Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Technology

How Holographic Storage Works 79

The Chef writes: "Tweak3D.net has yet another excellent article for nerds -- err, I mean, guys looking to fill their brain with technical know-how. This time it's on holographic storage for PCs. Yeah, that's right -- storing files using holography! Go here for the story." This is something that gets mentioned in passing frequently but it's nice to have the technology explained nicely. Thanks for the overview!
This discussion has been archived. No new comments can be posted.

How Holographic Storage Works

Comments Filter:
  • by delmoi ( 26744 )
    haven't we been talking about stuff like this for years?

    Wake me up when you have a product please. I'm sure all of this would be interesting if I was a computer engineering researcher of physics, or if I wasn't expecting exponential improvement for the foreseeable future anyway. But nether I'm not.

    Btw, weren't' we supposed to have florescent CDs by now? Anyone have an ETA on those?

    The majority of those who oppose the death penalty have never been a victim of violent crime -- CNN
  • 2. The vast majority of people have never been victims of violent crimes.

    I KNOW THIS, THATS WHY IT'S FUNNY. Or at least its supposed to be. When I heard that I laughed out loud, since it was such a stupid thing to say. No one gets it, though. So I'm going to start using my next favorite CNN quote: "We don't know how bad things are in north korea, but here are some pictures of hungry children. -- CNN "

    Sheesh, people.

    We don't know how bad things are in north korea, but here are some pictures of hungry children. -- CNN
  • If they achieve a 10Gb/cm capacity like they are planning we will finally be able to have a which can actually store a decent amount of MP3s!

    :)

  • Forgot about the HTML formattint! My post should have read ...

    If they achieve the 10 Gb/cm capacity like they are forecasting, then finally I will be able to have the a (insert favourite MP3 player) that can hold a decent amount of MP3s

  • Even with a 140G CD you're still bounce by DMA data transfer rates. I read about this company years ago but am not much impressed by their public information. Their CFM disk looks alot like a glass master disk. Holographic storage has taken many forms in the past ten years, mostly due to the fact that many of the crystals originally used would only work at sub-zero temperatures (thats Celcius boys and girls) and would degrade very quickly. Your CFMD for the most part looks like a bunch of bunk. Even if it proves feasible which I highly doubt, the cost of a single drive would be extraordinary, just look at what happened with DVD-RAM when it first came out. So 600$ for a new drive and however many dollars for a disk which would be lucky to have the write speed of a CD-R or I can go with a couple 75GB hard drives that have a much higher data transfer rate.
  • Access times all depend on the sort of retrieval you're going to use. One system proposes using sound waves to cause different oscilations in the crystal letting to hit it at different angles with a reference beam. Instead of holographic storage replacing RAM I think we'll see it start replacing magnetic storage. Once refined the components to read and write a holographic cube would cost about as much as a good CD-ROM drive with the cubes not costing too much because they can be easily mass produced. With 10GB per cubic centimeter it isn't hard to imagine larger crystals holding 100GB or more.
  • " I first heard about holographic storage research ten or twelve years ago"

    I think the first mention of holographic storage was 1968. The Hal 9000 computer in 2001 used it.

    If Mr. Mr. A.C. Clark's vision is correct (and he has had some good visions) holographic storage will come of age this year :)
  • Please refer to me as Holography Nazi for the remainder of this post:

    The external hologram surface may turn out to be sensitive to damage (though I can hope for the use of confocal optics, etc. to image the volume despite surface imperfections or contamination) and the volume may be vulnerable to sunlight.
    The external hologram surface may indeed be sensitive to damage. This is why Harddrives come packaged inside of metal cases. Sunlight would be blocked out as well. Additionally, magnetism, which damages non-optical drives, would not harm this drive. Generally speaking, optical storage devices have always been more reliable then magnetic media.
    The hologram technology used here showed promise because it can be multiplexed with different laser colors and at different angles, but the 'clarity' of the signal goes down with the square of the number of channels, until it is unintelligible. This does not bode well for rapid breakthroughs (though if we could predict them, they wouldn't be breakthroughs). Precise alignment is necessary to assure high density, reliable readings. it seems likely that the 2003 holographic drive will be larger, more expensive, and offer no appreciable advantage (aside from ?magnetic insensitivity?, if that counts)
    You didn't learn much about holograms when you made one in 1977, because much of what you said here is wrong. As an explorer, you probably made a reflecting-light hologram, similar to what is on a VISA card. These are inaccurate as the picture changes depending upon the angle and colors of light reflecting upon the foil. Using a transmission-light hologram is much more accurate. When a monochromatic light source (expanded laser) shines upon the film, the original image is created in 3 dimensions. The image is very insensitive to precise alignment of the lasers. The laser can be angled within a range of more then 10 degrees. The film can hold more information then one 3-D image. This is what IBM is refering to. Additional monochromatic light sources can store multiple images in the film, each constructing a 3D image when transmitted through the film. After too many images are added to the film, it can't reconstruct images as well. Think of this as overexposing regular film, it's almost similar to this. The solution to this is not to have many images. One true 3D image (which holograms construct - not false ones like 3D glasses construct) already holds much more information then it's 2D counterpart.
    Aside from the probability of actually seeing a production drive someday, I think that several other holographic technologies are more promising. and none of the holographic technologies show signs of exceeding the practical capacities of straight optical media in the predictable future -- i.e. the next three years. Standards, not technical capabilities, block DVD-R from coming out *this year*
    I'm a little unsure about the practicality of a holographic drive in the near future, but IBM is known for applying it's research to products. The copper technology and advanced layering of ceramic insulaters/semiconductors are examples of this. Holography has already met and surpassed traditional optical methods when it comes to nondestructive testing.

