Toshiba Boosts Hard Drive Density By 50%

An anonymous reader writes "Toshiba has unveiled a ground-breaking technology that boosts recording density by 50% on an 80-GB, 1.8", single-platter drive. Using what it calls Discrete Track Recording technology, Toshiba was able to pack 120 GB storage on a single 1.8" platter. The new development will hugely benefit media player, UMPC, and ultra-portable laptop segments where 1.8" drives with maximum possible capacity are in great demand."

whenever they talk about hard drive densitiy

[weirdcrashingnoises (1151951)]

i think of this...

George McFly: Lorraine, my density has bought me to you.
Lorraine Baines: What?
George McFly: Oh, what I meant to say was...
Lorraine Baines: Wait a minute, don't I know you from somewhere?
George McFly: Yes. Yes. I'm George, George McFly. I'm your density.

Damnit...

[feepness (543479)]

Now I have to wait longer for SSD to become the clear winner.

Re:Damnit...

[Kjella (173770)]

Oh how terrible if SSD has to actually beat the competition, that it didn't just lie down and die when SSDs announced they were taking over. As bulk storage, SSDs are still damn expensive. Think either-or, say an 8GB SSD soldered on the mobo (1.8" form factor is way overkill) with OS/apps and 120GB bulk HDD space on a 1.8" HDD. Sounds good to me, at least until 100GB+ SSDs are cheap and that may take a looong while.

Re:Damnit...

[mcrbids (148650)]

Now I have to wait longer for SSD to become the clear winner.

I've been waiting for well over 10 years. When I first thought that SSD was going to "win", 1 GB drive was huge. Now, it's $9, plugs into a thumb-sized slot in 5 seconds, and is available at the local Wal-Mart. The mechanical drives sport 750 GB for $200 that the 1 GB drive used to cost. (and that doesn't even account for inflation!)

I have a digital camera with video and sound. It's up to 800x600, and with my 2 GB flash cartridge, I get up to about an 30 minutes of video. It's very small, lightweight, and runs on a couple AA rechargables.

Still think that SSD hasn't "won"?

Re:

[jcr (53032)]

They are now and will always be more expensive than mechanical drives.

Always?

I wouldn't bet on that. At some point, no moving parts has to beat moving parts.

-jcr

Re:

[mcrbids (148650)]

I wouldn't bet on that. At some point, no moving parts has to beat moving parts.

Yeah, like how the Peltier cooler has replaced mechanical refrigerators. Or the thermocouple has replaced mechanical generators and steam in nuclear power plants.

Oh, wait....

Re:

[aliquis (678370)]

At some point, no moving parts has to beat moving parts.

Yeah that is what I'm beting when I read Slashdot ;D

Re:

[timmarhy (659436)]

you were always going to have to wait a long time.

yes spinning disks suck, but 100x the cost sucks even MORE. SSD is up there with those holographic drives we keep hearing about that are only 2 years away...

Re:Damnit...

[timeOday (582209)]

No, I think SSD really will win. Over the past 5 years, solid state has doubled in capacity over and over again, while hard drive capacity growth has slowed. The net effect has been tremendous gains for flash memory.

The other trend I see is satisfaction with hard drive sizes. Notice how the blurb for this article only mentioned 1.8" platters, as if capacity was only lacking in small devices? For most people, requirements for storage simply aren't growing. Even Vista is insignificant on a cheap, commonplace 500 GB drive. My PVR PC still has a 160 GB drive, I just can't be bothered to upgrade.

With near 0 access latency and higher reliability, flash doesn't have to beat winchester drives in $$/GB to win. It just has to be big enough and cheap enough, and it's getting there.

Re:

[Wicko (977078)]

I can't see it winning any time soon. There will always be need for larger capacities, especially with the relatively new HD formats becoming more and more common. I don't really see the two as truly competing, I doubt the capacities will catch up for a long time, perhaps some new format will appear long before then. I can see a lot of systems using SSD or hybrids for their main partition, including an OS (although that would probably kill an SSD much sooner than a contemporary disk), and hybrids/contempora

Re:

[Ramble (940291)]

Possibly they'll win one day but at the moment (and this will be true for some time) there is no clear advantage to SSDs. Rubbish storage capacity, similar throughput and expensive as hell. Latency is one of the only things they have going for them.

