REPORTS OF VISITS TO INDUSTRIAL LABORATORIES IN JAPAN

Nikon Corporation

Nikon's involvement with optical storage media is through a joint venture with Hitachi-Maxell, called MNO (Maxell-Nikon Optical). This company, which is nearly three years old, manufactures and markets exchange-coupled magnetic multi-layer MO disks capable of direct overwrite (DOW). Both 3.5" and 5.25" disks are produced by MNO. The high-end product is currently the 4X, 5.25" DOW disk with 2.6 GB capacity (double-sided). The 8X, ISO format disk was expected to be introduced into the market later in 1998 by Sony Corporation. The DOW version of the 8X, 5.25" MO disk will follow the non-DOW disk. (The ISO has almost finished specifying the 8X 5.25" DOW disk and has finished specifying the DOW 4X 5.25" and the DOW 5X 3.5".)

The current price of the DOW disks in Japan is approximately 50% higher than the non-DOW disk. For example the 3.5" DOW disk can be purchased for ¥2,400, whereas the non-DOW disks cost about ¥1,600. Despite this price difference, the general consensus at Nikon was that the overwrite capability is important to the user, and they believed that a large fraction of the users will prefer the DOW disks.

As far as compatibility with drives is concerned, all 640 MB drives for the 3.5" disks and all 4X drives for the 5.25" disks can handle the DOW disks as well as the ISO standard non-DOW disks. WTEC hosts also stated that all future-generation drives for MO will be compatible with the DOW disks.

The market for MO drives is still small, but most Japanese scientists/engineers with whom this author talked seem to think that there is a bright future in this field. In 1996 there were nearly 1 million 3.5" MO drives sold worldwide, and the number was close to 930,000 units in 1997. By far the largest share of the market (i.e, percentage of users) was in Japan. There does not seem to be a large market for the Zip drives in Japan, although the panel was told that Fuji film is now making Zip media, and NEC has been licensed to manufacture Zip drives. According to the WTEC panel's hosts at Nikon, Japanese consumers think very highly of optical technologies, perhaps because of the superior performance of CD audio and its resounding success in the market place. Any technology that uses lasers is likely to attract the attention of the Japanese consumers. As for MO technology, the panel was told that the shelves in Akihabara are full of MO disks these days.

No one at the meeting knew the exact size of the market for MO disks, but our panelists' guess (based on the number of drives sold last year) was that there is probably demand for 10 to 20 million 3.5" disks this year. Given that several manufacturers in Japan produce these disks, hosts were asked if this type of a market was suitable for their company to be involved with. The answer was a strong yes, with the comment that 10, 20, 30, 40 million disks per year is certainly a good enough market for Japanese disk makers to pursue.

There was some talk about the Windows 98 operating system and that this version of Windows will allow bootup from removable media. Also the fact that Win '98 will accept ATAPI interface (instead of SCSI) was considered a boost for MO drives in the near future. Fujitsu, for example, is already making 3.5" MO drives with ATAPI interface; Fujitsu has also announced 1.2 GB 3.5" (one-sided) MO drives, and reportedly has plans for Dragon I and II drives, which will accept both 3.5" MO disks and CD-ROMs.

Nikon researchers apparently are not involved in phase-change media development. The only information they provided on this type of media was based on their personal knowledge of the R&D efforts in Japan. They said that in April 1998 Hitachi and Panasonic would ship DVD-RAM disks and drives. These are 2.6 GB/side, 5.2 GB (double sided) 120 mm diameter disks. Sony's DVD+R/W is expected to have around 3.5 GB capacity, and Pioneer's DVD-RW will be about 3.4 GB capacity. The latter is expected to be introduced in 1999.

The participants had some reservations about the future of CD-R type media (CD-recordable, DVD-recordable, WORM-type media). They felt that with the mastering equipment makers' effort to develop a new method of mastering of CD and DVD, stamping as few as 100 disks will be cheaper than recording them individually on the CD-R type media. Of course CD-R will continue to serve a special niche market, but more and more people will gravitate toward the ROM type media on the one-hand and toward the rewritable media on the other hand.

WTEC's hosts stated that there are essentially two types of markets for MO disks and drives in the near future: (a) personal computer/consumer/portable applications, and (b) infrastructure applications such as servers and mainframes. From now on, they said, the 3.5" and 5.25" disks will move in different directions, partly because of differences in removability requirements and partly because of the price targets that manufacturers feel they can set for each product.

For portable applications it was felt that SyQuest and Jazz drives are not suitable because they can be dropped only about 40 cm or so. The same concern applies to the magnetic field modulation (MFM) method, which contains a flying head. The LIMDOW (light intensity modulation direct overwrite) technology, on the other hand, does not use a flying head and is therefore more suitable for portable applications. Technologies which use the magnetic field modulation technique are probably more suited for infrastructure applications (e.g., servers and mainframes) as well as for low-end consumer applications such as the MiniDisc.

The participants felt certain that the optical ROM format will be around in 20 years and beyond. They mentioned that semiconductor memories should be watched closely in the next few years. New semiconductor technologies such as one-electron DRAM might be able to compete with optical RAM storage products. The need of optical storage devices for some sort of mechanical actuator to access the data might be considered to be a drawback for this technology, when the recording densities would be much higher in the very distant future. The participants also felt certain that the optical RAM format will be around in 20 years and beyond.

Pioneer Corporation's R&D Laboratory

In the field of optical disk data storage, Pioneer is a manufacturer of CD-ROM, DVD-ROM and DVD-R media. Presently it is engaged to establish the DVD-RAM version 1.0 and 2.0 (2.6 GB and 4.7 GB) specifications in DVD Forum working group 5, but Pioneer's main focus is developing the 4.7 GB DVD-RW disc and drive as well as the next generation 15 GB DVD-RW disc and drive.

Pioneer researchers are engaged in a number of research activities at the leading edge of modern optical technology. For example, they are developing blue lasers for various applications using both the technique of second harmonic generation (SHG) and direct fabrication of blue semiconductor lasers based on III-V materials (GaN-based system). They are also active in developing new display technologies based on a class of organic electroluminescent materials. At the time of this visit the R&D laboratories employed a total of 165 researchers.

In the area of optical data storage, Pioneer's research seems to be focused on DVD-ROM and rewritable DVD systems. The company has an in-house mastering facility for producing high-density disks for research and development purposes. Pioneer researchers mentioned their plans to bring out the 15 GB DVD-ROM by the year 2001. This system, which is intended for high definition TV (HDTV), will use low-power blue lasers, and will employ actively controlled liquid crystal (LC) elements for aberration correction and for tilt servoing. The use of 410 nm blue laser and a 0.6 NA objective lens will allow a capacity of only 9 GB on a 12 cm platter. To get to 15 GB, Pioneer researchers plan to use a number of advanced techniques, including 2D equalizer, cross-talk canceller, adaptive tangential equalizer, and Viterbi decoder.

In its prototype 15 GB DVD-ROM system, Pioneer uses a three-beam cross-talk canceller, which also provides the feedback signal to the radial tilt correcting servo based on a LC element. Although in principle this liquid crystal element can also do automatic correction for the tangential tilt, in practice the speed of switching the LC is not sufficient for high-speed applications. The tangential tilt correction is therefore done by electronic equalization. The details of this equalizer (which they referred to as "super equalizer") were not discussed, because Pioneer is applying for a patent on this technology.

To produce the 15 GB master for second-generation DVD-ROM, Pioneer researchers used a photo-bleachable dye layer on top of the photoresist. Only the central region of the focused spot is strong enough to bleach the dye layer and, therefore, expose the photoresist layer below. In this way, the researchers could achieve super-resolution and create well-defined pits. They estimated that the use of the photo-bleachable layer had improved their cutting resolution by about 20%. Their mastering machine used a 0.9 NA lens and a 351 nm laser, and created DVD masters with a track pitch of 0.37 microns and minimum mark length (3 T) of 0.25 microns.

In the area of rewritable DVD the researchers mentioned Pioneer's DVD-R/W format based on rewritable phase-change media, with a limited number of write/erase cycles (around 1,000). This product is intended for the consumer market (as opposed to computer market), for which the limited cyclability is acceptable. Compatibility with DVD-ROM was highly emphasized. The material of choice for DVD-R/W is InAgSb, which is the same material used in CD-R/W. The researchers said that they have confirmed a jitter value of less than 7% in this material at 4.7 GB capacity and over 100 times cyclability, which is better than what DVD-RAM can claim at the moment. Pioneer researchers hoped to develop the 4.7 GB DVD-R/W before the end of 1998. They maintained that of all the issues facing rewritable DVD, probably the copy protection issue is the most significant stumbling block.

Pioneer researchers gave the WTEC team a tour of their laboratories, where the panel saw the various technologies developed for the second generation DVD-ROM and DVD-RW. In particular, panelists saw the improvement of the read signal due to the LC-based tilt servo, and the improvement of the eye-pattern by the so-called "super equalizer."

As for life after 2001, Pioneer researchers showed us preliminary examples of techniques that will be used in the 50 GB DVD-ROM. Unless some new technologies are developed in the next few years, it seems likely that electron beam lithography will have to be depended upon to create 50 GB master disks.

Pioneer Laboratories has done research on both magneto-optical and phase-change media in the past, but researchers seem determined to move towards phase-change technology in their consumer-oriented strategy. The researchers who talked to the WTEC panel maintained that advanced storage magneto-optical disk (ASMO, a.k.a. MO7) seems to have the support of the disk manufacturers but not the support of drive manufacturers in Japan. They also seemed to think that there are certain problems with the development of the ASMO technology, especially as related to the thin disk.

Sanyo Corporation's Hypermedia Research Center

One group some of the panelists visited is involved only in magneto-optical (MO) recording research, although Sanyo manufactures CD, CD-ROM, MiniDisc, pickups, and lasers for the optical data storage market. Advanced storage magneto-optics (ASMO) technology seems to be one focus of research at Sanyo. The current plans call for the 6 GB ASMO disk on a 12 cm platter (single-sided), followed by the second-generation disk at 12 GB using the MAMMOS technology, and leading up to the 30 GB disk in the third generation.

According to the Sanyo researchers, magnetic-field modulation (MFM) recording is attractive because it offers the possibility of recording domains as small as 0.1 microns in diameter. Although the minimum required data rate is currently around 4 Mb/s, the researchers foresee the potential of 50 Mb/s data rates in the next 2-3 years using the MFM technique. They also believe that in the near future access times will reach below the current value of 60 msec.

The substrate thickness for ASMO disks is 0.6 mm in the data area (1.2 mm in the central hub area). The tilt of the disk is not considered a serious concern here because MFM recording and the technique of magnetic super resolution (MSR) readout used in ASMO disks are not tilt-sensitive. Therefore high-numerical aperture objectives can be used in conjunction with the 0.6 mm substrate without much concern for tilt-induced coma. The preferred mode of MSR seems to be CAD-MSR. The GdFeCo layer used as the readout layer has a large Kerr signal, even at short (blue) wavelengths. As for future-generation devices using blue laser diodes, Sanyo researchers mentioned that they had researched superlattice materials such as PtCo, but they also felt confident that, using the MAMMOS technology, TbFeCo and GdFeCo media would offer sufficient sensitivity and acceptable levels of signal to noise ratio (SNR) at short wavelengths.

It was mentioned that the density of MO media is limited only by the width of magnetic domain walls in amorphous RE-TM alloys (on the order of 10-20 nm), and that data transfer rates are limited perhaps by the 10 psec time constant for magnetization reversal in these media.

Concerning the competition between hard disk drives (HDDs) and optical disk drives (ODD), hosts at Sanyo felt that both technologies will coexist in the future. While HDDs are superior in terms of cost-performance and data rate, ODDs have the advantage of removability and mass-reproducibility. It was mentioned that the rapid growth and the pace of change in HDD technologies is perhaps the reason why the Japanese companies are not leading in this area. Slow decision-making processes were blamed for the slow pace of change in Japan. The Sanyo researchers emphasized that while the hard disk drives are made in the United States, many parts and components come from Japan. The focus of magnetic recording in Japan seems to be on magnetic tapes for VCRs.

Sanyo researchers felt that the MO technology is superior to phase-change (PC), even in terms of compatibility with DVD-ROM. The wollaston prism, for example, costs under a dollar in large quantities, and silicon photodiodes are very cheap; in fact the split detectors needed in MO drives are about the same price as the single-detectors used in PC drives. Some versions of MAMMOS operate without a magnetic field whatsoever. So all the talk of MO being more expensive than PC, they contend, is unrealistic.

It was mentioned that the growth path for LIMDOW technology is through MSR and blue lasers, although the researchers felt that the commercialization of high-power blue lasers before the end of 1998 was not likely. They also emphasized the importance of the partial response maximum likelihood (PRML) technique for high-density recording. As for the use of liquid crystal (LC) devices in optical disk drives, they mentioned that Sanyo currently uses LCs in its DVD-ROM drives to achieve compatibility with CD-ROMs, but they said that the cost of these elements must come down before their use becomes widespread.


Published: June 1999; WTEC Hyper-Librarian