    I recommend Optical Methods of Engineering Analysis by Gary Cloud for more reading in the area of holograms and how/why they work. This book is very applied and mentions many practical examples of holograms in industry.

    Finally, my grammar nazi side is pointing out that you misspelled manufacturing. It's late and I probably misspelled more then just that, so I forgive you.
  • How about "people looking to fill their brain with technical know-how"? I am female and I very much enjoyed this article.
  • I doubt that this thing will be anywhere near a serious contender for replacing my hard drive for a few years, simply due to the fact that holograms require a nearly perfectly stable plane, but I find that this thing looks ipressive for use as a backup medium. You can store an incredible amout of information on a relatively nonvolitile small item, and retrieval times are very fast. Apply some massive redundancy in the encoding process so that you don't waste too many blanks, and you have the perfect medium for doing those daily backups. Plus, a cube a few CM in dimensions looks so much more Sci-Fiish than a spinning disk for storing information
  • I remember seeing articles about this sort of thing in PC Magazine years ago. They stated things like "costing pennies on the GB", and "just around the corner". Yet, it's still not here.

    Granted, many posters here and the article itself have pointed out many legitimate reasons.

  • Byte [byte.com] had this ages ago. :)

    Seriously though I see to recall an article back in the old, dead tree edition. Er, just a sec

    ...Man, I wish all the magazines had archives as complete as the old Byte ones. Anyhoo, they have a couple links about holographic storage there too:

    1996 [byte.com] and 1998 [byte.com]

    Not bad, but essentially saying the same thing as the tweak3d article. Some of the other, non-holographic versions of storage sund like they may actually see the light of day first.

  • equipment for the lasers and mirrors could be quite large, and while they might shrink it, it would still be large compared to one cm^3. But I bet you could use the same equipment to address every cm^3 in the hard-drive form fatcor.

    No more disk/RAM

    Let's face it: we have RAM to execute programs and disks to store data and programs permanently. When you have something as potentially fast as holographic memory (assuming it boils down to no moving parts), you have the speed of RAM combined with the capacity and persistence of disks.

    What this means is the whole paradigm of modern operating systems shifts. Your programs will maintian state between power cycles, files won't have to be read into memory (they can be used in-place), etc. The whole thing takes over "virtual memory". (This is sort of how the PalmOS works with programs maintaining state)

    This is not unlike the paradigm 64-bit memory addressing boasts (where you could map the entire drive into a memory address range and make dealing with disks a whole lot simpler).

  • I don't know what anyone would ever do with a hard drive bigger than 10GB

    Probably this doesn't deserve much of a response, but... let's see. Currently at home I'm working on a GIS (Geographic Information Systems) project which relies upon large aerial photographs that averge 250MB each. It doesn't take long to fill 10GB at that rate. My girlfriend is an artist and she regularly generates intermediate Photoshop files in the 20MB range. In the office, I regularly deal with databases which require several GB of storage space.

    So, for what it's worth (which is probably not much), there you have it: real-world evidence that people really can take advantage of large drives.

  • So if you had a CD-diameter disc, 1cm thick, with say a dozen lasers constantly illuminating a fixed area on any given cylinder therein... You could have full-motion real-time immersive 4-d senso-vid!

    Or like 27432 x 10 ^ 32 full-length MP3's :-P
  • IBM is known for applying it's research to products.