Re:

[owlstead (636356)]

Throughput is lower, but SSD's already provide faster startup times than HDD's. Latency, reliability, size, low power use, noise are all clear advantages for SSD's over hard drive technology. If prices come down a bit (or rather, a lot) for SSD's when they become more regular, HDD's will only have size and size per dollar left as advantages.

For my OS, flash seems a very logical choice. Price and some uncertainties about the flash currently on the market have withhold me from buying it so far, but this will

Re:

[DrMrLordX (559371)]

Just keep your eye on MTRON. They're pushing a lot of SSD harddrives out on the market, albeit at astronomical prices. Still, the performance on their drives is fantastic.

Re:

[Jeff DeMaagd (2015)]

So what? It's not as if the advancements in hard drives are now unexpected. They are going to be pushed as hard as possible as long as they are competitive. SSDs are getting larger too, and flash chips are continually dropping in price. I really don't expect a "clear" winner for some time. What is a clear winner in one circumstance is a clear loser in another. I use a couple machines that are running on flash drives, one is a 512MB CF drive, but that's because vibration is a problem because it's runni

One thing confuses me about SSD

[symbolset (646467)]

They're solid state. To me that means that every bit is as close as every other, near enough.

It should be possible to deliver far more bandwidth from an SSD than through magnetic media.

But the best claims I've seen for SSD are about 10MB/sec.

Where's my pen drive that's capable of 480Mbps? Where's the SATA attached SSD capable of 3Gbps?

Re:

[TheRaven64 (641858)]

You can make flash as fast as you want. Serial reads are limited by the speed of the individual elements. Flash is typically made as a block device, however, so you can read every bit in a block in parallel. This, however, gets expensive very quickly. You need to double the number of pins and the number of interconnects to each cell in order to double the transfer rate. Since the interface is serial, you will probably need to buffer the reads for a complete block into fast RAM and stream them out, slig

Re:

[suv4x4 (956391)]

Now I have to wait longer for SSD to become the clear winner.

SSD will likely never be the clear winner when we're talking about large storage. Just like HDD didn't replace magnetic tapes for backup in companies, because of their lower capacity and reliability.

You can expect SSD to make a boom (already is anyway) in mobile devices, and HDD's will become all hybrid: with 10-20 GB of SSD and a spinning disk to fill-in the capacity up to say 300-400 GB.

This way, you get the best of both worlds. Your startup and

Technological anachronism

[Circlotron (764156)]

Imagine if the humble telephone dial had received this much effort and technology. What would THEY be like now?

Re:

[Waffle Iron (339739)]

Imagine if the humble telephone dial had received this much effort and technology. What would THEY be like now?

Most likely, all of the possible 1e10 U.S. phone numbers would be encoded as individual nanoscale holes around the dial. You could dial anybody in the country with just a single flip of the wrist.

Re:

[Bill, Shooter of Bul (629286)]

Good question. Its been my longstanding dream of building a cell phone with a phone dial and no screen. It would be an old school handset with a dial, and a hook built into the belt holster. Maybe the dial should be on the holster as well. And I would tell everyone my "phone was off the hook", sometimes as a joke, sometimes as an excuse for not picking up the phone. Although, I really don't know how many people would get the joke.

Re:

[TheThiefMaster (992038)]

It would be voice activated. Like, uh, your mobile can do.

Yeah.

another infomercial

[semiotec (948062)]

1. Samsung had announced their 120 Gb 4200 rpm 1.8" drive a couple of weeks earlier (http://home.businesswire.com/portal/site/google/i ndex.jsp?ndmViewId=news_view&newsId=20070820005213 &newsLang=en), but there's no sign of it available yet.

2. The Toshiba 120 Gb drive, according to PC Watch Impress (http://pc.watch.impress.co.jp/docs/2007/0906/tosh iba.htm) is 3600 rpm with CE-ATA interface, not really suitable for notebooks, even ultra-portable ones.

I'd guess the new iPod Classic uses the Toshiba drive, since it supposedly uses even less power compared to their previous 1.8" drives. But if this is the case, it means I can't just rip it out of the iPod to plug into my laptop, since the interface doesn't appear to be compatible with their previous 1.8" drives.

