailing on the SCSI market, all for the same price as established SCSI drives of the same size or established ATA drives four times the size.

Hrmm.

Can they compete at that price?

[nolife]

Estimated Price: $160 (36 GB)
Manufacturer Specifications
Beta unit provided by Hypermicro.com
Remember, mention StorageReview in your HyperMicro.com order and receive free UPS ground shipping!

Tiger Direct has 36GB Ultra160 SCSI's for only $99. Anyone know if these are some type of rejects? Google did not reveal any obvious issues with this model.

Re:Can they compete at that price?

[Sivar]

IBM SCSI drives are generally the slowest, loudest, least reliable, hottest SCSI drives on the market. That said, $99 is dirt cheap, but you'd be happier with a fast IDE drive or a real SCSI drive, like a Maxtor Atlas 10K-3 (which are also quite inexpensive)
ALso, take a look at ResellerRatings.com for TigerDirect. I wouldn't order from them...

For SCSI drives, I have found HyperMicro generally has the best prices among companies that are trustworthy, and that them, Newegg, Mwave, or Dell have the best IDE prices.

Hate to break it to all of you...

[Xandar01]

This is pretty much a dupe. Even more amusing, note who posted the first article.

http://slashdot.org/article.pl?sid=03/02/21/055324 9&tid=137

Enterprise-class ATA Drives Posted by CowboyNeal on Friday February 21, @05:48AM

from the fast-enough-to-make-disk-heads-spin dept.

Re:Hate to break it to all of you...

[Sivar]

This is pretty much a dupe.

No. It isn't a dupe. The new /. article has a link to an actual review of the drive, not just an announcement that it happens to exist. Analog: If a story mentions that NASA has contracted for a new space vehicle, and later there is another story that covers the actual performance, mechanics, and statistics of a completed model, those two stories would be related, but not dupes.

I'm waiting for..

[euxneks]

Holographic data storage!!!

Yes, in the future, we will all have quantum computers with holographic data storage devices, communicating to us through 3d monitors!

Fast but Noisy

[zeekiorage]

Objectively speaking, the beta Raptor turns in impressively low noise floors, likely due to its single-platter design. A score of 40.4 approaches the noise floor delivered by the latest Barracuda ATA drives. Subjectively speaking, however, our sample emits an irritating high-pitched squeal reminiscent of early 10k RPM SCSI disks. The whine was audible even over the testbed's relatively loud drive cooler fans.

My current 7500rpm Segate drive makes noticeable amount of noise, this one is even noisier, why can't the drive manufacturers come up with some noise suppression case/jacket for the drives. For my new desktop I would rather go in for 2 low speed (4500rpm) drive in a RAID 0 configuration.

Re:Burnout?

[Sivar]

At these speeds, would you hardware be more likely to 'burnout'?

No. As I mention above, there are 15,000RPM drives which are more reliable than any 7200RPM IDE drive on the market today.
Of course, you pay for them... Even Hypermicro, a discount reseller, sells 18GB models of Seagate's X15.3 for over $200. That's 10x the cost per megabyte of a cheaper, slower, less reliable IDE drive, but that IDE drive is fast enough and reliable enough for the average user.

My SCSI Drives

[UserChrisCanter4]

SCSI isn't that expensive, especially if you're willing to analyze what you're actually going to use, and not just RAID 0-ing two 80GB drives like a lot of people I know.

I picked up two Western Digital 9.1GB 10,000RPM SCSI drives for $35 each, shipped. If you don't have a controller, U-160 Cards can be had for about $70. I stick my OS on one drive, swap and applications on the second, and have a 45GB IDM Deskstar (75GXP and still running after 2 years, I like living on the edge) handling mass-storage tasks.

According to WD's site, these drives have transfer rates comparable to the 8MB Cache IDE drives, but seek times in the 5 ms range (vs. around 8.5). Oh, and they're not particularly loud either, at least not anything I've noticed.

