Data Storage Devices: Are Hard Disks Obsolete?
by Simon Mansfield
The makers of data storage devices have been hitting the personal computer market with an avalanche of new products and better technology. For users, this means faster, easier, and less expensive data storage and retrieval. For potential buyers, though
, the path can lead through a bewildering maze of possible solutions.
Last year, for purchasers of a new system, it was generally a toss-up between a 200MB or 5OOMB hard disk drive (HDD). This year, it's a choice between a 1 gigabyte (GB) HDD or a magnetic-optical (MO) drive. Next year, it will be all of these, and sever
al more.
PCR for PD
To confuse the data storage market even further, Matsushita Electric Industrial Co., Ltd., has unveiled a 650MB rewritable drive and quadruple-speed CD-ROM, all combined into one package. Dubbed the PD (for Phase Drive), the system uses a new optical t
echnology -- called phase change recording (PCR) -- that allows the storage medium to be optically rewritten. PCR is similar to the magnetic-optical system. Like MO drives, PCR uses a laser to read and write data, but there are two key differences: PCR do
esn't use a magnet, and the optical laser employed in phase change drives operates at a much lower temperature than the lasers in MO drives.
According to Takeo Ota, general manager for optical research at Matsushita, the cooler laser reduces heat leakage on the disk surface. This enables the data tracks to be written much closer together, increasing the disk storage capacity. In addition, b
ecause magnets aren't used, PCR drive heads are cheaper to make than the heads used in MO drives.
The PD requires some trade-offs in the quest to combine both a rewritable drive and a CD-ROM, however. Standard PCR drives have an average seek time of around 45 milliseconds and a data transfer rate of 1.5MB per second. With the addition of a quadrupl
e-speed drive for the CD-ROM side, however, access time in the combined unit is 180 milliseconds (four times as long), while the data transfer rate is 870KB (one-half the speed). While this is a significant compromise on the PCR side, it represents a grea
t improvement over current CD-ROM drives. Overall, current PD drive performance is comparable to a hard disk. In an unofficial benchmark test, Matsushita demonstrated a Windows PC processing ten SOOKB screen saver files in consecutive order. For the hard
disk, it took 15 seconds, and for the new PD drive, 17.5 seconds, to process and display all 10 screens (using a software cache in both cases). Retail prices for PD have yet to be announced, but the company says that OEM samples will be available for abou
t $1,000 early next year. In addition, the company is seriously considering offering the key components to other storage device makers
The Matsushita product development people have made it clear that they understand lust how important pricing is going to be. "We want to try and get prices down quickly," said a Matsushita official, while adding that the US multimedia market is the key
market the drive will be aimed at. A company spokesman acknowledged that the US market would demand prices below $600 for it to be successful. This would drive OEM prices down to the $300-400 range, however, and force production costs down to the $200 (2
0,000 yen) mark. One company official reacted to these figures with a tight grimace, saying, "That's what we are afraid of; we are under a great deal of pressure to get costs down this far quickly."
Price is not the only concern, however. Asked about the prospects for PD, officials at other leading storage makers agreed the product is innovative, but most cautioned that the lack of standards for phase change is a major drawback. Since Matsushita i
s the only company currently making phase change systems, though, if its phase change technology is successful, the company is in a good position to set a de facto standard.
MO
After several years of identity crisis, MO has firmly split into several disk sizes, while at the same time consolidating international data format standards for the 3.S-inch drive segment. This is the disk size that companies like Fujitsu Ltd. are tou
ting for the everyday desktop user.
The larger 5.25- and 22-inch magnetic-optical disks are generally regarded as professional products aimed at network managers, audit managers, and professional multimedia houses. Hitachi has pushed capacities for 5.2S-inch drives to 2.5GB (currently,
the best 3.5-inch MO drive offers 230MB capacity). At the very high end is NEC, which is aiming for the video-on-demand market. NEC has announced a 12-inch, 23GB drive that will reportedly go on sale by the end of the year.
With all this activity happening in MO, it's not surprising that many users are considering buying an MO drive. But while MO disks for data storage are looking increasingly attractive, engineers at Fujitsu I,td. and Toshiba privately concede that PCR
probably offers the better long-term technology option, given its inherent advantage of higher densities. Toshiba Corporation is planning its own 3.5inch PCR drive, and Matsushita can be expected to lobby hard for Toshiba to base its phase change product
s around Matsushita's approach.
