Guest Eugene

Hitachi Deskstar 7K1000

Recommended Posts

Guest Eugene

Through its unique five-platter assembly, Hitachi Global Storage is the first manufacturer to hit the prestigious one-terabyte mark with the remarkable Deskstar 7K1000. How does it stack up? Join StorageReview as we pit Hitachi's giant against the best that the SATA interface has to offer.

Hitachi Deskstar 7K1000 Review

Share this post


Link to post
Share on other sites
Posted May 29th, 2007 by eugene

Wow, this drive is so fast, it sent Eugene into the future! Seriously though, nice drive. I know what the benchmarks say, but do you think this drive would really be faster than the Raptor in "real world" computers. ie... ones that aren't run off a clean image every time you boot up? Imagine one of these about half full with a typical amount of fragmentation for a home user, and I picture a drive that won't come close to the results it posted in this review. Obviously the Raptor would suffer performance degradation as well, but one would think it would be to a lower degree. Any thoughts?

Share this post


Link to post
Share on other sites

Where's the quick reply option for the Article Discussion forum?

KingGremlin, you could always partition off the first 150 GB of the drive to reduce the spread of important OS & app files across the disk, then use the rest for bulk data storage.

It would be interesting to find out if a RAID 0 of two Raptor 150s (or OS on one Raptor, games & data on another) worked out faster than a single 7K1000, given the price. Of course, you'd still be about 700 GB short in the capacity department!

Share this post


Link to post
Share on other sites
With a 43.5 dB/A objectively-measured noise floor, the Deskstar T7K500 does not break any records when it comes to idle noise.

What about the 7K1000? :)

Under a full seek, the drive's actuator chatter is noticeable, perhaps slightly louder than perhaps slightly louder than the Barracuda's and Raptor's output but not overly heavy in any way.

Share this post


Link to post
Share on other sites

Don't forget to add 7K1000 to Performance Database. Unlike as usual, this time review was out before database entry.

Share this post


Link to post
Share on other sites

And how are the scores with NCQ disabled? It beats Raptor with NCQ enabled but I would assume it would fare even batter in Single-User Suite if it was disabled.

7K1000 versus 7200.10 750GB (or more accurately it's identical twin: ES 750)... I think when the difference between closest competitors (capacitywise) is this big, it's obvious there'll be a noticeable difference in performance (if you have fast enough system to bottleneck the HDD).

Office: 917 vs 575. High-end: 735 vs 535. Farcry: 877 vs 671. Sims 2: 962 vs 645. WoW: 741 vs 485.

59%, 37%, 31%, 49% and 53% performance edge over competition, respectively.

History repeats itself. 2nd gen Kurofune vs. 2nd Raptor. 3rd gen Kurofune vs. 3rd gen Raptor.

Share this post


Link to post
Share on other sites

Great review! And a major relief to finally see one on a desktop drive after half a year~ as usual, kudos to SR for another quality effort. The 1TB drive may be bleeding edge, but imho the 750GB probably offers greater bang for the buck (at this level anyway).

History repeats itself. 2nd gen Kurofune vs. 2nd Raptor. 3rd gen Kurofune vs. 3rd gen Raptor.

I wonder why Hitachi keeps two lines of high-cap drives (Kurofone and Vancouver) instead of merging them together; or is Kurofone specifically for 5-platter designs only? Otherwise they could use the Kurofone platform and just lose the platters as needed, as per usual - after all, IIRC Vancouver uses a newer (?) DSP.

Hopefully the 7K1000 series proves to be reliable - that's an awful lots of eggs in one basket, as far as data storage is concerned. As it is, for some reason, Hitachi T7K500 and 7K500 drives appear to take quite some fire over reliability issues from user reviews at various online vendors compared to WD or Seagate.

Share this post


Link to post
Share on other sites

FINALLY! Excellent review, by the way.

But I do have one question about another front page news item. The mention that the Atlases got tested, but "never quite got the go-ahead from the manufacturer for a formal review." Uh, they sent it to a review site, the drives are out in the wild, so any NDA will have expired. Even if you got them direct from Maxtor, unless to get them you signed an NDA with no end date (which aren't even enforceable,) then you're clear to publish the review. And if you got them from a public source (like a distributor,) and never signed ANYTHING with the manufacturer, then you're well within your rights to publish a review.

Share this post


Link to post
Share on other sites
With a 43.5 dB/A objectively-measured noise floor, the Deskstar T7K500 does not break any records when it comes to idle noise.

What about the 7K1000? :)

Under a full seek, the drive's actuator chatter is noticeable, perhaps slightly louder than perhaps slightly louder than the Barracuda's and Raptor's output but not overly heavy in any way.