    As a grammar nazi, you ought to know that this is supposed to be "its", not "it's". :-)
    --
    Patrick Doyle
  • no, you've got it backwards. A geek is someone who is awkward or weird. Such as those interesting people in circus sideshows (like the human pincushion and whatnot), they were sometimes called geeks. I'm not sure exactly what a nerd is, but I think its what you described to be a geek.
  • One of the issues with holographic storage is throughput speed. You can store a tremendous amount of data, but unlike a CD-ROM, where the medium moves very quickly and the laser is relatively fixed, holographic storage is relying on moving lasers throughout the medium, which is inherently slower for the time being.

    The trick will be to find a way to use many lasers at various points to illuminate the storage medium at near simultaneous speeds. It's essentially parallel retrieval, but it will speed things up.

    JHK
    CASCAP, Inc. [cascap.org]

  • by NaughtyEddie ( 140998 ) on Monday June 26, 2000 @04:03PM (#974613)
    A lot of people are asking why we don't have this technology yet.

    One possible answer is because of the sensitivity of holographic equipment to vibrations. A hologram encodes phase differences between laser beams. Errors in the phase encoding mean errors in the data retrieval - you get a blurry or disjoint hologram, or you lose your data.

    Light is in the hundreds of nanometers range of wavelength. This means a vibration in the equipment (a movement of one part relative to another) of only a tenth of a micron can completely throw the phase encoding out of alignment. Imagine a tape deck whose heads needed positioning to submicron precision.

    Making holographic images is therefore rather difficult if, say, a large lorry rolls past your window. A hard-drive with the same problem would be absolutely useless.

    So until a suitably hard substrate can be found on which to engineer this equipment, it's only a pipedream. Maybe nanotechnology will create such a material ... I doubt it'll happen before then.

  • by Anonymous Coward
    A major issue with holographic storage is throughput speed. One can store a tremendous amount of data, but, unlike a CDROM (where the medium moves very quickly and the laser is relatively fixed) holographic storage relies upon moving lasers, which is inherently slower.

    The trick will be to find a way to use many lasers at various points to illuminate the storage medium at near simultaneous speeds. It's essentially parallel retrieval, but it will speed things up.

    Tupac

  • Knock the monkey flat on his ass next time. Junkbusters. [junkbusters.com]

    No, I don't work for them, yes I use it, and yes, it kicks major ass. I haven't seen an ad in months.
    --
    -jacob
  • by konstant ( 63560 ) on Monday June 26, 2000 @04:09PM (#974616)
    Wonderful article. I'm so glad /. still posts stuff like this occasionally.

    The author mentions, enticingly, that the potential of the technology is to store 10 gigs or more in an area roughly the size of a "single gambling die". This, clearly, is a fantastic dream.

    Regrettably, the real problem that the article doesn't really touch is the space and more importantly, the precision and energy, required by the laser that is needed to read and write to the medium. Just glancing at the interior of my relatively rudimentary CD-ROM drive, I can see that its mechanism consumes considerably more area than a die. And it doesn't even rely upon the sophisticated network of lenses described in the PRISM research project.

    You all know how inconvenient it was/is to transport a CD player through rough terrain and expect it to work continuously. Imagine trying to get any kind of ruggedness out of this badass!

    However, 10 gigs smaller than the last joint in my thumb.... yum.
    -konstant
    Yes! We are all individuals! I'm not!
  • I can confirm this. I recently went into the future (simple: I just travelled at .99c for 2 years - the hard part was getting back again). This was happening as a daily occurence, just as you describe. It was fantastic. I spent 30 years in the future collecting all the albums I ever wanted to own (full precision, too - none of this MP3 crap) and writing a fantastic book about the whole experience. It was great.

    Unfortunately, my girlfriend dropped my holocube into a cup of tea, thinking it was a sugarcube, and I lost the lot.

  • Finally a really good article about holograms in general and holographic storage specifically. Most people are not really aware how close many companies are to production of holographic storage. While magnetic hard drives are getting really dense they are still overly mechanical which adds greatly to their price. A holographic drive that would fit into a 3.5" internal bay could hold easily more than 100GB of data and have few or no moving parts. This means a greater mean time between failure and much less power consumed by the system (so we can power our GeForce 7s and Voodoo 9s). Besides desktop and server storage you could have a shitload of information encoded into a small plastic chip on a driver's license or credit card. The next generation of smart cards could carry user preferences for various computers, personal files, and cryptographic signatures. On top of that you can stick holographic drives in TiVo like toys or portable MP3 players. A single memory card the size of a sony memory stick (which is some of the best looking portable media ever) could hold a week's worth of music or an entire HDTV quality movie. How about a Palm XV with a gig of storage, that would be something to show off at an office party. Wow, I seem to have wet myself.
  • I think its time the penguins said-So long and thanks for all the fish-
  • what the hell is 'holography?'
    Sigh. I'm sure that this is a troll, but I have to do my duty.