CD-R
Another up-and-coming player in the data storage market is Ricoh Corporation, currently one of the few makers of Compact Disk Recordable (CD-R) systems. A spokesman for Ricoh said that the company is working with other companies to update the Orange Bo
ok standards (which govern the CD-ROM format) to enable CD-ROMs to store over a gigabyte of data. If this new standard were then applied to a phase change disk, it would effectively make the CD-ROM erasable and rewritable. Code named CD-PC by Ricoh, a com
pany official confirmed in a telephone interview that Ricoh is aiming to have a CD-PC drive on the market by early 1996.
MD
Then there is MiniDisk (MD) Data from Sony, a low-end optical system offering a capacity of 140MB. Based on Sony's MiniDisk audio system, the launch of MD Data was anything but perfect; it is only now hitting the US and European markets. Sony's indecis
ion in getting MD Data out the door may well leave the format dead on arrival as other low-cost, higher capacity options hit the market.
Sony spokesman Georges Gerad notes that MD Data is just one of many projects underway, and it's difficult for a company as big as Sony to say, "Yes, this is the product we should back." But with 12 million PCs to be sold in the US market alone this yea
r, Sony may have dropped the ball on getting MD Data adopted by the industry as a viable low-cost option for low-end storage needs.
OF
Fuji Photo Film Co. is drawing up plans to attack the low-end of the market with a 100MB optical-floppy system. According to a recent announcement by Fuji, the product is now in the final stages of development; the company is still to decide whether ca
pacity should be doubled to 200MB. No matter which size is adopted, Full plans on an early 1995 release of the system.
But is it cheap?
Whether Full a optical-floppy, or Mitsubishi's PD, or any of the other new technologies succeed ultimately depends on price. To win in today's digital world, a manufacturer must capture the attention of PC makers straight off. And for PC makers, the bo
ttom line is whether users will embrace the new technology -- which is a function of both system and media cost.
For storage systems, this means that manufacturers have to take a risk and invest heavily in capacity and introduction. If the manufacturer bets on the wrong system, the loss can be substantial. But if prices are low enough on first release, users will
soon be hooked and filling millions of disks with gigabytes of multimedia.
Sony fights back in storage battle
At a press conference held in Tokyo in late October, Sony Corporation, in conjunction with Hitachi Ltd. and the 3M Company, announced a prototype magnetic-optical (MO) disk drive that packs 650MB onto a 3.5-inch disk. Company officials were hesitant to
commit to a time frame for product release, saying only that it would be available sometime during 1995.
The Sony MO system uses a new format standard based on 2048-byte sectors for writing and reading data. (Conventional data systems, such as hard disk and floppies, use 512-byte sectors, which has also been adopted by the International Standards Organiza
tion for MO systems.) A Sony official claimed that using 2048-byte sectors is more efficient and was the key to increasing capacity to 650MB. Sony and Hitachi will attempt to get the backing from Fujitsu Ltd. and other MO makers for a new ISO standard --
which would make existing drives incompatible.
The timing of Sony's announcement was a direct reaction to Matsushita's strong push to get phase change technology adopted by other makers. (Toshiba is expected to make a public commitment to phase change in the near future.) An official from 3M told C
omputing Japan that Sony and Hitachi, in making the announcement, wanted to alert other makers that MO drives remain a serious contender for high-capacity storage. The official cautioned, however, that 3M remains neutral in the storage technology battle,
acknowledging that the phase drive (PD) has a number of advantages, ranging from the ability to quickly increase storage capacity to the low production costs for drive heads.
The technological key behind Sony and Hitachi's new MO system is the use of pulse lasers rather than continuous lasers. Pulse lasers consume substantially less power and, more importantly, make the development of room-temperature, long-life green and b
lue lasers much easier. While Sony still regards development of green lasers as a long-term option, company officials suggested that incremental improvements in existing technology could enable the capacity of the new 3.5-inch system to be doubled.
Where this leaves the storage device industry remains to be seen. All indications point to a battle between Matsushita and Sony, much like the videotape VHS and Beta wars of the early eighties. Sony currently has the upper hand, with a growing number o
f companies backing MO over PD in the current marketplace. But as an official from 3M suggested, "in my opinion, phase change is the future."
|