Hitachi 7K1000 1TB a silent beast? :O

rpsgc · Post by **rpsgc** » Mon Mar 19, 2007 10:34 am

Hitachi 7K1000 1TB a silent beast? :O
So says Anand.

According to Anandtech's review, the 7K1000 is the quietest HDD they've ever tested, even quieter than the Western Digital KS models...

What do you think of that?

Link to review: http://www.anandtech.com/storage/showdoc.aspx?i=2949
Link to acoustics: http://www.anandtech.com/storage/showdo ... i=2949&p=5

gb115b · Post by **gb115b** » Mon Mar 19, 2007 11:39 am

if these figures are right...makes it a bit of a no-brainer (aside from cost)

man, i was looking to upgrade my RAID 6 array...but not sure if i can spring for 8 of these!

over to you Mike...

J. Sparrow · Post by **J. Sparrow** » Mon Mar 19, 2007 11:53 am

If a five-platter design was really so quiet, wonder how good a single- or two- platter 7K1000 might be.

Jumper · Post by **Jumper** » Mon Mar 19, 2007 12:42 pm

200 GB/platter... As fast as a Raptor... AAM supported...

*droop*

Erssa · Post by **Erssa** » Mon Mar 19, 2007 12:53 pm

Looks promising...

grambo · Post by **grambo** » Mon Mar 19, 2007 1:20 pm

Sounds pretty awesome, for $399US too. I hope Seagate gets a 1TB model out soon (250GB x 4 platters apparently) to compete and drive prices down.

andyb · Post by **andyb** » Mon Mar 19, 2007 4:13 pm

If a five-platter design was really so quiet, wonder how good a single- or two- platter 7K1000 might be.

The big question is IF, that sound testing methodology is utterly retarded - 5mm testing is laughable, why not gp the whole hog and test at 1mm

The idea is that they get consistant results at that distance because their testing facilities are from the stone-age simply means that their testing facilities are utter shit, and therfore cant be trusted - hence the results are as worthless as a guess from someone who thinks a HDD is a brand of mobile phone.

As far as the performance goes, this is not that unexpected - Anandtech are only testing the first 1% of the drive (10GB!), therefore all of the seeks are going to be a few tracks - short/slow/quiet seeks, this is the same as testing a 10GB HDD using DOS, all of the data is going to be within the first 10% of the drive, therefore all of the seeks are going to be as above.

We have consistantly seen larger drives getting quieter, this will continue to happen so long as a very small amount of drive space is being used and tested.

The performance (if its true) is phenomenal, and Shitachi need praise - however I am cynical and stopped trusting Anands reviews a long time ago, I will believe its performance and accoustices when I see other reviews that I trust.

Andy

Shadowknight · Post by **Shadowknight** » Mon Mar 19, 2007 6:04 pm

I'm pretty skeptical. Other than SPCR, pretty much all review sites blow when it comes to analyzing noise. A lot of times what they find "whisper quiet" is just not audible over the rest of their system, which is probably louder than an average SPCR member.

Hopefully MikeC will get a review sample, then we'll see for sure.

Natronomonas · Post by **Natronomonas** » Mon Mar 19, 2007 8:47 pm

Regardless of their methodology, the fact it's quieter than anything else they've tested that same way surely bodes well...

I wanna SPCR review

andyb · Post by **andyb** » Tue Mar 20, 2007 2:47 am

Regardless of their methodology, the fact it's quieter than anything else they've tested that same way surely bodes well...

Not at all, the problem is that their environment and testing equipment is so poor its the equivelent of a + or - 25% accuracy on a benchmark.

What happens if the mic is 1mm to close or too far away - what happens with a drive that produces a noise (tone) that their crap equipment cant detect, it is therefore NOT measured. Likewise it is very possible that some drives make more noise than other at certain points of the drive, at 5mm the mic is only going to pick up noise from a tiny fraction of the HDD.

Have the measured vibration/what is the HDD attached to, is it sitting on foam, rubber, bolted into a case.???

Its worthless.