Hehehe :)

That aside, perhaps SR could make greater use of their noise-measuring setup by adding a condenser microphone right next to their sound meter to record a clip of the HDD's noise, with, for example, the first 3 seconds of the clip recording with the drive powered off (to give an idea of the ambient noise), then recording the noise of the drive as it goes through its power on, idle and seek stages during the usual objective sound pressure level meter measurements.

That, imho, might be a good complement to SR's subjective comments as well as the objective sound pressure level meter readings, since the meter readings may not be able to indentify the strong presence of particular (and potentially annoying) frequencies - which a sound recording could. Frostytech.com does this for their heatsink reviews, for instance. It's not perfect, but it's a useful relative measure between HDDs and works along the same lines as "a picture is worth a thousand words" :)

Share this post


Link to post
Share on other sites
Guest Eugene
That aside, perhaps SR could make greater use of their noise-measuring setup by adding a condenser microphone right next to their sound meter to record a clip of the HDD's noise, with, for example, the first 3 seconds of the clip recording with the drive powered off (to give an idea of the ambient noise), then recording the noise of the drive as it goes through its power on, idle and seek stages during the usual objective sound pressure level meter measurements.

That, imho, might be a good complement to SR's subjective comments as well as the objective sound pressure level meter readings, since the meter readings may not be able to indentify the strong presence of particular (and potentially annoying) frequencies - which a sound recording could. Frostytech.com does this for their heatsink reviews, for instance. It's not perfect, but it's a useful relative measure between HDDs and works along the same lines as "a picture is worth a thousand words" :)

This is something I'd be willing to explore further... at the time of Testbed3's introduction and Testbed4's development, it seemed to me the only ways to capture the sound were bulky DAT recorders or another dedicated PC with a high-end sound card and/or mic.

I suppose now we could look into either a USB stand-alone mic with a high-quality A/D (are there such things?) or a standalone poertable recording device. Anyone have any recommendations?

Edit: We currently use a stand-alone PC to power up drives for sound pressure measurement as some rare drives also require a data signal in addition to power to spin up. It's equipped with a fanless power supply so we're able to unplug the 2.4 GHz P4C and NL35.2 bootup drive when taking measurements... however, running the PC long enough for a software-based recorder to capture drive sounds would require powering up a drive as well as running the CPU, potentially without a fan, for an extended period of time. So, I'd be most interested in any good-value, relatively inexpensive standalone recorders.

Share this post


Link to post
Share on other sites

I wonder why Hitachi keeps two lines of high-cap drives (Kurofone and Vancouver) instead of merging them together...

I think making them all Kurofunes with less platters wouldn't be a good thing.

- Kurofune has sensors to detect rotational vibration: added price, useless feature for simple web-surfing.

- Kurofune has a spindle that is supported with a screw on top. This unconventional spindle (unconventional as it is FDB) might cost more than typical FDB motors used in Vancouver (2..3 platters) and Pathfinder (1 platter) models. This unconventional motor also has smaller coils, draw less current on start-up, but require a long period of time to spin up the platters (of course some of this is due to added inertia of more platters).

- Leaving one platter out doesn't make the actuator any lighter, just replacing heads with counterweights: all Hitachis would be as noisy seekers as the biggest variant (idle noise would depend on platter count, though).

- probably other reasons as well.

I like the way those models are kept separate, keeping the low-end models cost-efficient and offering something unique to premium drives. Sure, all IBMtachis have been quite unconventional when compared to other manufacturers.

Share this post


Link to post
Share on other sites
This is something I'd be willing to explore further... at the time of Testbed3's introduction and Testbed4's development, it seemed to me the only ways to capture the sound were bulky DAT recorders or another dedicated PC with a high-end sound card and/or mic.

I suppose now we could look into either a USB stand-alone mic with a high-quality A/D (are there such things?) or a standalone poertable recording device. Anyone have any recommendations?

Edit: We currently use a stand-alone PC to power up drives for sound pressure measurement as some rare drives also require a data signal in addition to power to spin up. It's equipped with a fanless power supply so we're able to unplug the 2.4 GHz P4C and NL35.2 bootup drive when taking measurements... however, running the PC long enough for a software-based recorder to capture drive sounds would require powering up a drive as well as running the CPU, potentially without a fan, for an extended period of time. So, I'd be most interested in any good-value, relatively inexpensive standalone recorders.

A good standalone recorder, imho, would be sufficient... 320kbps 192kHz 24-bit quality recording is probably rather overkill for our purposes, which is merely to give a decent, audible idea of what the HDDs roughly sound like and compare them inter se. 192kbps 44.1kHz MP3 would be great, I think.

