backtop

Print E-mail del.icio.us 27 comment(s) - last by TheGreek.. on Sep 9 at 11:01 PM

Both models of Toshiba's new 1.8-inch HDD as compared to the size of an AA battery  (Source: Toshiba)
Toshiba crams 160GB into 1.8-inch HDD

Storage capacity never ceases to increase, while form factor appears to move in the inverse direction. Toshiba Storage Device Division announced its 160GB 1.8-inch hard disk drive. The new drive uses the CE-ATA interface – designed for optimal performance and power utilization in small devices – and marks the first drive from Toshiba with the technology.

In addition to the two-platter 160GB drive, Toshiba will also offer a single-platter 80GB model. Both drives achieve an areal density of 353 megabits per square millimetre (228 gigabits per square inch). In addition to increased capacity, Toshiba has reduced the power consumption of the 160GB HDD to 0.002W/GB, which is 33 percent less than the previous 1.8-inch 100GB model, said the company.

"Toshiba's new 1.8-inch HDDs build on our long-time leadership in this key product segment and gives CE manufacturers the tools to create smaller, lighter and more compelling handheld devices," said Scott Maccabe, vice president and general manager of Toshiba Storage Device Division.  "Storage is an integral part of the digital experience, and these small form factor HDDs deliver the capacity, power efficiency and throughput to support, if not accelerate, consumer demand for sophisticated portable computing, multimedia and entertainment devices."

Toshiba’s new 80 and 160GB drives are expected to be used digital audio players, digital video recorders, copiers, mobile PCs and more. Interestingly, the capacities of Toshiba’s newly announced drives match up with the storage sizes of the newest iPod classic players announced yesterday – though neither company has made any indication of which drives Apple is sourcing.

The new 1.8-inch HDDs are being integrated by OEM partners for consumer products slated for the holiday season.

Comments     Threshold


This article is over a month old, voting and posting comments is disabled
W/GB ???
By jak3676 on 9/6/2007 7:37:15 AM , Rating: 5
Did they just make this metric up? I've only ever seen drives measured in watts at idle and under load. I do appreciate the larger capacity and the increaded density alone should improve performance. The watts per GB just seems like total marketing BS to me.




RE: W/GB ???
By TomZ on 9/6/07, Rating: 0
RE: W/GB ???
By fleshconsumed on 9/6/2007 10:53:54 AM , Rating: 3
It is a useless metric because idle/load power does not depend on the platter density.


RE: W/GB ???
By TomZ on 9/6/2007 11:11:23 AM , Rating: 1
Sorry, I don't follow... W/GB is only a measure of the energy efficiency of the drive, it is not meant to relate to the platter density.


RE: W/GB ???
By fleshconsumed on 9/6/2007 12:26:46 PM , Rating: 2
W/GB ration is useless because all it does is it tells you that one drive is bigger than the other, well, I don't need W/GB number to tell me that, I've got capacity specs for that.

W/GB is simply Idle Watts divided by Total Hard Drive Size measured in GB. It is totally worthless when designing PMPs. The press release boasts 33% advantage in W/GB ratio compared to previous 100GB model, well duh, of course it's going to have 33% advantage because previous model was 100GB and new one is 160GB.

Look at it this way you got two hard drives, one is 5W idle 80GB model, second one is 5W idle 160GB model. When you divide one by another first model has 0.0625W/GB, the second one is going to be 0.03125. It looks like the second one has much better efficiency but it's not, a PMP with 160GB drive will have exactly the same battery life as the one with 80GB drive because the idle/load power which is what matters for battery life. Yes, you will be getting twice the storage, but you will not be getting twice the battery life which is what most people think when company says it is an "energy efficient drive" compared to the old model.

Like I said in the beginning, W/GB is a useless number because what it mostly tells you is that one drive is bigger than the other, it tells you nothing of how energy efficient this drive is, this is just another derived number that was invented purely for PR reasons.

*small caveat, in reality 80GB model will have slightly better battery life because it's a single platter model and 160GB is a double platter model, and in the latter case the hard drive motor will have harder time spinning up the platters, thus consuming more energy and being slightly less efficient. However, that difference is relatively small that it can be ignored.


RE: W/GB ???
By Oregonian2 on 9/6/2007 1:16:02 PM , Rating: 2
quote:
Like I said in the beginning, W/GB is a useless number because what it mostly tells you is that one drive is bigger than the other, it tells you nothing of how energy efficient this drive is, this is just another derived number that was invented purely for PR reasons.


