140GB CDROM: Fact or Vaporware??

[paulw]

Hmmmm... A video delight if true.
http://www.theregister.co.uk/991130-000011.html

[DrMordrid]

I've heard of ths technology before and it's got a great future IF they can pull it off.

IF they do there will be some very PO'ed members of the DVD consortium, not to mention the current makers of removable media. ;-)

Another technology involves a cube of crystaline material that is read and written to by quartz crystal aimed lasers. The theoretical capacity of this one was stated to be about 1.5 TeraBytes per cubic cm.

Dr. Mordrid

[Marshmallowman]

How cool, they are transparent.
Labels might been a pain tho

[Patrick]

Wow, can you imagine a variation of this technology being used in a camcorder?

Screw the MiniDV vs Digital8 debate, eh Doc?

[dchip]

DrM, is anybody (company) out in front in the development of that storage medium you describe, or is it in R&D with more than one company? I also like to do a little investing now and then.

[DrMordrid]

That came from a Scientific American article about two years ago. I *think* it was IBM.

Dr. Mordrid

[gas]

FACT!

[gas]

FACT!

------------------

[gibverse]

I'm just going to say this. The Register is about 95% bullshit hot-air, and 5% fact. I have read about this elsewhere, so it may be fact, but I wouldn't count my chicks before they're hatched.

John

[jrr]

There is a bit more information/propaganda at
http://www.c-3d.net/tech.htm

If they are using a 1mW 640nm semiconductor laser and each layer re-emits 0.1% of the light at various IR wavelengths from say 1000-1500nm then there are certainly detectors on the market that can work at these sensitivities and speed.

Oh and Brian, by the way there is an infra-red laser in the Lab next to mine which is passed though some quantum optics to convert it to green. These optics wouldn't work on a consumer level as they only work when the instensity is high enough for 2 IR photons to excite an atom in a short enough time before it re-emits a new photon with double the energy. (The laser we have is 40MW compared to a typical CD with is less than 1mW)

James

[DrMordrid]

Yes, we do indeed have to take into account quantum effects when we analyze things these days. Common sense breaks down in Heisenberg & Schroedingers little world.

Did you see the one where some researchers directed microwaves into a solid block of brass and the signal came out the other end ater a time interval that indicated a superluminal speed of information transfer? They're still trying to figure that one out. Most likely quantum tunneling?

Dr. Mordrid

======

Q: What's the difference between a quantum mechanic and an auto mechanic?

A: A quantum mechanic can get his car into the garage without opening the door.



[This message has been edited by DrMordrid (edited 07 December 1999).]

[Brian Ellis]

Taken at its face value, as reported, it is a load of male bovine excrement. Any scientist will tell you that fluorescence is a phenomenon whereby a substance stimulated with a short wavelength light (e.g. UV) emits a longer wavelength light. It is physically impossible to do the reverse, as its a matter of the energy state of the molecules. As the "simply adapted" CD drives are stimulated by an infra-red laser, which already has a very long wavelength, this means that the fluorescence (if it exists: I know of no compound which will fluoresce under IR) must be of a longer wavelength. This would probably mean that the optical resolution would be largely insufficient to record that amount of data on the few tens of cm2 of a CD disk, even over a number of layers which would fluoresce different IR "colours", necessary to differentiate the "tracks". Also, there is no indication how the write process modifies each layer independently. In addition, fluorescent materials must be opaque to the incident light which causes the fluorescence. It IS possible for some transparent glasses to fluoresce under UV, but only because they are opaque to the UV wavelength. I would be hard put to credit that you could have several different materials in layers, all of them transparent to visible light. Incidentally, if the fluorescence is IR, as is suggested, then the materials would most likely be opaque to visible light, as well, because there are very few substances which are opaque to one wavelength and significantly transparent to a shorter wavelength: that would involve complex resonances.

I'm not saying that I'd take this report with a pinch of salt, but - pending further scientific clarification - there is not enough salt in the Dead Sea to make me believe, at this time. If anyone wishes to invest in it, I respectfully suggest they use only fun money.