Andy

alfred · Post by **alfred** » Tue Mar 20, 2007 6:57 am

Sure. Furthermore, we're still waiting AnandTech to include some T133 and T166 Samsung drives in their reviews.

Anyway, I'll probably try this 1TB drive when it's available; it's going to help me cleaning my old backup server from some 200~250 GB drives, and if it's really that quiet I'll post some noise recordings, comparing it to my favorites.

Live · Post by **Live** » Tue Mar 20, 2007 9:59 am

andyb wrote:what is the HDD attached to, is it sitting on foam, rubber, bolted into a case.???

Its in a case.

We even removed the rubber mounting grommets in our drive cage and did not notice any differences in acoustics.

While I agree that the testing they have done is far from perfect I wouldn't call it worthless.

Further I don't understand your reasoning about testing the harddrive only to the 1%. How did you come to that conclusion? How do you propose to test the performance of a drive?

gb115b · Post by **gb115b** » Tue Mar 20, 2007 12:23 pm

yeah...as i said...over to you Mike.. the reviews here (SPCR) are in my experience pretty much definitive with regards to sound...

storagereview used to be ok... but not great.. (and they haven't doen a review for an age it seems)

anandtech as far as i'm aware never claimed to be a specialist site re: silence so i'm nto going to critisize their methodology. it's fine for the majority of readers that they have. though its not for SPCR ones...

they're more concerned with performance / price / capacity...

anyone got any further corroborating (or contradicting) evidence?

szrc · Post by **szrc** » Tue Mar 20, 2007 1:15 pm

Live wrote:While I agree that the testing they have done is far from perfect I wouldn't call it worthless.

No, it's not worthless. But for the various reasons mentioned already (loud, inconsistent environment, strange methodology with mic @5mm, etc.) it's also not reliable. More data points are needed to draw any solid conclusions.

Live wrote:Further I don't understand your reasoning about testing the harddrive only to the 1%. How did you come to that conclusion? How do you propose to test the performance of a drive?

PCMark05 (used by Anandtech) plays back a prerecorded disk trace into a "dummy" file. If the disk is properly defragmented, this dummy file will consist of a small, contiguous piece of space on disk in which all of the seeks will occur. Especially as platters become increasingly dense (like 200GB each) the distance traveled by the heads while seeking within the dummy file is very small. The noise made by such seeks will be quiet and not necessarily representative of larger seeks that one may encounter in real life.

The tool used by SPCR specifically causes full-range seeks designed to make sonic differences as noticeable as possible.

andyb · Post by **andyb** » Tue Mar 20, 2007 3:50 pm

szrc put that perfectly, I had no idea of the tool/benchmark that they used, but I do knot that data gets put onto the start of the drive first and (generaly) works through it in sequential order.

I supposed that they used 10GB for XP, the Swap File and all of their programs, thats 1% - the other 99% of the drive was never tested, they never did a single "long seek".

Again I will say its worthless.

Andy

EsaT · Post by **EsaT** » Tue Mar 20, 2007 11:44 pm

In StorageReview forums one poster told interesting point.
That drive has some kind low RPM power conservation mode so drive might have been in that because they just didn't measure noise immediately when drive was idling. That would also explain low idling temperature.

andyb · Post by **andyb** » Wed Mar 21, 2007 7:34 am

It would also explain why there is a large power consumption drop on the official Hitachi documentation when the drive is running in "silent" mode.

Andy

J. Sparrow · Post by **J. Sparrow** » Wed Mar 21, 2007 8:07 am

EsaT wrote:In StorageReview forums one poster told interesting point.
That drive has some kind low RPM power conservation mode so drive might have been in that because they just didn't measure noise immediately when drive was idling. That would also explain low idling temperature.

AFAIK all recent Hitachi drives have the same features, it's called APM. I'll go check SR, though.