    holography
    Pronunciation: hO-'lä-gr&-fE
    Function: noun
    the art or process of making or using a hologram

    And because I know that you're gonna ask:

    hologram
    Pronunciation: 'hO-l&-"gram, 'hä-
    Function: noun
    a three-dimensional image reproduced from a pattern of interference produced by a split coherent beam of radiation (as a laser); also : the pattern of interference itself

    All definitions are taken from Merriam-Webster's Website [m-w.com]. It's ironic that I steal MP3s but I still try to reference information that I use. Can anyone explain those morals?
  • I rest my case.

    -----
  • Holographic storage will be/is damn close to instantenous... you'll still need ram though.

    -----
  • I first read about this technology in the September 1990 issue of Byte magazine ("Experimental holographic system promises massive data storage, rapid access", pg 24 - 25). Back then, they were still trying to figure out how to perform nondestructive reads from holographic storage.

    And they were suggesting that within 5 years, we'd all have gigs of this stuff on our desktops. I guess that's what happens when you let managers decide on project release dates instead of the engineers who are actually doing the work!

  • Polls showed that the common public was not interested in video phones at all so only one company in Germany dared to try and make them. Apparently it did not go that well for them.
  • They could use magnetic suspension to damp out some of the major vibrations the way telescopes do.

    You probably wouldn't use a strictly mechanical system to accomplish this degree of accuracy. Trying to damp the lenses so that they don't jitter even a nanometer would be impossible. They are probably modulating multiple beams to adjust the interference pattern.

    Using a closed loop system that constantly tracks and adjusts the aim every couple of nanoseconds would resolve the rest of the vibration problems.

    If anybody has any good links to closed loop technology please post. I have never found any good information on this stuff, probably due to people protecting their research.
  • Ok, so who is the supplier for Blue-Green SolidState Lasers under 50$ ?

    I'd imagine no one right now, but there isn't that much demand either. And you're missing the point with your DVD comparison: holographic storage is (or at least has the potential to be) *really* fast, with (most importantly) no moving parts.

    And yes, costs will eventually come down. Remember, cds weren't cheap in 1985 either.

    -----
  • I mean cd media and hardware. Drives cost in the tens of thousands and pressing cds was incredibly expensive when cds first came out. Now you can buy a 30x cdrom for $20, and companies press cds by the thousands for 10 cents a piece.

    But geez, 18 pounds? That's almost $30! Is that for a music cd? And I thought they were outrageous here!

    -----
  • As you may have guessed, I flunked materials science ;)
  • What type of closed-loop technology do you mean? General Systems Theory includes a lot about closed loop. The CD tracking mechanism uses a similar feedback mechanism to keep its head on track - this is really very clever, but I read about it in a book so I don't have a link. The basic principles of error-controlled negative feedback can be found in the basic op-amp circuit, too. And even sound compression technology utilizes similar ideas. Does any of this help at all?

    The main problem I can see with using a servomechanism to damp vibration is that the tiny vibrations happen very happen, and a mechanical damper that could react quickly enough would be hard to engineer. Maybe active magnetic suspension would be the way to go?

  • by Anonymous Coward
    i dont need more storage right now. what i need is fast storage. instantenous accesss to lets say 100 GB. yeah, that be sweet. no need for RAM -- super fast storage.
  • but a breakthrough nonetheless. 48 MB/cm^2.. but 10 GB/cm^2 possible. Desktop units by 2003.
    Looks like IBM is breaking down the doors of high storage yet again. Their 45-75 GB drives are impressive, but this? Wow.
    Access rates look pretty nice long-term too.. "a gigabit-per-second data rate appears reasonable for holographic storage".
  • Ok, so who is the supplier for Blue-Green SolidState Lasers under 50$ ?

    Not to sound like a nay'say'er but this is an expensive and complex solution. Take an Epoxy Cube of Layered Alumn'/Seaweed (ala DVD) and What is your Total MB^3cm? Is it cheaper?
  • I'm terrible when it comes to physics with light, and I was even able to understand that. Really read the article.
  • You know, I'd love to see this as much as the next guy, but isn't this a tad like every other story we've seen that promisses us a technological marvel "just around the corner" so to speak that never arives? It's interesting, but I wonder if we'll ever actually SEE one? I've been hearing about this kind of optical storage for at least 5 years now, and have never seen a single bit of technology make it into production. Has anyone else?