Unfortunately, from what I could dig up, it seems that portable recording devices (with direct interface and transfer capability to PC) are generally split into either voice recorders or (expensive) professional portable recorders. The voice recorders - Sony is probably as good a choice as any other - are simple and fairly inexpensive; however, the majority of these devices only have a decent frequency response around the 60-3500Hz region (human voice) - not quite enough to catch especially the high-frequency whines of some HDDs. The higher end ones like Sony's ICD-MX20 are pretty good and can effectively cover 60-13500Hz, which should be sufficient for our purposes, but costs about USD250-300. :P

whiic, thanks for the info, it does make more sense with those factors in mind. I'm still debating between getting a 7K500 or hunting for a T7K500 - where I live there is only the former, but I was thinking of the T7K for its increased ruggedness (as far as shock is concerned anyway) inherent in a design with fewer platters. Guess it all balances out in the end :P

Now if only 7k1000 came in 500GB and replaced the 7K500 altogether (bringing in 7k1000's higher areal density, 32MB buffer, thermal fly height control and other technological improvements), I'd snap it up in a heartbeat! :D

Share this post


Link to post
Share on other sites
Guest Eugene
Unfortunately, from what I could dig up, it seems that portable recording devices (with direct interface and transfer capability to PC) are generally split into either voice recorders or (expensive) professional portable recorders. The voice recorders - Sony is probably as good a choice as any other - are simple and fairly inexpensive; however, the majority of these devices only have a decent frequency response around the 60-3500Hz region (human voice) - not quite enough to catch especially the high-frequency whines of some HDDs. The higher end ones like Sony's ICD-MX20 are pretty good and can effectively cover 60-13500Hz, which should be sufficient for our purposes, but costs about USD250-300. :P

A few hundred dollars wouldn't break the bank actually... but do such devices have the sensitivity required to capture the noise of today's drives? Voice-oriented recorders seem to assume a minimum noise floor far in excess of the 38 dB/A that today's quietest drives emit at a paltry 3mm distance.

Share this post


Link to post
Share on other sites

Here's what SPCR does....

http://www.silentpcreview.com/article683-page2.html

1) Thorough Acoustic Testing & Analysis

We use three methods of assessing noise:

* Sensitive lab-grade sound level meter (SLM) to record sound pressure level (SPL) from one meter distance in a test room with very low ambient noise (<20 dBA ambient) at many power levels from 40W on up to full rated output.

* Audio recordings made with a sensitive pro quality microphone and prosumer digital sound system to capture the noise made by the power supply at selected points, from 1m and 30cm (1ft) distances. These recordings are converted to high quality MP3s for readers to download for listening comparisons. It's the next best thing to actually hearing the PSUs yourself in our lab.

* Careful listening and detailed descriptions of the sound level and quality. We consider this the most important part of the acoustic analysis. 90% of what we know about a products acoustics can be learned with careful listening under varied conditions. In combination with our measurements, listening helps us to identify and confirm the effects of measured parameters on noise, and any other effects not documented or detected in other ways. These include instances of tonal noise, periodic or cycling noise,

(Note: When measuring or recording the noise, the power supply loader's internal fans and all other noise sources are turned off.)

Page 7 shows they use a separate, dedicated, low-power silent PC for the recording.

They have a dedicated-- but not special, just in a low-noise area-- for testing. 38dBA is many orders louder than what SPCR typically endures.

A quiet environment is a prerequisite to low noise testing; the lab has been measured down to ~17 dBA at night, and a <16 dBA adjacent room is also available for any PSUs that are quieter.

Share this post


Link to post
Share on other sites
And how are the scores with NCQ disabled? It beats Raptor with NCQ enabled but I would assume it would fare even batter in Single-User Suite if it was disabled.

7K1000 versus 7200.10 750GB (or more accurately it's identical twin: ES 750)... I think when the difference between closest competitors (capacitywise) is this big, it's obvious there'll be a noticeable difference in performance (if you have fast enough system to bottleneck the HDD).

Office: 917 vs 575. High-end: 735 vs 535. Farcry: 877 vs 671. Sims 2: 962 vs 645. WoW: 741 vs 485.

59%, 37%, 31%, 49% and 53% performance edge over competition, respectively.

History repeats itself. 2nd gen Kurofune vs. 2nd Raptor. 3rd gen Kurofune vs. 3rd gen Raptor.

Exactly. Eugene I don't know if it is a fair statement to say the Deskstar 7K1000 is the fastest drive, when it has not been compared to the top scores put out by the Raptor with NCQ disabled. Most of us have probably seen firmware optimizations have varying impact depending on performance, depending on how they are implemented.

The Raptor still beats it all but one test with NCQ disabled...

http://www.storagereview.com/php/benchmark...48&devCnt=2

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now