HedgeHocker · Post by **HedgeHocker** » Fri Mar 23, 2007 11:53 am

The good people at Storage Review say Dell and Alienware have exlusive rights to sell this drive anyway and only offered as an upgrade option on thier systems... looks like a long wait for this thing so I will have to forget about getting one. I'm looking for a very energy efficient high capacity drive right now myself and I had high hopes to get one until that wall was hit. I'm reading forums and tables on sites for that 500Gb of higher drive for a 24/7 mostly idel PC. Tell me if you know it! please...

I wanted to mention this availability since I didn't see it in the Anandtech review (but I didn't read all of it either) so back to the more in depth discussion I suppose. Drop me a note on the drive I need.

whiic · Post by **whiic** » Fri Mar 23, 2007 12:42 pm

It's mentioned in Anandtech review:
"What we did not expect was that this drive would be offered exclusively through Dell or its subsidiary Alienware in select XPS, Aurora, or Area 51 gaming desktop systems before general retail availability in the next two to three weeks."

But there's something I managed to ignore the first time reading: it's only going to be "two or three weeks" of exclusive Dell retail and then released to smaller retailers. Or maybe "general availability" means retail availability through Dell only (i.e. not bundled-only as it might be now).

I sure hope it's going to be available at all retailers willing to sell Hitachi drives, because I couldn't care less what Dell offers.

madshi · Post by **madshi** » Mon Mar 26, 2007 2:54 am

The drive is already listed by number of German retailers - and for a good price, I should mention. Anyway, most don't have specific shipment dates. One shop lists 04.07.07, which I think is probably a typo. Probably they meant to write 07.04.07, which would be in about 2 weeks.

jojo4u · Post by **jojo4u** » Mon Mar 26, 2007 6:18 am

On the whole, I find the test quite thorough. Many real-life benchmarks and no IOMeter crap. One must have in mind that the tests are performed within the first GB, so a 100 GB disk is of course very fast.
The low-rpm idle sound reasonable and is referred to "low-power" in the datasheet. Power figures in idle/unload/low-power: 9.0/6.0/4.5 W. It's unclear wether activating AAM also activates APM, since anandtech still hasn't heard about this feature (can be checked with hdparm or hddscan or Hitachi's own tools). Btw, in the at this time 71 comments are only two which have AAM/APM in the text

[/b]

whiic · Post by **whiic** » Mon Mar 26, 2007 1:27 pm

According to 888 at SR:
In example 7K1000 Datasheet lists for power draw:
* Idle/normal = 9.0W
* Idle/unload = 6.9W
* Idle/lowRPM = 4.5W

It's probably 6.9 watts and not 6.0 watts. Of course I could check the datasheets now, but I prefer reading the OEM specs when the come available (and that's 200 pages of techno babble).

6.0 watts just appears too low for me to believe. It's simply because 7K500 (previous generation Kurofune) had:
Idle average 9.0 watts
Unload idle average 6.8 watts
Low RPM idle 4.4 watts
(according to 7K500 PATA OEM specs version 1.5)

7K1000 is quite an improvement as SATA variant of 7K500 consumes ~1 watt more in all operating modes than PATA. At least in specs... 1.0 watt extra for SATA/300 (drives like 7K500) and 0.7 watts extra for SATA/150 (drives like 7K400). I believe SATA supports some interface power saving mode. If I remember correctly, older Hitachis didn't support it. Maybe the new ones does? If they didn't, it'd quite be a no-brainer as it'd partially negate Hitachi's excellent efforts at bringing mechanically(*) highly efficient drives to market... plus other manufacturers (like Seagate) support it. (Though, despite Seagate's interface power saving, they are still noticeably hotter than Hitachi's that have the same number of platters.)

(*) Well they are still extremely hot seekers. If you plan to use them in a server, give them adequate cooling. When idling they should be relatively cool. Then again: relative to what? There's no 5-platter competitors to relate to. Comparing them to 4-platter competition, it might appear "unfair" but they actually fair suprizingly well and in some cases operate even a tad cooler.