    Fawking Trolls! [slashdot.org]
  • WHEN THE HELL DO WE GET THIS TECH?! I first heard about holographic storage research ten or twelve years ago, and they had ALREADY figured out how to put a gigabit of data in a one-centimeter cube. If this has followed the same curve -- hell, even a smaller curve than normal magnetic hard drives have, then we should be popping terabit plastic cubes into our computers by now, with ridiculous access speeds in the microseconds, and so on and so forth.

    So my question is, WHAT'S TAKING SO DAMN LONG?! Can someone just PLEASE come out with a holographic storage drive already? Fer chrissake.

  • By the way .. a standard cellophane roll was found to be able to contain 92 Gigabytes. Impressive.

    Wouldn't this have more to do with the way the information is layed down (metaphorically speaking) rather than the actual physical material used?

  • Perhaps in the next five years we will see the advent of the non-spinning drive, and portable disk space approaching the petabyte. Enough storage to keep most of what we now consider to be human knowledge.

    Maybe it's enough to keep most of human knowledge on it, but I bet M$ can still come up with an OS that will fill most of that storage capacity.

    Of course this has been predicted for a while. OD course they kept saying we'll have it soon and then not coming through. Same with a lot of stuff. Still, the engineers usually get it right in the end. But it's always later than they said it would be. 2003 seems a little optimistic t'me.

  • by IAmATuringMachine! ( 62994 ) on Monday June 26, 2000 @04:18PM (#974638)
    The slashgods repeatedly rejected a submission about a scientific american article that talk about upper limits on magnetic storage and puts forth a marvelous discussion of holography (http://www.sciam.com/2000/0500iss ue/0500toig.html [sciam.com]). My question is, why this article and not the other? I personally preferred the Scientific American article.
  • You can already fit every piece of music ever written into a few gigs <sic>
  • haven't we been talking about stuff like this for years?

    1.Sure have. However, keep in mind that the astronomical speed at which computer tech has been developing for the last 30+ years is all quantitative growth. The basic concepts of hardware design haven't changed dramatically, we can just do it bigger, faster, cheaper... What we're looking at with holographic "disks" is a quantitative change... and that's going to be slower. It's like saying back in 1920 "We've had super-fast trains for years, and look how fast the automobile is developing... what the hell is taking those aeroplane guys so long?"

    2. The vast majority of people have never been victims of violent crimes.

    3. The vast majority of people who support the death penalty have never been falsely arrested/chagred/convicted of any crime.

    4. Columbia University Law department study of all dp convictions since 1977 shows 68% were "seriously flawed" (read: mis-convicted) cnn blab here

  • http://slashdot.org/articles/00/04/11/0832216.shtm l

    I just did a quick search in the slashdot engine (available on every page) for "Scientific America" -- even with the missing N I got that one, and from the headline (The End Of The Road For Magnetic Hard Drives?)I thought it might be it.

    So ... that's why. :)

    timothy
  • Well... it at least seems simpler than Rambus. If they just let the Taiwan folks make it, we'll have it cheap.
    blessings,
  • well dip me in shit and call me stinky :)
  • Main problem I see is that for now it would be read-only, like CD's... and we already have DVD storage. Still though, these would be great for movie storage. You could store the entire star wars trilogy in 3 different languages, THX and the original analog audio, all on one cube.
  • Holographic storage, it's a science fiction, as of now, What more can I say.

    I worked on it in Grad. School 8 years ago. It seems 8 years passed by without significant progress in this field.

    One major limitation is that data per page is limited to the resolution of the LCD display. The most economical supply was those computer projectors. We took it apart, tore off its polarizing plastics, and all we could get is 640x480 (maybe they make higher resolution now). You can spend BIG money and get a special designed LCD SLM, it has higher density, but they can't make it bigger. There are limitation of numbers of angles (or wavelengths) you can use to store different pages (in reality, maybe around 20, best under 100), so data size per page is really important.

    Another problem is the recording material. The commercial read-out device has to be small diode laser (like CD-ROM, laser printer), so the data has to be recorded in the same frequency - RED - means low energy (like, black-and-white dark room can be illuminated with dim red light?). I never heard of anyone produced a material that could record under red, or anyone made a blue diode laser.

    On the side line, if they could make R/G/B diode lasers, then we can make super-sharp projectors. TV or monitor... now that's something I want.