However, I still hope that at least one of these make it to the retail market. It would be nice to be able to double my current 80 Gb drive.

Re:

[Wicko (977078)]

Err, it did say that it was per platter, not a single drive. I'm not sure how many platters 1.8" drives have, if I had to guess, that Samsung might use 2 platters. Just a guess anyway, too lazy to look that up.

Re:

[Nimey (114278)]

I'm pretty sure that some manufacturers put 1.8" drives into ultraportable laptops and tablets. When you want something physically small that sips power, you put out all the stops.

Re:

[semiotec (948062)]

"You don't put 1.8" drives in notebooks, you put them in iPods. 2.5" drives go into notebooks."

Thanks, I'd be sure to pass that onto Toshiba (R200, R400), Sony (TZ series), HP (nc2400, nc2510) and Dell (D420/430) and tell them that they really should listen to some AC on Slashdot.

"some sort of crazy ultraportable"

and that also means they weight 2 lb instead of 5 lb. The last three of my laptops have all been ultraportables, and all together they still weight less than 15" notebooks out there.

Re:

[Zero__Kelvin (151819)]

"AC with good reason: "You don't put 1.8" drives in notebooks, you put them in iPods. 2.5" drives go into notebooks, 3.5" drives go into desktops and servers. (Anything bigger than that has been obsolete for decades.)"

semiotic replies: "Thanks, I'd be sure to pass that onto Toshiba (R200, R400), Sony (TZ series), HP (nc2400, nc2510) and Dell (D420/430) and tell them that they really should listen to some AC on Slashdot."

... and even if semiotic could not find and or name a single manufacturer who does so

I found the patent info on this...

[dada21 (163177)]

...Toshiba's patent just says to take out the MFM hard disc controller, and replace it with their new RLL controller. I tested this myself and got my 10MB drive to a full 15MB without a single problem!

Re:

[slittle (4150)]

Did you know that you can also drill a hole in it and get 30MB?

Honest.

Re:I found the patent info on this...

[Headcase88 (828620)]

dada21 casts JOKE at Anonymous Coward
Woosh! The attack misses.
Headcase88 taunts from afar
Headcase88 equips flame shield

Better information on this

[eclectro (227083)]

There is an article about discrete track [eetimes.com] recording that explains it pretty well. Using materials with different magnetic properties they are able to map channels onto the platter (hence the 'discrete'). Presumably this might would be cumbersome to manufacture for larger discs, but less so with smaller disks.

Re:

[Anpheus (908711)]

God, I'd really like it if they'd come out with discreet track recording technology so I can hide all my porn!

(But at 120GB, that's not nearly enough space!)

Re:

[tomstdenis (446163)]

It appears that they have to etch something on the platter to get this to work, probably the yield of 3.5" platters isn't worth the cost, since you can't just use multiple platters (3.5" == desktop remember).

Re:

[by Anonymous Coward]

density & capacity aside, i am wondering why don't we have drives with multiple independent R/W heads?

Re:well...

[toddestan (632714)]

I would think that it would be cheaper and easier just to take two standard drives and RAID them for the same effect rather than to build some two servo monstrousity.

Re:

[ChrisMaple (607946)]

I've read that this has been done, but it's not cost-effective. I guess that would be because there isn't much demand for this technique, so low production quantities mean high prices. OTOH, I don't think anyone has tried it recently, and it might be worthwhile for some very demanding applications. I suspect in most cases that another 1 Gbyte of RAM would be more effective than cutting seek time from 5 ms to 3 ms.

Multiple read-head drives

[Kadin2048 (468275)]

I think they used to do this on some very old drives. (Big ones, not 3.5" AFAIK.) Also, when you think about the evolution of hard drives, remember that they came from drum systems where there was one head for every track on the drum, so it wouldn't have been a particularly foreign concept to the guys designing the earliest ones.

IIRC, it's not as effective as you might think it would be at first glance. Although it does help some workloads (ones that are seek-limited), I don't think the improvements were en

Re:

[v1 (525388)]

And yes, I am a seagate/maxtor fanboy. I still have a 1.6gb maxtor from 95 that works fine.