------------------
Brian (the terrible)

[Brian Ellis]

James

Thanks for the info. I have read it carefully and am still more than sceptical. I hope I'm wrong, but I'll not revise my thinking for the moment. Firstly, they talk about a 50 nm shift between excitation and fluorescent emission. I cannot believe that an optical filter can have a low-pass feature with such a sharp cut-off. Secondly, as you say, an emission of 1000 nm minimum would seem practical. At that wavelength, the capacity of a CD ROM would be reduced to about 420 Mb. OK, if they used a visible light laser, this could be rectified, but we are talking about a different economic ballpark. Thirdly, the succes of the system would depend on the focussing of the beam onto a specified layer. How can the planarity of a pressed multilayer disk be assured over changes of humidity and temperature. (In the printed circuit industry, where multilayer pressing is common, bow and twist is a constant problem and there we are talking of an acceptable aplanarity of 5% of the linear dimension, not a µm or two.) Fourthly, as I mentioned, fluorescence depends on the excitation of the electron state of molecules, implying absorption of the light. Even if the fluorescent material is transparent to visible light, it cannot be to the excitation wavelength, therefore there is going to be much absorption and scatter through the layers. Even if I grant that the system works over a few surface layers, I cannot see it working over tens of layers. Furthermore, the pressed pits would introduce considerable refraction and diffraction, even if by some miracle all the rest was true. Etc.

Let us assume, now, that the technique really does work and can be commercially introduced. Starting with pre-programmed disks. They speak of 20 hours of video per disk. Great, all the works of Spielberg or Wender on a single disk. At what price? Two factors are involved. The manufacturing yield and the royalties. Who could afford such a disk as a one-time purchase? Whose taste is eclectic enough to looking at a preselection of films of such varying subjects, that they would pay relatively astronomic sums? The average Joe would rather buy a DVD or VHS tape with just one film, at a time. The market would therefore become quite restricted to such things as hotels, hospitals, airlines and suchlike offering a video-on-demand viewing, IMHO. This is a niche market.

Now let's look at what may interest us: WORM and WMRM disks. As described, a writer with a single laser would not work faster than a current CD writer, say 4 X. Or 55 hours to fill a 140 Gb disk! It would be conceivable, at a price, to have a system writing 10 layers at a time, so this could reduce it to 5,5 hours. This would require a basic data transfer rate of 7 Mb/s, which is not excessive, except that the multiplexing of such signal rates may be very difficult, as your would be transferring 10 separate signals at 700 kb/s, in reality, simultaneously.

I still require a lot of convincing.

Yes, your example of re-emission at a higher frequency is interesting, but is it fluorescence? I think not.

Doc, how can you measure superluminal transmission speeds when electrons beetling along a conductor move much slower? In reality, I think we still have a lot to discover when the wavelength is comparable to molecular dimensions. Common knowledge is restricted to X-ray diffraction, but a whole new world will open up when the frequency of the radiation goes up by an order of magnitude and the amplitude is high enough to cause other undiscovered phenomena. I think we are at the threshold where the fundamental research done in accelerator physics will start to tell us some practical applications may come about and then we shall be approaching the possibility of atomic data storage with all the knowledge in the world stored in a pinhead, and instantly accessible, at that. But not in my lifetime.

------------------
Brian (the terrible)

[jrr]

Sorry everyone if this discussion has got a bit technical and slight off the point of the forum but...

Brian,

As you suggest I think the main problem they will have is developing a cost effective manufacturing process. Also I agree with your point about the film market being quite limited. However if they could get the readout speed high enough they would make the ideal medium to disribute high resolution films to new digital cinemas, ie a film saved as a 4000x2000x32bitx24fps MPEG2 stream with minimal compression

Filtering out the source laser light (or picking off different flourencent peaks) will be no problem - narrow band interference filters which allow only a 10nm bandpass are common in laser devices. See companies like: http://www.ealing.com/sales/Product.asp?main=3&cat=501

I hadn't read the part about what material they were using, but they would indeed have a low activation energy of around 0.1eV corresponding to a wavelength shift of 30-40nm.

The storage density would be limited by the source wavelength not the re-emisson

Also focusing a semicondutor laser to different layers 50microns apart relativly easy with the kind of feedback mechanism they propose. The only problem being if you want to quickly refocus from one layer to another.
To improve the read speed I would think it easier to have two or more laser paths reading from different layers simultaneously - although this would obviously put up the price.

I hope they manage to get them working although as with the holographic storage dreams I think it will be a little while before we see them in the shops.

James

[sweeper]

Hi Brian,
I have a few technical details. Present color filters are really very nice. Take a long pass filter from Schott with the cutoff 695 nm. It has transmittance below 0.001% at 650 nm and shorter wavelength, but already 60% at 700 nm and 90% for wavelength longer 750 nm.
It is a standard thing, which I use every day.

If you excite a commercially available dye compound by GaAs laser at 640 nm, you can get fluorescence up to 1100 nm, depending upon compound. I guess polymers can be even more effective.

You should not care about diffraction limit for re-emitted light as it anyway spreads in all directions.

Absorption is not a problem at all. Take Ruby, famous material for lasers, it is almost 100% transparent, almost no color. The reason for this is that transition to fluorescent state is optically forbidden, i.e. electron cannot be excited into this state from the ground state by light, otherwise the lasers would not work.

Alex