Although video disks will not need to require capacities much beyond 30 GB, most people believe that future computer applications will continue to utilize as much capacity as can be made available at a reasonable price. Furthermore, video disks have only modest requirements for transfer rate (under about 5 MB per second). But, in the year 2005, desktop computers will be expecting transfer rates of 40-80 MB per second, and backbone networks will be operating at better than 1 GB per second.

Neither phase change or magneto-optic technologies are likely to extend their capacities beyond 100 GB per disk (Fig. 6.2). Neither is likely to achieve transfer rates in excess of 40 MB per second without using parallel optical pickups at significantly higher drive costs.

For capacities well above 100 GB on a 12 cm disk, one must look for an optical technology that utilizes the volume of the disk and not just the surface. There are two candidates currently in the research community: two photon and holographic. Whereas phase change materials may be possible with sufficient transparency to allow for 2-4 layers, two photon materials are completely transparent except at the intersection of two light beams. Thus, hundreds or maybe thousands of layers are possible. Capacities could be more than 1 TB per disk. Holographic recording also utilizes the volume. Light is imaged coherently by the selected hologram and incoherently by all of the others. It could also achieve more than 1 TB per disk. The key issue for both is the development of suitable materials. For commercial applications, an erasable material is preferred. However, a very inexpensive write-once material could also be a commercial success. The WTEC panelists saw little work on either of these technologies in Japan. Most of the work is in the United States.

Fig. 6.2. Areal density vs. technology.

Probe storage is another new technology that is being pursued by many companies in Japan and a few companies and universities in the United States. The panel saw programs at Cannon, Toshiba, Hitachi, Matsushita and others in Japan. Carnegie Mellon University, IBM and HP have programs in the United States. Probe storage offers the potential of really high areal densities, perhaps as high as 10 Tb/in2. In addition, it also holds the promise of small size devices that will operate at low power. Representatives of several companies mentioned that they see probe storage for portable applications such as camcorders. Chapter 7 of this report covers some of these longer-term options in more detail.

Published: June 1999; WTEC Hyper-Librarian