But they aren't exactly something for silent PCs. For silence and 1 TB, it's better to keep a separate home server in another room no matter what solution you'd go for... 1x 7K1000 or 2x T166 or 12x P80 80GB or 12x-48x Barracuda IV. I don't consider SPCR's new focus to flagship capacities a bad one. Some of us need the capacity and going for multiple small drives usually is not better. Going with the "noise per drive" focus suit the people who don't need GB. Those who really need all the GB (or TB) the world has to offer, "noise per GB" has some validity. And since home theater as a function of computer has become quite a bit more important lately with all the warez... correction: legimate back-up copies of movies and TV series.

Enma Ai is calling me~ <3 so good nitey or whatever --->

regmentor · Post by **regmentor** » Mon Mar 26, 2007 2:12 pm

they're pulling out the stops but i still can't forgive them for the deathstar line

madshi · Post by **madshi** » Tue Mar 27, 2007 6:13 am

whiic wrote:But they aren't exactly something for silent PCs. For silence and 1 TB, it's better to keep a separate home server in another room no matter what solution you'd go for... 1x 7K1000 or 2x T166 or 12x P80 80GB or 12x-48x Barracuda IV.

I'll use 4x 7K1000. I'm considering waiting for the CinemaStar version of the drive, though.

Does anybody have any information about how Hitachi DeskStar and CinemaStar drives differed in the past? Is the CinemaStar version worth waiting for?

Post by **MikeC** » Tue Mar 27, 2007 8:33 am

I've asked for a sample. Hopefully, we'll see something soon. I suspect review samples might be in short supply right now.

But I have to say I am not so hopeful about its acoustics. It's a 5-platter drive, and Hitachi's own specs say....

Idle: 2.9 bels (typical)
Silent seek: 3.0 bels (typical)
Seek: 3.2 bels (typical)

This is definitely better than the 3.1 bels idle of the 500GB Hitachi we reviewed last year. That one we at measured 26/28 dBA@1m (idle/seek) and was no challenger to the Samsungs. The single platter Hitachi 7K80 is rated at 2.6 bels idle and measured quite well at 20/24 dBA@1m (idle/seek) though we did not like its noise quality as much as some others.

So can a 5-platter drive Hitachi rates at 2.9 bels idle be the quietest drive we've tested? I doubt it very much.

whiic · Post by **whiic** » Wed Mar 28, 2007 7:27 am

"So can a 5-platter drive Hitachi rates at 2.9 bels idle be the quietest drive we've tested?"

The point is, it doesn't have to be. To me it's enough it's quieter than 2x T166... or 12x 7K80.

7K1000 vs T166 or WD5000AAKS would give more perspective than just comparing single drives. Of course if sample of 7K1000 happened to arrive before a sample of T166 or WD5000AAKS, it's probably not worth delaying the review. (Comparing 7K1000 to 2x T133 wouldn't be entirely accurate but could give a hint whether two-drive solution is better or not. I don't know if such a comparison should be made formal or not.)

Post by **MikeC** » Wed Mar 28, 2007 7:31 am

whiic -- you have a point.

I was simply referring back to Anandtech's comment.

madshi · Post by **madshi** » Mon Apr 02, 2007 1:48 am

Multiple shops here in Germany claim that the drive will be released here in mid July!

jojo4u · Post by **jojo4u** » Mon Apr 02, 2007 4:24 pm

Reply in comments by the author about APM/AAM:

RE: AAM and APM by Gary Key on: Mar 30, 2007 4:05 PM wrote: We had a lengthy meeting with the Hitachi engineers this week to go over APM and AAM modes along with the firmware that is shipping on the Dell drives. I hope to have some answers next week as testing APM capabilities on a Dell based system resulted in a slightly different behavior than our test bench. I have completed the balance of testing with various AAM/NCQ on/off combinations and some additional benchmark tests. I am hoping to update the article next week. Also, I ran acoustic tests in a different manner and will have those results available. Until, then I did find out that sitting a drive on a foam brick outside of a system and taking measurements from the top will mask some of the drives acoustic results. The majority of noise emitted from this drive comes from the bottom, not the top.

Hitachi 7K1000 1TB a silent beast? :O

Hitachi 7K1000 1TB a silent beast? :O

power consumption

Re: power consumption