At $160, this drive doesn't seem like a good idea. I've seen numerous 10K ~36GB SCSI drives for about $30 more. I guess you can factor in the card cost if you honestly want to, but if you're talking about RAIDing these things, you're probably talking about buying a good SATA or IDE RAID card anyway.

If you have plans to archive every friggin' CD you own in FLAC format, then SCSI isn't a cost-effective method to go. I don't. YMMV, but I've found that I can beat the hell out of the computer and I don't see the nasty drive access issues that I used to. For a site where a lot of people piss and moan about not needing this many mhz or that DX9-capable card, I'd say the logic of smaller faster drives when you probably aren't gonna fill the giant ones is pretty evident.

tom's hardware

[TerraFrost]

tom's hardware has some additional info on this harddrive:

http://www.tomshardware.com/technews/20030210_0836 51.html

Re:I know it's a joke, but

[Sivar]

Have anybody ever actually thought about it? For the amount of extra money to blow, why not spend more for memory and have EVERYTHING run from there? what, 4G is not enough for your desktop system? x86 only addresses that much right now, ya know...

Compare the price of 4GB of RAM with a 10GB hard drive. Also note that all memory used for a RAMdisk (as disk which will vanish once power is turned off) will be unavailable to applications.

Notice that computers run on multiple tiers of increasingly large and decreasingly expensive storage. This has been found to have the best performance/cost ratio. First we have registers, then L1 cache (except for Pentium IV's), then L2 cache, then on some systems L3 cache, THEN RAM, then the hard drive.
RAM is simply not cost effective for mass storage, and the performance benefits of using a RAMdrive really aren't very noticeable for many tasks. They help immensely for extremely random I/O, like running a mailserver, but Office and Diablo2 aren't going to run so much faster that it justifies the huge jump in cost and huge increase in risk (RAM drive dying when power goes out).
Besides, if we used a slow hard drive to load 4GB of data into RAM, can you imagine how long booting the system would take?

That said, there are companies offering battery-backed RAMdrives which fit in a PCI slot, and there are those (Armadillo comes to mind) which offers huge, fast FLASH-RAM drives in both IDE and SCSI flavors, but they are very expensive. There's more to making one than simply collecting a bunch of DIMMS together, ya know. :)

I'm still trying to figure out what was wrong. . .

[kfg]

with 78 rpm. If it was good enough for Sachmo it's good enough for me.

KFG

Re:I know it's a joke, but

[lingqi]

that wasn't my point, and you don't have to explain memory hierachy to me, I know it well.

The point is - hard drives with high transfer rate (okay, so 10k will afford you a few microseconds of access time too) have very few benefits, and only in a very few areas (that *I* can think of, anyway):

1) video-edit
2) system boot
3) kernel compile; maybe
4) swap

now, with a large enough memory, you shouldn't ever NEED to swap, or worry about using massive space for kernel compile (and really now, you gentoo kids need to chill out a little), etc etc.

For video edit, you can use the extra space anyhow so for similar price a RAID 7.2k drive array would work out better price/perf wise, I'd think (and sorry but a raided 7.2k would get better rates than 10k single, while probably not costing much more). (with raid card, you can get probably three 7200 drives while only two 10k drives)

so, besides boot-time, WHY would you need a faster hard-drive; or the question being, why invest the money into a faster drive, instead of a LOT of memory?

can you imagine how your system will scream if it never have to page, ever again? (technically, you can't really "page" anyhow since you already filled all 32 bits - that's AFAIK, correct me if you know better; it's been a while)

so, it's more like a economics question.

I am not really suggesting RAMDISK, btw - I just think that you can compile your OS / programs with option like "I have massive memory so use it lavishly and don't touch that drive."

Re:hot-swappable IDE drives - what's the fuss abou

[ocelotbob]

You've merely managed to get lucky, as hot swapping IDE drives is potentially a very bad thing. As it's not designed to be done, protections aren't in place to keep transient voltage, such as static shocks, from wiping out your controller and/or drive. Yes it may work for you most of the time, but I would not trust doing such a procedure on a computer which I'd put important information on.