  • by orpheus ( 14534 ) on Monday June 26, 2000 @04:40PM (#974646)
    I do not in any way mean to detract from the accomplishments presented here (though, of course, it's been presented in varying shades of "Any time now" for decades. I made my first hologram as a Explorer scout in 1977 or so, and at that time it was already a staple of such classics as "The Adolescence of PI")

    However, the storage capacity cited is 10GB/cm^3 cubic centimeter) not cm^2 (square centimeter) as you stated. By comparison, given how thin the magnetic films are, I doubt the 75GB HDD even has 1 cc of active storage volume so holograms do not approach the volumetric density of magnetic media.

    The key of course is that holograms offer the promise of true volumetric storage, where magnetic media is limited to the thin film [though who can forget the hedelberg group who used a thin film -- namely a roll of commercial adhesive tape [villa-bosch.de] to as an optical medium for up to 10GB a few minths ago?)

    So, since it comes down to form factor, I'm not excited. The problem with the HDD is the overall associated mechanism, and with 1 GB matchbook 10G-resistant HDDs out there *today*, I'm not sure when I'll ever be excited. There is no reason to expect we'll be carrying 'naked' (or packaged) holographic media, any more than we carry naked (or packaged) *high density* platters today -- and holographic drives may well be larger, more expensive, more fragile, etc. than HDDs in 2003, as well. In 2003, you won't be able to *buy* a new HDD as puny as 10GB, if indeed they are still maanufacturing that size, today.

    The external hologram surface may turn out to be sensitive to damage (though I can hope for the use of confocal optics, etc. to image the volume despite surface imperfections or contamination) and the volume may be vulnerable to sunlight.

    The hologram technology used here showed promise because it can be multiplexed with different laser colors and at different angles, but the 'clarity' of the signal goes down with the square of the number of channels, until it is unintelligible. This does not bode well for rapid breakthroughs (though if we could predict them, they wouldn't be breakthroughs). Precise alignment is necessary to assure high density, reliable readings. it seems likely that the 2003 holographic drive will be larger, more expensive, and offer no appreciable advantage (aside from ?magnetic insensitivity?, if that counts)

    Aside from the probability of actually seeing a production drive someday, I think that several other holographic technologies are more promising. and none of the holographic technologies show signs of exceeding the practical capacities of straight optical media in the predictable future -- i.e. the next three years. Standards, not technical capabilities, block DVD-R from coming out *this year*

    Hey, I want my petabyte encrypted keychain as much as the next guy... but, you know, 'fire' still has many unparalleled uses, in the nuclear age. And I'd rather grill than irradiate my dead cow this weekend anyway
  • You are right, but on the wrong ground. Yes, recoding hologram need absolute stillness, preferably the setup is in a basement, and on a air floated table.

    But that's about writing data, you really can't selectively change only one or two bit in hologram anyway. It will be more like CD-ROM thing. The recording material (a crystal at the size of a sugar cube) being manufactured under high power (bigger) laser and you use low power laser to read it.

    And it's a fantasy anyway.
  • What you read probably is from Scientist's report to Pentagon so they can get more money to fund their graduate student and publish papers, which enhance their fame and status.

    And they knew it's a unworkable technology.
  • Ah about the No moving parts parts bit...
    Optical storage is not SolidState nor is it Non-Mechanical.
    The "Feature" here is Density not zero moving parts. Speed is/could-be great when the "head" is on the data and assuming a lot of preconditions.

    Cheap CD's are based on the Cheap LED-lasers. Sorry the media could be armpit hairs or gold it does not matter to the "Effect" just how well it works.

    I hope you read some of the 3+ posts as they point out these issues much better than I.
  • The tracking ability of a contemporary CD player to track and error correct the tiny data path on a wobbling Compact Disk is amazing. And yes your examples make the general concept of optical feedback mechanisms easier to understand.

    No mechanical servo mechanism is going to be free of vibration at this accuracy. As is the case with Compact Disk, error correction plays a big part. The logical path for optical storage would be an extension of the existing spinning disk technology only making the CD tracks multilayer.

    The precision that this article seems to imply seems to go way beyond that. If you used active magnetic mounting you would have to run the whole device in a vacuum for it to work otherwise even sound waves would cause vibration. This approach sounds too much like work.

    You could store data in stationary cubes rather than use a rotating disk and servo mechanism to move the laser along a path. I imagine it would be easier to focus and modulate a large matrix of stationary lasers. Feedback loops and error correcting controled by a very fast clock could take care of any vibration.