I was a seagate fanboy until 3 months ago. Lets just say that evening I could hear the (2 month old) 500gb seagate in my basement before I put the key in the door. (sounded like a circular saw)

Re:

[edwardpickman (965122)]

I was a seagate fanboy until 3 months ago. Lets just say that evening I could hear the (2 month old) 500gb seagate in my basement before I put the key in the door. (sounded like a circular saw)

It was just happy to see you and welcoming you home.

Re:Its still a toshiba

[nmb3000 (741169)]

Your hard drive was likely still under warranty through Seagate -- did you look into it and see if they would replace it for you?

I imagine that is the least of his worries. When I lost an 80GB drive a couple years ago I would have gladly paid several times the price of a new one if I could only have gotten the contents back. While a free replacement drive might lessen the blow somewhat--as geeky as it might sound--losing a hard drive with gigabytes of content you really care about is a gut-wrenching experience. Everything from my high school days (homework, projects, work, programming, games, music... everything) was gone in one fail swoop.

The only thing similar to it is having your house burn down. Sure insurance should cover it all, but there is no way to get back what was really lost. I suppose if nothing else it taught me the importance of hardware redundancy, though it seemed a high price to pay at the time.

Re:

[benow (671946)]

Yeah, well, backup is a good thing to do. I've built myself a script to find 4G, prompt for disk and burn it, validate the burn and, if burned correctly, nuke the files, eject and move on to the next 4G. It works well, and at >$.30 for 4G it's cheap and fast. Not reliable enough for the paranoid, perhaps. If not, suck it up and mirror a couple cheap smaller drives.

Re:

[benow (671946)]

Heh, sorry, it's not good enough to release (currently broken). I integrated it with my sql disk indexer and it got kind of big. The core is not much more than growisofs and dir compare. The rest is java to sort by filesize and create growisofs command. Perhaps I should get it good enough again.

Re:

[Kijori (897770)]

I wouldn't worry about doing anything so complex if you want to backup your files. Simply archive your filesystem - or as much of it as you can store at a time - and then split the files and burn them to DVDs. Rather simpler and you don't have to hunt through everything to find the files you wanted, they're exactly where you'd expect on the FS after you restore.

Re:

[benow (671946)]

Yeah, the sql indexing is useful for finding a single file among many when mass archiving... rather a different solution than straight backup. Using the sql disk index I had actually built up a Fuse VFS for NSM serial controled optical jukebox. ie: 'cd /jukebox/disk32/; ls; mplayer file.mp3' would set disk32 as the curr, list files from sql index, then load disk 32 and stream the file. It's not far off working, actually...

Re:

[The Master Control P (655590)]

Well, if you didn't mind paying thousands of dollars, you could certainly have had a data-recovery service remount your dead drive's platters with a new r/w head and get the data out...

Re:

[jsebrech (525647)]

While a free replacement drive might lessen the blow somewhat--as geeky as it might sound--losing a hard drive with gigabytes of content you really care about is a gut-wrenching experience. Everything from my high school days (homework, projects, work, programming, games, music... everything) was gone in one fail swoop.

With external hard drive prices these days, you have zero excuses for not having a backup.

I keep everything that must be kept in a separate folder, and I drag that to my external backup drive

Re:

[timmarhy (659436)]

err, i have 2x seagate 200gig sata drives sitting at home, both under 2 years old both of them died with the same symptoms, clicking noise followed by filesystems that wouldn't complete booting.

I'm not the only one with this experience either.

safe to say, i wouldn't be crowing too loud about seagate drives.

Re:

[timmarhy (659436)]

It will be faster. seek time is a major factor in hard drive speed.

Oblig.

[r_jensen11 (598210)]

Sure, you could say that now. But just remember, one Bluray disc will be able to hold that amount of data in a few years. Personally, I think it would be sweet to be able to have all three extended editions of Lord of the Rings in HD and 7.1 audio tracks on one device.

Re:

[Headcase88 (828620)]

And if there's one thing that R&D isn't good for, it's improving upon an imperfection.

Re:

[tomstdenis (446163)]

No. See Density. Last I checked there aren't any affordable 500+GB flash disk systems.

And even still, flash wears out too. So it's not like we have a never-failing non-volatile memory system yet.