SPCR

Another HDD review based on our new HDD test methodology: Western Digital drives -- Raptor 74 & Caviar SE 250

Good review.

Are SATA connectors really failing enough to warrant developing (and marketing) a proprietary cable? I can't recall ever reading about someone having troubles with them coming unplugged over time.

Is that young Devon's voice we hear on the mp3's?

Do I sound that young?

I don't know whether the cables come out magically over time, but I do know that I've accidentally ripped SATA cables out while working the a case. Pretty tough to do that with IDE cables. I think some cables are better than others though, so it might just be an issue of proper fit.

Wow. So Raptor really is as quiet as the reference drives when idle? And seeks comparable when no AAM is enabled? (Well, obviously the sharpness of the sound (energy distribution on seek) is different, which can cause different noises when/if the drive is hard-mounted.)

Also, the power measurements show interesting results. Based on power consumption, Raptor shouldn't be that much hotter than other drives. Any chances of testing Raptor inside a Smart Drive, or in the new Scythe Silent Box?

All in all, certainly interesting facts. Seems that all HD makers are getting closer to Samsung/Seagate regarding noise levels. Any word on updates for recommended hard drives -list?

Cheers,

Jan

Rusty075 wrote:Are SATA connectors really failing enough to warrant developing (and marketing) a proprietary cable? I can't recall ever reading about someone having troubles with them coming unplugged over time.

I've had problems with that happening. Just another one of the reasons I've been staying away from SATA so far.

Any info or future review of the Caviar SE16 series that were just presented at E3? The Tech Report has some news on these drives and they claim the platters spin at 7,200 RPM, have 16MB of onboard cache, and pull data from that cache at rates up to 300MB/s with a SATA-II interface.

The Tech Report also quotes WD's site that says "noise levels are virtually below the threshold of human hearing."

http://www.westerndigital.com/en/produc ... riveID=133

They are also trying to confirm whether or not these new drives will support Native Command Queuing. I would very much like to learn more about these high performance and supposedly near silent drives.

ImJacksAmygdala wrote:The Tech Report also quotes WD's site that says "noise levels are virtually below the threshold of human hearing."

http://www.westerndigital.com/en/produc ... riveID=133

If WD's own specs are correct, noise is not a strong feature:

WD wrote: Idle Mode 28 dBA (average)
Seek Mode 0 33 dBA (average)
Seek Mode 3 29 dBA (average)

But then again... the Caviar SE model we tested is spec'd as follows:

WD wrote: Idle Mode 34 dBA (average)
Seek Mode 0 36 dBA (average)
Seek Mode 3 35 dBA (average)

So maybe the data is incorrect and the thing really IS quiet. I'll see if they can send a sample...

Love the HD reviews so far Mike. You covered off an important area of current drives that all seem quiet. I have a 200JB non-FDB WD (actually 2) and they do suck combared to these. BTW I think there is a typo on page 2 under the Raptor section :

Command queuing is crucial to server performance but offers little benefit to single users, further establishing the drive as a high performance single-user drive, not a server drive.

MikeC (in the review) wrote:The metallic ringing during seek (and the intermittent squeal during idle) is a sound we've never encountered in another drive.

I have 2 newer WD 160 GB SATA drives and do not notice any metallic ringing during idle or seek. I do have a very good soft mounting system (http://forums.silentpcreview.com/viewtopic.php?t=10450), and use AcoustiPack, so I don't know if that accounts for the fact that I can't hear it.

With my case open, I cannot hear the drives idle. With the case open, I can only barely hear the drives seek if I get my ears within 12" of the drive. With the case closed, I cannot hear anything from the drives, even with my ear next to front input of the case (I do not have a fan on the front of my case).

very informative read! Thanks!

Sata cables do come out much easier than PATA cables. A slight tug can remove the cable from the electronics. I've found this to be a good thing. When removing the sata cable, i dont have to pull with such force that my hand jerks backwards and hits my graphics card (god forbid) or something like that.

Now that is really a surprise for me WRT to the raptor. I'd have expected SCSI like shreeking (painful, almost, definitely painful once you reach 15K RPM) with it.

Now there's one more on my list: supposedly, Samsung is about to release 200GB 5400 RPM drives that are speced with 2.1bel (their older 5400 are somewhere around 2.4bel according toSsamsung themselves)

mongobilly wrote:Now that is really a surprise for me WRT to the raptor. I'd have expected SCSI like shreeking (painful, almost, definitely painful once you reach 15K RPM) with it.

I run a Seagate Cheetah 15k.3 in a very quiet system and would have to work to further reduce the noise from the Artic Cooling VGA silencer set to low on my 9800 pro before any additional reduction in hard disk noise would be effective. This is in a system which is essentially inaudible while under my desk during the day time. Seeks are audible, but muted.

