A Terabyte of Data on a Regular DVD?

Roland Piquepaille writes "This is the promise of the 3-D Optical Data Storage system developed at the University of Central Florida (UCF). This technology allows to record and store at least 1,000 GB of data on multiple layers of a single disc. The system uses lasers to compact large amounts of information onto a DVD and the process involves shooting two different wavelengths of light onto the recording surface. By using several layers, this technique will increase the storage capacity of a standard DVD to more than a terabyte. Read more for additional references and a diagram showing how this two-photon 3D optical system reads data."

Re:maybe Im not getting it

[itlurksbeneath]

Much higher density and multiple layers. TFA mentions 33 x-y planes (layers) of information. With that many planes, the density of each layer is comparable with a single side of a Blu-Ray disc. Can't remember if Blu-Ray is multi-layer or not.

Article is wrong

[Andy Dodd]

By the definition of a DVD (yes, just like the various "color" Book standards that defined CDs, there are standards that define DVDs), this new technology will not result in a standard DVD by any means.

More proper terminology might be "in a standard form factor 12cm optical disc".

Firehose Data Rates

[goombah99]

The data rates when reading or writing one of these would be pretty high. If there were 1000 tracks then that would be 1GB/track if it were spinning at 3600 RPM that would be arounf 500 Gigabits per second. You'd need sever times that in electrical bandwith to keep the squarewave sharp. That's terrahertz modulation rates even for the shortest reads.

So to do this at all your going to need 100 or more read heads and data channels to get the modulation rate down, or there would have to be orders of magnitude more tracks. Or possibly there's some way you could encode the bits in different overlapping frames such that the data rate of any one frame was lower. For example by using different reconstruction laser spatial patterns for different frames could use physics to select which frame was being selected.

Otherwise this is drinking from a fire hose even for the shortest reads, and the equipment needed would be prohibitively expensive.

The same problem happens when writing: how do you buffer a gigabyte of data to deliver it that fast. It ain't gonna be in the main RAM.

What about HVD?

[stonesmith]

HVD is supposed to be out as early as next year and have 3.9 terrabytes of storage.

Not a standard DVD

[NiteShaed]

The summary says "By using several layers, this technique will increase the storage capacity of a standard DVD to more than a terabyte.", yet UFC's website offers the following description:
"Depending on the color (wavelength) of the light, information is written onto a disk. The information is highly compacted, so the disk isn't much thicker. It's like a typical DVD."
A disk that "isn't much thicker" than a standard DVD isn't a standard DVD.

Re:light on details...I'm a skeptic

[drinkypoo]

and finally, due to the nature of holography (in that small sections of a hologram contain the information needed to reconstruct the entire hologram), a disc with holographic storage should be much more resistant to read errors resulting from scratches, whereas with one of these, a scratch could render data on several layers unreadable.

You are incorrect - you're almost right but your interpretation of the durability of a hologram is unfounded.

Small portions of a hologram contain the information needed to produce an approximation of the original image. The difference between traditional and holographic storage is that a scratch on a CD renders the information under the scratch unreadable, while a scratch on a hologram degrades the entire image.

In other words, you lose just as much data, it's just unevenly distributed. In the end, it will help you with durability by making it so that a certain percentage of the disc must be damaged before the data is unreadable; but at the same time, if you start with a 10cm square hologram, and you want to be able to still read the data faithfully if you only have 1cm square area left, your data will have to be written across 100x the area that it normally would in order for you to be able to read it out later.

If a 700MB CD without ECC is 800MB then an audio CD is ostensibly one-eighth error correction. Assuming the same density, you would get the same amount of data on the CD, but you would still be able to read data from any part of the CD as long as no more than 1/8 of the media was destroyed. In theory you could drill some symmetrical, balanced holes in such a CD (assuming a rotating-media holographic system, which is probably not a safe assumption) and lose nothing, not even the data you punched out.

Anyway, the REAL problem with optical disc durability is that the top layer is vulnerable. Scratches on the bottom can be polished out and minor scratches don't even have a significant effect because the laser is focused on the metal layer, not on the disc surface. It's diffuse when it passes through the layer where the scratches are. If the top of the disc were protected, I'd probably have lost about 50% less discs. I just had to throw about five discs away because their metal layer stuck to my CD binder and peeled off... And the first CD I ever killed died because I laid my arm across it for a couple minutes and sweated on it, which caused big chunky pieces of the metal layer to delaminate and stick to my arm like gold flakes.

Re:A Terabyte... For How Long ?

[smallpaul]

Microwave ovens are made to be ovens. That the idea arose because of radar systems is irrelevant to this discussion. You're right that CDs were intended for audio. I would argue that they are not great for computer systems precisely for this reason. Unlike the old floppy disks, CDs do not do random-access writes and are not covered by a dust and dirt-blocking shell. On the other hand, standadizing on a non-optimal solution has had the great advantage of making CDs and CD players cheap as dirt, as well as allowing computers and audio systems to share music. The original poster in this thread presumably values reliability over cost savings based upon his negative experience with CDs.

Re:A Terabyte... For How Long ? depends on $

[denis-The-menace]

Maybe it's the quality of media that you buy.
look up your media here to see how it rates.
http://www.digitalfaq.com/media/dvdmedia.htm [digitalfaq.com]

Re:maybe Im not getting it

[evilviper]

is there actually any significant degradation when you use the laser to read, or is the power level so much lower that there's really just not enough energy to cause it?

No. There is some degredation of the dye with exposure to light, but with a low-power reading laser, it's so extremely small as to be negligable.

There's much more energy in a few seconds of exposure to sunlight than in numerous full reads of a CD-R.

Re:maybe Im not getting it

[Anonymous Coward]

Um...parent is wrong, 1024 GB is not a TB, 1000 GB is a TB, don't let idiots who think 1024 is "near enough" get control of a well defined magnitude.