You are probably correct. However I infer a tone in your statement that the public ('P' in PR) isn't important regarding the value of disk drives such as this one. Seeing as how it's the public who buys them and uses them and cares about GB storage vs battery life (in the new iPod "Classics" for instance) it seems that this "efficiency" is a VERY important thing. And yes, to the end customer, the "P" in PR.

P.S. - That 'P' also compares Flash versions with GB vs battery life as well. For consumer goods producers, the public is sometimes considered important to have good relations with.

P.P.S.- I'm one of those 'P's looking at the new iPods and looking a the battery life vs GB issue -- something which those drives are participating in. I want both to be big, which is to say I want that GB/watt number to be high.


RE: W/GB ???
By TomZ on 9/6/2007 2:26:57 PM , Rating: 1
quote:
W/GB ration is useless because all it does is it tells you that one drive is bigger than the other, well, I don't need W/GB number to tell me that, I've got capacity specs for that.

Uh, no, because both W and GB can be different when comparing two different models. Therefore, if you are a designer and your goal is to find a very energy efficient drive, it would make sense to compare W/GB.

Anyway, this argument is boring, so I'll just end by correcting your statement to something I'm sure we can both agree on.
quote:
W/GB ratio is useless to me

OK, I can agree with that, can you?


RE: W/GB ???
By TheGreek on 9/9/2007 10:58:17 PM , Rating: 2
quote:
Sorry, I don't follow...

It's still a worthless metric, easily understandable to any technicalminded person, whether you understand it or not.

I realize you won't ever get it, it's like sending GWB to a business ethics class. Don't bother trying.


RE: W/GB ???
By Madzombie on 9/6/2007 6:36:12 PM , Rating: 2
W/GB would only be useful if the power consumed by a device depended on how full the drive was. However, this is not the case, so the raw idle/load power consumption rates are all that affect battery life. Lets say you have a 10GB drive which uses 1 watt and a 100GB drive that uses 2 watts. The 100GB drive will have 5x higher W/GB, yet still uses twice the power. W/GB tells you nothing about the power efficiency of a device.


RE: W/GB ???
By Oregonian2 on 9/6/2007 7:49:03 PM , Rating: 2
The buyer of the drive-containing-device will look at the device's GB and battery life. Both are being looked at simultaneously, and the drive's ratio, although not the only thing contributing, certainly is a contributor to the customer's evaluation of the product overall. So it's not necessarily a biggie stat, it's important in devices like iPods that the GB be high and that the battery life be long (low power consumption). Having a high GB/watt ratio contributes to what the overall end-customer wants, so it's a "good thing" in that regard.


RPM?
By sibbor on 9/6/2007 8:49:14 AM , Rating: 2
Anyone know how fast these platters will spin? 4200 rpm or 5400 rpm? Fast 1.8" drives isn't really a big market...




RE: RPM?
By TomZ on 9/6/2007 10:09:21 AM , Rating: 2
These drives are 3600RPM. That may sound slow, but think of the target application: small, battery-powered devices. Not PCs.


RE: RPM?
By Dactyl on 9/6/2007 11:48:16 AM , Rating: 2
Further, the higher density means you get decent bandwidth at low RPM.

Compare this drive to a 7200RPM 3.5" 160GB drive.

These drives are half as fast as 7200RPM, which means seek times are twice as long.

BUT if the platters are four times as dense, you would get twice as much bandwidth out of the little thing.

In other words, it would be an okay drive for laptops--once you got past booting up!


RE: RPM?
By Samus on 9/6/2007 12:23:27 PM , Rating: 2
Your physics doesn't compute. RPM has little to do with seek time in comparison to physical size. There's less area to seek too. The immense aerial density of these drives probably improves their seek time, too.

I have a 40GB 1.8" Hitachi in my IBM X40 (purchased here on AT forums no less) and although its transfer rate is in the 30-40mb range, it has great seek time of 9ms (actually about 5.3ms when taking into account bus overhead, or in other words, 5.3ms track-to-track)


RE: RPM?
By TomZ on 9/6/2007 2:24:00 PM , Rating: 2
Physical size doesn't enter into it. Seek time includes rotational latency - you have to wait on average 1/2 the rotational period for the data to rotate to under the head. Since the rotational latency dominates other factors in seek time, a halving of the rotational speed approximately increases the seek time by 2X.

At least that's how I understand it. Please correct me if I'm wrong.


iPod + Toshiba
By Lonyo on 9/6/2007 7:10:36 AM , Rating: 2
Well apart from the fact that this announcement makes it fairly obvious, most DAPs that I know of have used Toshiba drives.
My iRiver did, my Philips did, a friends iPod did, and my iAudio does.

I was also expecting someone to announce a 160gig 1.8" drive given the Apple announcement. Nice that it's Toshiba.