Someone said I must have hearing damage when I posted about this before. This review perhaps will open a few minds to the possibility of high performance AND quiet mass storage.

patricksumner wrote:Someone said I must have hearing damage when I posted about this before. This review perhaps will open a few minds to the possibility of high performance AND quiet mass storage.

Some have probably thought the same about my hearing, because I cannot hear my disk drives in a quiet system that sits under my desk about 3-4 feet away from ears. I am constantly amazed at all the discussions about using laptop drives, RAM drives, etc, in a desktop machine to quiet the disk access.

The sudden seeks and ineffectiveness (or lack?) of the AAM implementation are not characteristics of a truly quiet drive.

Have you guys tried asking WD about that? Looking at the numbers, my first thought was that there simply is no AAM present - the differences are so small that they might just be a fluctuation or testing error. Of course, if there is no AAM now, there might be AAM later (new models or firmware) - although I'm not sure if it'd make sense to pay the price premium for extra fast seeks only to manually slow them down.

m0002a wrote:Some have probably thought the same about my hearing, because I cannot hear my disk drives in a quiet system that sits under my desk about 3-4 feet away from ears. I am constantly amazed at all the discussions about using laptop drives, RAM drives, etc, in a desktop machine to quiet the disk access.

There's nothing that amazing about this -- notebook drives are a lot quieter and if you have something like just one or 2 Nexus 120 at 6V, then any current 3.5" drive will be audible in seek, especially if the system is situated any closer or it's in a lower ambient.

m0002a wrote:

MikeC (in the review) wrote:The metallic ringing during seek (and the intermittent squeal during idle) is a sound we've never encountered in another drive.

I have 2 newer WD 160 GB SATA drives and do not notice any metallic ringing during idle or seek. I do have a very good soft mounting system (http://forums.silentpcreview.com/viewtopic.php?t=10450), and use AcoustiPack, so I don't know if that accounts for the fact that I can't hear it.

With my case open, I cannot hear the drives idle. With the case open, I can only barely hear the drives seek if I get my ears within 12" of the drive. With the case closed, I cannot hear anything from the drives, even with my ear next to front input of the case (I do not have a fan on the front of my case).

I also have a Western Digital Caviar drive. Mine is the WD3200JB and I have it mounted on EAR Grommets. I can't hear any metallic ringing either (even with my ear to the drive). Seeks can be heard (with AAM on) but it is very slight and when I had first enabled AAM they seemed inaudible (that is how quiet they are).

My drive also doesn't seem to have a whine.

Jan Kivar wrote:Wow. So Raptor really is as quiet as the reference drives when idle?

I'd say so. Based on SPCR reviews, I've got a Samsung P80 in my main computer, along with a 72GB Raptor (I bought it back when they were still about $200). I'd say at idle they are approximately equally quiet. And the P80 is the quietest HDD I have ever seen (and not heard). The Raptor really IS quiet at idle. (I used to have two WD "screamers" in my last system, so I know just how loud the older WD IDE drivers were. The platter whine nearly drove me crazy after a while.)

However, when seeking it's a totally different story. The Samsung is darn near silent, while I can clearly hear every seek of the Raptor (which is my system drive). I don't mind the sound too much, though, because it reminds me of the sound of the really old HDDs (think Micropolis RLL drives!), but maybe a little deeper in pitch. When pushed harder the drive makes a very SCSI-like chatter. For me, the noise of the drive is outweighed by the speed (it really does seem snappy).

Actually, I think it's still pretty quiet when seeking if you have the noise dampened somehow. I believe it was a little quieter sitting open on my desk when I was first assembling the system. I have mine mounted now in an Antec 3700AMB (purchased after Ralf's excellent review!) using the rubber grommets in the drive cage. However, the case sits in a CPU tower cabinet build into my desk which seems to have bad acoutic properties (amplifies the noise a bit).

I wholly agree with MikeC's synopsis: the Raptor is fine if you don't care about the seek noise (which might be somewhat tameable), but it's not silent by any stretch of the imagination.

Based on power consumption, Raptor shouldn't be that much hotter than other drives.

It does get hot faster than drives like the Samsung (just finger testing when I was assembling the system originally), but I don't think it should be too hard to keep cool with good airflow and/or conductive material in contact with the case.

Yet another very interesting article. Good reading.

Thanks again for a great site.

Can you tell us whether the Caviar SE 250 you tested used a Nidec or JVC motor please Mike?

Cheese78CA wrote:Can you tell us whether the Caviar SE 250 you tested used a Nidec or JVC motor please Mike?

AFAIK, neither. The Nidec / JVC motor question only applies to the Samsung P80 series.

Oops, you're right Mike. It's been a while since I read about the Samsung using both Nidec and JVC motors.

patricksumner wrote: Someone said I must have hearing damage when I posted about this before. This review perhaps will open a few minds to the possibility of high performance AND quiet mass storage.