    This shifts the emphasis of isolating components from vibration to quickly adjusting and error correcting in response to feedback. Theoretically you could use a cube of Jello on a paintshaker to store your data using a closed loop optical system.

  • I don't know if there's a systematic mistake in the article, I would have estimated the GB/cm^3 ratio much higher. Consider this:

    A plain DVD stores 5GB on a one-dimensional area. If you look at a hologram, you might get around 100 angles with different information... and that's still a flat hologram! So you could store the information of approximately one DVD just by taking an image of a cm^2 of its surface for each angle allowed by the hologram. So in my opinion the theoretical limit should more be at 10GB per cm^2 than 10GB per cm^3.

    If I remember right, I read an article in a local newspaper about this, where they were able to read out one flat area out of a cube (technically, by applying a reference laser beam only in one layer of the crystal). So we could have hundreds of layers in a cube, which would more be like a terabyte in a cm^3.

    if the physical limit was around 10GB/cm^3, why would they spend that much money on it? We would just wait until IBM has its Flash-Drives out with a GB, and there's no optical stuff involved.

  • strangely enough,

    I figure there are plenty of side show freaks with GREAT people skills.

    I still think a nerd is personality inept.
    The modern definition of geek, to me at least, implies interest in cerebral, scientific stuff.

    A nerd wears tape in the middle of his glasses and takes his sister to the prom.
  • Just as a matter of interest, have you ever run a CD player with the top off, and watched the lense float over the surface of the CD? It's an amazing sight.

    And yeah, with error correction you can handle any amount of environmental noise, but in return you lose data storage to redundancy. It would be a shame if the only practical implementations of holographic memory had so much redundancy that the actual data storage capacity was reduced back down to hard-drive levels ;)

  • Dude, there's a difference between NERD and GEEK.

    I don't mind being called a geek... it just means my interests are eccentric compared to the mind-numbed hoards.

    But a nerd implies that I don't have any people skills.

    Anyone else know the difference?
  • "Remember, cds weren't cheap in 1985 either."

    Huh! Its 2000, and cds cost 18pounds in the u.k.

    :)
  • The word is "geek" you politically correct bastard.
  • I knew what you meant, i was being sarcastic... ;)

    but yes, a single music cd. I wont go on about it because its been discussed before here :) And you can get them cheaper, sometimes, from independant stores (rather than Tower, HMV etc).

    Sort of puts you off of impulse buying though.

    a.
  • You can't be using MP3's then... all my music use up 5-6GB. Granted, when I encoded it, I didn't use 128kbit... but still, getting ALL the music ever written onto a few gigs would be quite a feat (unless you discount 90% of the music out there as nothing but a clone, which with today's pop music wouldn't be too far off :)
  • Yeah! I want my video phone, my hover car, and my electronic windowshades! Let's not forget the bionic implants, paint-on monitors, etc.

    After all, it _is_ the year 2000.

    On a serious note, holostores have been worked on forever. You need to be a bit more patient. My book on digital storage predicts they'll be in widespread production by 2015-2020. This seems reasonable.

  • by Vladinator ( 29743 ) on Monday June 26, 2000 @03:12PM (#974660) Homepage Journal
    Maybe so, maybe so. Can we get some articles about SERIOUSLY cool stuff that will occur in this decade?

    Fawking Trolls! [slashdot.org]
  • Don't forget the requisite comments about more MP3 and p0rn storage...
  • by fluxrad ( 125130 ) on Monday June 26, 2000 @03:19PM (#974662)
    is that you have to outfit your computer with special paper 3d glasses. one side red, one side blue!


    FluX
    After 16 years, MTV has finally completed its deevolution into the shiny things network
  • Just glancing at the interior of my relatively rudimentary CD-ROM drive, I can see that its mechanism consumes considerably more area than a die.

    Take a look at the mechanism in a laptop drive, or a walkman, or a minidisk player.

    Desktop drives only come in two sizes: big, and bigger. Since CDs are rather wide, CD-ROMs fall into the "bigger" category. It's cheaper and easier to make the electronics big, so they use the available space.

    If your drive had a smaller (and more expensive) laser thingy, you'd be complaining at the wasted space.


    My mom is not a Karma whore!

  • I give you proof: Comments #12, #2, #11, #15, #17, #16, #13, etc.

    The only comments so far that did not fit into these categories are ones that say "Even I understand it!" And this while a bunch of trolls are still at 0 or 1 (although they will probably be at -1 by the time you read this). Please, moderators, spend your time shoving down the trolls before wasting points on borderline stuff that you may not like but someone else might.