You're playing russian roulette by swapping out drives. You're probably best off getting a good case, and an extra controller card for your spare drives.

As I said on the SR discussion forum.....

[AbRASiON]

I find this drive a disapointment.

This drive uses 36gb per platter.
On a 10,000 rpm drive the platters need to be somewhat smaller, due to them physically spinning faster it can cause "problems" with a full size 3.5" disk @ that speed.

Hence they physically make the disk smaller, so I can totally understand them NOT acheiving the current 7200rpm "flagship" 83gb per platter, however 36gb per platter is quite old for 10,000 rpm drives, which is quite a shame, 50 gb per platter would have been magnificent.

Unfortunately due to this, it indicates (at least to me) that this is nothing new technology wise, but a 10,000rpm disk with an SATA controller strapped on to it, they may have even licensed it from their buddies @ IBM (since they used to be chummy in the early 7200rpm days of ATA)

I'm having some guys on the SR forums claim the drive isn't that bad at all and that claiming it's a disapointment is silly because it still does X, and that's fine it's their opinion.

However MY opinion is that this drive LOGICALLY should be *THE* fastest ATA drive in existence, bar none in all benchmarks - that's what we enthusiasts want and what we will pay for - we want it to not only be faster than all 7200rpm drives (bar none) but be the fastest ATA drive period - if they can acheive this and truely blur that SCSI / ATA line - the "geeks / losers and enthusiasts" (read: myself and many others) will glady drop the same money we would normally drop to receive 2.5 and even 3x the space.

As I've said previously, most people (I feel) who initially saw the PR for this disk approx 3 weeks ago would have thought this: - "that drive will be the fastest at everything ever besides scsi" that's their expectations, and that's mine - and unfortunately it's not the case.

So some of you may like the drive, but after reading SR's review I'm not down with that drive at all - also take note the drive makes the distinct "10,000rpm whine" sound which is disapointing as well.

Big sigh from me...... these damn "hacked up" tape drives (which is all a hard disk is a "SUPER" tape drive) should be long gone by now! - magnetic spinning media has been holding the PC back for a long time, the second we have 100mb a second (slow by todays memory standards) and sub .2ms access times we will really start to see some impressive things.

Re:The roof... The roof...

[Sivar]

Any faster, and AMD will grow angry at their loss of the Heat Champion Throne. ;)

AMD already lost that title to Intel's 3.06GHz P4, which can output over 100W of heat (compared to 74 for the hottest AMD chip).

Re:Can they produce these with a serial ATA interf

[cheezedawg]

My impressions of the drive:

-It is very heavy. It surprized me how much heavier it was than the other 7200 RPM drives.
-It has what look like a built in heatsink in it's case, and I didn't notice it feeling much hotter than other drives (maybe because of the heatsink)
-It was a noisy environment, but the drive seemed almost as quiet as the other drives (again, kind of surprizing to me)

About the speed, we mostly run proprietary traffic generation programs to the drive, so I haven't really been able to use it in a real world environment yet (no Windows or games or anything). That being the case, it is hard for me to compare.

Re:Can they produce these with a serial ATA interf

[matguy]

it's like this:

Spindle speed and areal density have a give and take relationship, faster spindle = less areal density being the data can only be reliably read so fast by a generation of drive head and disk platter technology. Improving the head and platter designs gives you higher areal density at a given spindle speed, and sometimes a higher top spindle speed, but increasing that spindle speed will quickly reduce your areal density, sometimes to the point that the spindle speed increase can actually lower your sequential read and write speeds.

more areal density generally produces:
-faster sequential reads and writes
-higher capacities

faster spindle speed generally produces:
-faster sequential reads and writes
-less rotational latency
-lower areal density = lower drive capacities
-more noise
-more heat

Of course there are excpetions to these attributes, but generally they are cause and effect.

In general your big fast 15k SCSI drives aren't always that fast in big sequential read/write tests, at least not reletive to their spindle speed, but they generally fly in random small read/write scenarios being they're generally designed for servers and most servers require that kind of data transfer and that's where the shorter rotational latency is going to really help.