Yup
By phaxmohdem on 9/6/2007 10:54:36 AM , Rating: 2
I kind of assumed this technology advance had happened when the new iPods were released yesterday @ 80gb and 160GB.

Last time 1.8" drives were updated, the iPod was also among the first user using a single platter 30GB drive, and a thicker dual platter 60GB drive.




Ipod?
By rsasp on 9/7/2007 7:03:49 AM , Rating: 2
perhaps the new generation Ipod is using this already




who cares....
By LQ on 9/6/07, Rating: -1
RE: who cares....
By mdogs444 on 9/6/2007 8:23:38 AM , Rating: 2
You've got quite along time for SSD to be mainstream.

And when they do, I dont think you're going to be seeing 160gb, - more like 128gb, 192gb, 256gb, etc.


RE: who cares....
By TomZ on 9/6/2007 10:35:57 AM , Rating: 2
One important factor between flash and magnetic media is that magnetic media is cheaper per GB. While the gap is expected to continue to narrow, flash will still continue to command a price premium for many years to come.


RE: who cares....
By Blackraven on 9/6/2007 12:25:58 PM , Rating: 2
I mean, we have 16 GB SSDs this year and 32 GB ones too.

How come they don't sell them in units of 10 GB (like in magnetic Hard drives)???

Is there a technological explanation to this???


RE: who cares....
By Oregonian2 on 9/6/2007 1:19:17 PM , Rating: 2
Memory chips are binary things. They use address fields that are binary in nature. They come in binary sizes. hard disks have little rotating things with heads and however many tracks fit. No inherit binaryness. So they can be "whatever".


RE: who cares....
By Oregonian2 on 9/6/2007 1:20:23 PM , Rating: 2
P.S. - And flash is a memory chip, just one that's non-volatile.


RE: who cares....
By TomZ on 9/6/2007 2:35:42 PM , Rating: 2
Well, I'd have to disagree with you slightly on that. Discs are inherently binary - their sector size is typically 512 bytes. But the total capacity is not a power-of-two because there is some non-power-of-two number of sectors.

Flash chips are built in powers-of-two, e.g., 1GB chips for example. Therefore, if you are making a device like a flash key, you might produce a 1-chip 1GB or a 2-chip 2GB or 4-chip 4GB flash key. Again, powers of two total capacity.

But I would argue that there is no reason that flash SSDs need to follow this pattern. This is because there are not a small number of flash chips in an SSD. For example, if you are making a 32GB SSD out of 1GB chips, then you would have 32 chips. That opens the possibility of other capacities, e.g., 24GB, 48GB, or whatever.

Therefore, I would guess that the capacities for flash SSDs being 16GB, 32GB, 64GB is maybe influenced more by marketing considerations, based on the idea that having intermediate sizes doesn't add much value. In other words, if you already have a 32GB drive on the market, you might as well introduce a 64GB one that either doubles the number of chips or uses chips with twice the capacity, rather than making a 48GB drive.


RE: who cares....
By Oregonian2 on 9/7/2007 7:30:31 PM , Rating: 2
quote:
Well, I'd have to disagree with you slightly on that. Discs are inherently binary - their sector size is typically 512 bytes. But the total capacity is not a power-of-two because there is some non-power-of-two number of sectors.


For the device to be sized inherently binary, it'd have to be binary on "everything" related to size. The number of sectors per track probably isn't inherently binary nor the number tracks per surface.

quote:
But I would argue that there is no reason that flash SSDs need to follow this pattern. This is because there are not a small number of flash chips in an SSD. For example, if you are making a 32GB SSD out of 1GB chips, then you would have 32 chips. That opens the possibility of other capacities, e.g., 24GB, 48GB, or whatever.


Still all binary numbers. May cuts down on later cost reductions though when the same capacities can be made with fewer and larger parts.

Also when a lot of parts are used, it's nicer to use narrower parts meaning few(er) data-bits per part. IOW if you needed 16 parts to do a 32-bit data width, you'd use 16 2-bit wide parts doing 2-bits per part rather than using 16 32 bit parts where you could have used ony 12 parts for a smaller sized unit. This both makes the packages smaller as well as lowering the loading on each dataline (in addition to signal integrity issues being easier to handle).


RE: who cares....
By TheGreek on 9/9/2007 11:01:24 PM , Rating: 2
quote:
For the device to be sized inherently binary, it'd have to be binary on "everything" related to size. The number of sectors per track probably isn't inherently binary nor the number tracks per surface.

<smirk> Don't confuse him with the facts.


"So if you want to save the planet, feel free to drive your Hummer. Just avoid the drive thru line at McDonalds." -- Michael Asher









botimage
Copyright 2009 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki