The MOD Squad

Is There a Magneto-Optical
Drive in Your Future?

Computing's traditional magnetic storage media -- hard disk drives and floppy disks -- face competition from several fronts, but especially from magneto-optical (MO) devices. The high-speed, large-capacity (and, traditionally, pricey) 5.25-inch MO drive has established a firm foothold in the business world, but the lower-priced (smaller capacity, slower) 3.5-inch MO drive finds favor among the typical end user. We look this month at recent developments in the storage media market and the future of MO drives.

by Virginia Kouyoumdjian

The world of computer peripherals has become increasingly bewildering for the average user -- and no area more so than data storage devices. The number and variety of possible options seems to grow by the month. (Just when we had become used to FD and HDD, then CD-ROM and MO, along comes CD-R, PD, Zip, Jaz....)

Development of magnetic
technology slows

The development of the PC's first convenient storage media, the floppy disk (FD), more or less ground to a halt after the high-density 1.4MB format was developed. Half-hearted efforts were made to introduce a 2.8MB format, but enthusiasm for magnetic technology development was quickly transferred to hard disk drives (HDD).

From the early days of low capacity and frequent crashes, hard disk drives have made enormous strides in terms of reliability, speed, capacity, and -- of greatest interest to most consumers -- price. In the United States, the average price per megabyte of storage for hard disks is now reckoned to be under $0.40, and even in Japan prices are dropping rapidly toward that level.

With the fall in prices has come an increase in the average size of drives being purchased by consumers. In the 1980s, an 80MB HDD was considered "spacious." By 1994, typical hard disk drives were in the 300MB to 500MB range, and in 1995, the "average" has moved up to the 500MB to 700MB range. By 1996, many consumers will be opting for a 1 gigabyte drive to start with, and business will routinely be buying 1.6GB and 2GB HDDs.

The reason that consumers are driven to buy ever-larger hard disk drives is not just falling prices; operating systems have become more sophisticated and require ever-more capacity. Multimedia computing is at last becoming a reality, and the graphic and sound files it brings with it require enormous amounts of disk space. Even programs themselves are becoming larger, with a single Windows-based application, such as a word processor, taking up 10MB to 20MB or even more. A disk drive that a few years ago was considered more than enough is pitifully small for today's needs; several hundred megabytes is now regarded as a minimum.

CD-ROMs take over
software distribution

The increasing size of hard disks has dealt adequately with the larger data storage requirements for the end user, but it does not offer a solution to the growing problems faced by software developers in getting their merchandise to the users. The standard hard disk's greatest limitation is that it is a non-removable storage device. Hard disks have traditionally relied on their puny cousins, the floppy disks, to supply them with new applications.

With typical application packages now consisting of eight or more (compressed) floppy disks, software distribution is increasingly becoming the territory of CD-ROMs. Some software programs are now available only on CD-ROM -- or at least are cheaper that way.

CD-ROMs have been particularly welcomed by the entertainment industry, since they broadened considerably the scope for game manufacturers in terms of what can be included in a software package. By their nature, however, CD-ROMs face their own major limitation -- the inability to record data. In cases where data needs to be recorded (as in game scores, or saved games), this must still be done on the hard disk.

The use of CD-ROMs, therefore, is limited to that of a playback medium, a limitation that has been more than offset by their low price. Since the technology to press data onto CD-ROMs is the same as for audio CDs, they have benefited from economies of scale from the beginning. This is an unusual state of affairs for a new medium, and one that has contributed greatly to the fast popularization of CD-ROMs. Affordable writeable CD-ROMs, though, are still far in the average user's future.

The increasing inadequacy of floppy disks

There remains a real need for a removable, portable medium to supplant the floppy disk -- one that can offer sufficient capacity to handle today's needs. Over the years, a number of removable media have been offered to consumers (the best-known being the Syquest drive and Iomega's Bernoulli drive), but these have suffered from being generally high-priced and single-company proprietary products that are compatible with nothing else (and often with little compatibility between generations of the same product). As a result, while some of these options have found use as a data back-up device for office servers or power users, they have not achieved generalized recognition as a viable storage option.

It is into this picture that the 3.5-inch magneto-optical drive (MOD) appeared in the early 1990s. Earlier generations of MODs had been expensive and bulky, regarded mainly as specialized office filing tools with little benefit to offer the average user. Initially, MODs faced the combined obstacle of offering the consumer (and office manager) something new and unfamiliar, while also of being relatively expensive. (This was the time, remember, when prices for hard disk and CD-ROM drives were beginning to fall substantially.)

On the plus side for MODs, however, was the fact that a number of manufacturers had worked together to develop and adopt an ISO standard. This immediately placed MO drives closer to the familiar floppy drives than the Syquest or Bernoulli in terms of potential mass appeal.

MO market poised for rapid growth

The MO market has so far developed slowly. Some suspect that part of the reason has been that, with the exception of IBM, most of the MO drive manufacturers are Japanese (this in contrast to hard disk drives, where some of the biggest names in the business are American). Even today, it is variously estimated that 30% to 50% of the global MO market is in Japan.

The first MO drives took disks that held 128MB of data; while this was a respectable capacity, it was an awkward size -- too small to conveniently back up a hard disk's worth of data (even when compressed), yet too large (and expensive) for the tasks normally relegated to floppy disks. The MO market in Japan only really started to take off in 1994, with the introduction of the 230MB format. According to IDC Japan, the Japanese market for MO drives went from 110,000 in 1993 to 304,000 in 1994. Estimates are that the market will reach over 475,000 units in 1995 and surpass 700,000 in 1997.

Within this fast-growing market, 94% of the market share is held by just five companies, and one of these, Fujitsu, accounts for 51% of that total. In an interesting reversal to the standard pattern of what happens when a new product gains momentum and competition stiffens, Fujitsu has quickly and substantially increased its market share each year. Within Japan, for example, Fujitsu held a 57% share in 1994, up from a mere 17% market share in 1993.

Fujitsu's winning strategy

How has Fujitsu achieved a position of such dominance in Japan's MOD market in such a short time? In part, it is the result of a strategy that has aimed to establish the company at the forefront of consumer awareness, both in hard disk drives and new technologies such as MO.

In the past, Fujitsu was a big company that concentrated on big products. Two things then happened: Fujitsu became aware of the scale of possibilities in the consumer market, and hard disk drives themselves moved towards Fujitsu's area of specialization. This offered an excellent opportunity for the company to establish itself; increasingly, it has moved away from being an OEM supplier to selling its large capacity drives under its own name.

In the MO market, an early determination to make the product work through development and production facilities has paid off with growing market share. The company was well aware that in an environment where prices of other types of storage were virtually collapsing, little could be achieved by maintaining high levels of pricing for MO drives. By cranking up volume, Fujitsu became both a price leader and a market leader.

None of what has happened so far, however, has really started the ball rolling in terms of making the MO an integral part of every computer system. In the same way that hard disk drives really took off when almost every computer came fitted with one, the MO needs to be thought of primarily as a large-capacity internal storage device rather than an external peripheral add-on. And, if the MO disk is to replace the floppy disk or hard disk, the MO drive needs to become even cheaper and be a "standard" item with new computer purchases.

MO offers unique solutions

The advantages of the MO over the floppy disk are numerous. Says Mike Beirne of Fujitsu's corporate PR department, "MO offers unique solutions at many levels. They can be used as universal drives and for data transfer between different systems -- something which is increasingly needed in the current multi-platform environment. They offer high levels of security, and they are of course an excellent solution as a back-up utility."

The cross-platform side of the MO can already be seen in a series of applications marketed under the RingoWin name in Japan. These make it possible to use Macintosh-formatted MOs in a Windows environment, and vice versa. Another asset of an MO disk is that it can be used in a Partial-ROM configuration -- with an application, for example, written on a Read-Only section, but leaving the remainder of the disk blank and rewritable.

Until recently, the smallest MOD model available in the market was 25.4 mm (exactly one inch) in height, which was a little too large to comfortably fit in the typical computer. With the announcement of Fujitsu's micro-ISO compatible 230MB drive, though, the equation changes. The new drive is only 17.3 mm high and can easily be integrated into both desktop and notebook computers. Its seek time is 70 ms, which -- while considerably slower than the 35 ms of the larger model --is much faster than any floppy drive ever was. Fujitsu was the first manufacturer to announce a desktop model with an internal MO drive earlier this year, but other makers stood poised to follow suit.

The matter of seek time is frequently brought up when comparing the performance of MODs to that of HDDs. The very large hard disk drives being marketed today have seek times as low as 8 or 9 ms, while the standard 3.5-inch MOD seek time is more like 30 to 40 ms. To what degree this really matters, though, depends on what you are comparing the MO to. The MO compares very favorably in seek times with floppy drives and CD-ROMs.

Olympus and
fast-spinning MOs

One company, Olympus, has addressed the problem of speed from a different angle. As opposed to its principal competitors in this field (Fujitsu, IBM, Matsushita, and Sony), Olympus was originally a manufacturer of optical products. It therefore tackled the issue from that point of view, and has succeeded in decreasing the seek time of its own drives to 27 ms by increasing the rotational speed of the drive. In contrast to the Fujitsu 230MB drive, which spins at 3,600 rpm, the Olympus Deltis 230MO Turbo spins at 4,200 rpm, while conforming to the same ISO standards. According to Toshihiko Katsuyama, manager of the Marketing Department, Information Peripheral Division, at Olympus, "High-speed rotation was made possible because of the high performance of the Olympus head, which has very low levels of optical loss. So, performance can be maintained even when the speed is increased. Without the right head performance, such an increase in speed could result in data loss."

Performance is undoubtedly enhanced by Olympus' approach, but this also means a higher price for the drives. This is something the company is philosophical about, pointing out that there will always be a user base for which performance warrants a premium. Which is not to say, however, that prices won't also fall for these high performance products.

The 640MB MO format, and others

One development that will spur a further decline in the price of 230MB MO drives and disks is the introduction of the next generation format, the 640MB MO. This format is currently under consideration by the ISO (with approval expected as early as December 1995), and some two-dozen companies have expressed support (including major forces in the MO world, such as Fujitsu, IBM, and Olympus, as well as Canon, Kyocera, Philips, and 3M). The 640MB MO is not expected to kill off its 230MB predecessor entirely, though; the two formats are likely to co-exist for a while.

Some major names missing from the 640MB MO supporter list, though, include Matsushita, Sony, Toshiba, and Pinnacle Micro. All of these have their own competing formats on the market. Matsushita's format is known as PD (Phase-change Disk, or PowerDrive), Sony's as SSD (Super Storage Disk), and Toshiba's as PCR (Phase Change Recording). Neither of the first two conforms to ISO standards, nor are they compatible with one another or with MO. Toshiba has presented its standard to the ISO for acceptance, but its format, too, is incompatible with everyone else's. Pinnacle Micro, meanwhile, leapfrogged the pack this fall with a 5.25-inch 4.6GB disk drive dubbed the Apex.

Another arrival on the scene, one proving very popular with budget-conscious consumers, is Iomega's Zip Drive (which currently sells for about one-third the price of the cheapest MO drive). The Zip drive (its disks are as thick as two 3.5-inch floppy disks) is based on the older magnetic technology and has been described by one MO manufacturer as, "a great product, easy to buy -- but offering little long-term reliability." And following the Zip will be the Jaz, Iomega's 1GB removable-disk drive. These, too, face the issue of non-compatibility with anything else, something that is irking consumers more and more.

In fact, it is the growing presence of CD-ROM drives that has been described by some as the real threat to the possible success of MO. Increasingly, computers come with a built-in CD-ROM drive, and as a pre-recorded data medium, CD-ROMs are relatively cheap. Once users have built up libraries of CD-ROM, they are unlikely to want to throw them out to invest in something more costly -- even if it does offer the possibility of recording.

This issue is being tackled by makers like Fujitsu, which is working on a hybrid drive that can both read CD-ROMs and read/write MOs, satisfying two needs. As Mike Beirne points out, "Products that merge differences, rather than present the consumer with yet another choice to make, are ultimately better."

A potential "super-floppy"

The whole area of storage devices and prices is still very much a Wild West experience for the consumer. Trying to make an educated choice on what to buy, when something new is always around the corner, is incredibly difficult. As it picks up momentum, the 3.5-inch MO format is moving away from the "here today, maybe gone tomorrow" zone into the "I might be able to use this" zone. The fact that ISO standards ensure a high degree of compatibility in this case can only help the process, and competing formats that don't conform to ISO seem to be on fairly shaky ground, even if they do offer short-term draws like very low prices.

There is no doubt that a viable alternative to the floppy disk is badly needed. At the moment, all indications are that magneto-optical disks will fill this need, becoming tomorrow's "super-floppy."ç

Today's data-storage options

For the prospective consumer of a removable-disk data storage solution, the choices can be confusing. Among the products now available in Japan that fill the gaps left by standard hard disk and floppy disk drives are:

Zip -- Iomega Corporation's Zip drive is an external unit that uses "fat" 100MB 3.5-inch disks. Zip uses magnetic technology, and the data transfer rate is faster than a standard MO (about 1.5MB/second). Seek time is 29 ms. Typical prices in Japan are ¥29,800 for the drive unit (parallel or SCSI interface) and ¥2,680 for a disk (for a storage cost of about ¥26.8/MB).

RHD -- What is commonly called a Syquest disk (after its originator) in the US goes by the acronym RHD (removable hard disk) in Japan. Currently on the market are two (non-compatible) versions: 135MB and 270MB. Seek times are comparable to a standard hard disk, about 13 ms. The 135MB version sells at about ¥30,000 for the drive unit and ¥3,000 for the disk (¥22/MB). The 270MB version is priced at ¥80,000 for the drive unit and ¥8,000 per disk (¥30/MB).

MD -- Sony's mini-disk format was developed for audio use, but is being adopted for data storage as well. Its merits are small size and 2-hour battery life; major drawbacks are a 500-ms seek time and high per-megabyte storage cost. The drive unit sells for ¥50,000, and a 140MB disk costs ¥6,000 (¥43/MB).

PD -- Matsushita's Phase-change Drives, OEMed for several manufacturers, can also accommodate 4X CD-ROM use. This is the technology receiving the most attention recently from Japanese press and consumers. The drive is typically priced around ¥120,000, and a 650MB PD disk sells for ¥6,500 (¥10/MB). Seek time is about 65 ms.

CD-R -- A recordable (write-once) CD drive sells for about ¥120,000, while recordable CDs in the 600MB to 680MB range sell for ¥6,500 (about ¥11/MB). While great if you want to "master" your own CD, the write-once limitation makes this a specialized rather than general data storage solution; the PD is priced similarly and can read CD-ROMs.

MO -- For comparison, typical 3.5-inch MOs are priced between ¥80,000 and ¥160,000 for the drive unit; a 230MB disk costs about ¥2,000 (¥9/MB). Seek time is in the range of 30 ms.






(c) Copyright 1996 by Computing Japan magazine