  • If that's the case then hardness isn't really the issue, stiffness is. Stiffness is a close relative of hardness... diamond is a heck of a lot stiffer than steel.

    One way to engineer it would be to use the hard (stiff) substrate that you speak of, and combine it with a damped enclosure.

    The difficult part may be in determining what the response of the system would be. The read mechanism would have a different elasticity than the storage medium... what you would have to do is couple them stiffly (vague enough for you?), and add an external shock damping mechanism... two lines of defense.

    -- Phenym
  • by Wah ( 30840 ) on Monday June 26, 2000 @05:09PM (#974666) Homepage Journal
    The, um, backup crystal is almost full...

    inspired by this paragraph.

    However, as you keep recording more data pages slightly away from previous pages, the holograms will begin to appear dimmer and fogged up because their patterns must share the material's finite dynamic range and the data page is physically etched into the crystal. Eventually you will run out of space to store because the crystal has depleted all of its physical storage capacity, sort of like write once, read many media such as CD-R.

    This is how they get so much data, you can shift the angle ever so slightly and have a new canvas. Mix this with a billion nano-bots with flashlights and miniature crystal-zamboni's and you've got some serious disk space.
    --
  • I see a lot of info about information density and price points, but what about access times? How long can we expect data requests to take? I understand that this is all still in its infancy, but I'm just interested in orders of magnitude. I would assume that it's faster than magnetic media requiring moving parts, but how does it compare to RAM and other silicon storage?
  • I was making holograms in my undergraduate physics
    courses. And there are explanations of them in
    even basic optics books. But since this is /. and
    you don't need to take even basic physics to be
    a coder I guess most /.er haven't heard much
    about them. BTW they knew about holograms before the laser was invented. Altough noone could make it work until the laser did come along.

    Anyway, I am surprised they didn't mention the coolest
    thing about holograms. You can physically break a hologram
    in two and you will have two copies of the same
    hologram (although there will be a lose in quality.)
    Photographs store information about the image --
    holograms store information (phase and intensity)
    about the light that bounces off the image. Cool
    huh? even if they aren't useful everyone must agree
    that they are seriously cool!
  • Install this on the spaceshuttle Discovery and program it to do missions.Can we spell H.A.L.9000? (Good Morning Dave)
  • i remember reading about this in high school (3 years ago) in a book published in 1982. IBM is putting a lot of resources into this. and if you think about it, the payoffs are tremendous... all of the worlds computers using data storage created by IBM

  • The main problem with holographic storage is creating a usable and stable reader/writer. People have managed to create holographic storage devices from such fanciful things as spinning glass rods and tanks of supercooled gel ... to such ordinary things as a roll of clear cellophane tape.

    By the way .. a standard cellophane roll was found to be able to contain 92 Gigabytes. Impressive.

    There have been some recent advances in fixed holographic storage, which would allow a 1-6 terabyte hard disk to be made, with no moving parts. However, the cost/gigabyte is still well over that achievable by magnetic media (but under that of copper memory).

    Perhaps in the next five years we will see the advent of the non-spinning drive, and portable disk space approaching the petabyte. Enough storage to keep most of what we now consider to be human knowledge.


    Reach out, extend to, and embrace the universe.
    -Einstien
    -----
    Embrace, extend, and engulf the universe.
  • by Anonymous Coward
    Given enough density of storage... and cheapness
    of availablility.....

    I wouldn't mind designing machines with a 'Brain'
    So to speak....

    Think of a computer that's just cpu, ram(??) and
    your basic graphics and I/O devices...

    Drop in your personalized "brain" cube and you have
    your complete home away from home system (that
    you burned to cube for just this purpose) at any
    convenient 'puter Kiosk...

    If you figure they can pack 140G's into a 2 inch cube... (though wonder if that might be too thick?) that would allow you to customize
    your linux (or other os if they figure it out)
    setup to probe the type of machine and boot up accordingly...

    It would also leave room for more storage...
    Wouldn't even have to be read/write.. *shrug*
    At least not for daily stuff....

    All this hype about .Net?? Screw that... I'm much
    more inclined to carry all the
    info/applications/etc.etc. I need WITH ME...

    Not leave it sitting on _your_ server.. *shrug*
  • Some Troll seems to have temporarily "borrowed" my account to post some rather OT messages (trolls), I've changed the password so hopefully that's the last we'll see of them...

    Sorry about that guys.

    --
    Jon.

Disclaimer: "These opinions are my own, though for a small fee they be yours too." -- Dave Haynie

Working...