If you claim basically not to hear a Cheetah, that might be the reason (there's something like selective hearing damage to certain frequencies, too). I once had a 1U IBM box with 12K RPM Hitachi drives on the desk next to me, boy did I work fast to get it to a state where it could go to the data center. This thing was LOUD. The fans alone were a pain but the shrieking of the HDs very nearly got he crazy.

Personally, I'm 22 and I still (many lose it some time in their puberty hence the still) hear the high frequency whine that some CRT TVs emit (basically, I can tell some TVs to be on without actually SEEING them). My mother never quite believed it until I once demonstrated it to hear (TV on mute, I could still tell if it was on or not).

Performance and noise are not perfectly linked in the hard drive world. I had a 4200rpm 20gig IBM harddrive (Travelstar series) that was as noisy as my really really old Quantum 2gig and 4gig hard drives.

I believe you can make quiet spinning 15k rpm hard drives if you put in the engineering and money for quality parts.

MikeC wrote:

m0002a wrote:Some have probably thought the same about my hearing, because I cannot hear my disk drives in a quiet system that sits under my desk about 3-4 feet away from ears. I am constantly amazed at all the discussions about using laptop drives, RAM drives, etc, in a desktop machine to quiet the disk access.

There's nothing that amazing about this -- notebook drives are a lot quieter and if you have something like just one or 2 Nexus 120 at 6V, then any current 3.5" drive will be audible in seek, especially if the system is situated any closer or it's in a lower ambient.

I do realize that notebook drives are generally quieter than desktop drives. It is just that I can't hear my desktop drives at all. My system is fairly quiet since I can barely notice the difference between when it is running or powered down (and can only notice it when the ambient noise level is very low).

Maybe some people have their system on top of their desk, or on an uncarpeted floor, which makes it possible to hear their disk drives, I don't know. I think there are a lot of things that one can do with better drive mounting and sound absorption around the drive cage before resorting to a laptop drive.

Or perhaps I am going deaf. However, I can hear my desk clock ticking from 4 feet away, and it drives me nuts (but it was a gift, so I can't easily get rid of it).

pipperoni wrote:Performance and noise are not perfectly linked in the hard drive world. I had a 4200rpm 20gig IBM harddrive (Travelstar series) that was as noisy as my really really old Quantum 2gig and 4gig hard drives.

I believe you can make quiet spinning 15k rpm hard drives if you put in the engineering and money for quality parts.

I do too. Simply make a hard drive that has platters that exbihit 50% less friction with the air when spinning, a well designed motor and a <insert number here>" form factor that is designed to minimize noise and vibrations escaping the drive to the bare minimum.

Well, the review of the raptor certainly goes well with my experiences. I've found the idle noise to be excellent, on par with the quietest drives I've heard. The performance is excellent as well. Though the seek noise is clearly audible, I've never found it annoying.

I know I'm not the only person here who doesn't consider seek noise a problem; if you are such a person, the raptor is an great, if expensive, choice.

I know the plural of anecdote isn't evidence, but I've seen/heard lots and lots and lots of SATA horror stories, no real problems yet myself though. Most seem to be QC issues, like the mb connectors snapping off entirely when unplugged since they're relatively small, or flaky connections since SATA connectors don't mate with much force (cough). I think it's kind of neat that they have the same intermittent connection problems for the opposite reasons as molex connectors when you use multiple molex y-cables (precise pin location/loose connection vs loose pins/tight connection between the plastic bits). Rather this than having to use pliers to remove molex connectors, or lining up PATA connectors in the dark though.

The beginning of the article mentions the market rate for a 74GB Raptor to be in the ~$130 range but no reseller has a price within $20 of this minimum.

Was this figure entered in error?

I can mirror the negative experiences with conventional SATA connectors... some drive SATA inputs are so loose that even a slight jarring of the case or brushing the cable while re-seating a video card can cause a disconnection.

Some of my friends have even reported the plastic L breaking off. I like the SecureConnect system a lot although the cables can be quite expensive when compared to conventional SATA cables ($9-16 for the SecureConnect depending on your reseller).

The review comments about the WD CAVIAR SE:

• "FlexPower™ – connector technology that accepts power from both industry-standard and new SATA power supplies.
  This last feature is a byproduct of using a PATA to SATA logic bridge."

I don't know if that is universally true. Although the SATA power connector provides 3.3VDC, it is probable that most (if not all) current SATA drives only need 12VDC and 5VDC. Otherwise the SATA power adapters and many SATA drive carriers would be incompatible with the drives.

At this point, the provision of a legacy power connector is likely motivated by other factors rather than being something dictated by the chipset logic.

ky wrote:The beginning of the article mentions the market rate for a 74GB Raptor to be in the ~$130 range but no reseller has a price within $20 of this minimum.

Oops

You're right, the prices listed for the two drives are reversed; it should be ~$180 for the Raptor and ~$130 for the Caviar SE.