What are you looking for ?
Infinidat
Articles_top

History: FDD Contributions to HDD

For example DOS (disk operating system) for FDD and form factors later adopted in large part by first HDDs

This article was published by WikiFoundry.

FDD contributions to HDD

Why it’s important
If it look like a duck, acts like a duck, and tastes like a duck … it’s probably a duck.”

The FDD is more of a HDD with a limp substrate than any kind of a tape drive. The FDD set many of the HDD standards, and is a direct ancestor of 8-, 5.25- and 3.5-inch HDDs which at various times dominated the PC industry. It’s often debated (by people who like to argue about such things) that the FDD is a tape device, but most organizations who have come to grips with the issue (ANSI, CBEMA, standards committees) have concluded that the FDD is really a disk drive, and so it is included here for your consideration of its importance in the history of hard disk drives.

Discussion
Prior to FDDs most large disk drives were standalone boxes the size of a washing machine; some standardization around 19-inch rack for small disk drives.

The original FDD was created by IBM as a program loader for a disk drive controller. As such it was designed as a 8-inch read-only device which enabled boot-up software to be updated electronically rather than using hard-wire programming. It was followed by 8-inch read/write versions which established media standards but because IBM never sold their 8-inch FDDs separately they established neither packaging nor interface standards.

8-inch FDD details
Shugart Associates SA800 success establishes form factor and interface standards for 8-inch FDDs. FDD manufacturers compete by offering SA800 compatible devices

Shugart Associates SA1000 pioneered HDDs with floppy like interface and floppy form factor. The reason was simple, many minicomputers had cabinets designed to hold two full size 8-inch FDDs, a HDD upgrade option had to fit in one of those holes to be mechanically and electrically compatible, which is just what happened.

5.25-inch FDD details
The 8-inch version was followed by the smaller form factor in the 5.25-inch pioneered Shugart Associates with its SA 400 full height FDD drive; it was the direct ancestor in form factor and similar in interface to the first desktop HDD, the ST506, also by Al Shugart, then president of Seagate.

The importance of the FDD to the history of computing cannot be underestimated. Flex disks became the primary means of software program distribution (Visicalc, Lotus 123, and WordPerfect were extremely popular), selling computer games, word processing storage, etc.

Early FDDs sold for about $600 each (I bought 2 for my Apple-2plus), and the 5.25-inch ‘Mini’ disks became huge sellers for 3M, Memorex, Dysan, BASF, among many vendors. Over time the capacity was increased first by using both sides of the disk, then increasing areal density. The final 5.25-inch version had 1.2MB capacity, but technical problems appeared. Concentricity was an issue with increased track density, so a plastic support ring was added to the ID hole of the media to minimize damage. The DS/DD (double-sided/double-density) version also used a cobalt substituted high coercivity iron oxide for improved performance, but these disks were not compatible with older drives. The packaging resembeled a greeting card in an open envelope, which became a reliability issue for some users.

3.5-inch FDD details
A new and improved FDD media was desired which would be smaller, more protective of the recording media, not subject to ID wear, and be inexpensive. In the early 1980s various manufacturers tried unsuccessfully to establish a smaller-format standard, notably Sony in 1981 shipped a 3.5-inch (actually 90mm x 94mm) FD using a rigid plastic cartridge with a sliding metal shutter. This original Sony format was only used by HP and Sony itself.

In late 1982 the Microfloppy Industry Committee standardized on a media design derived from the Sony design and shortly thereafter (early 1983) compatible drives were offered by a number of manufacturers. The issues were openly discussed at the X3B7 and X3B8 jointly held ANSI standards meetings, with competing proposals from IBM, MIC/Sony, Dysan and a few others. At a meeting in Lake Tahoe, things came to a conclusion when Dick Pricer of GE made a motion to adopt the MIC/Sony design, which met all of the desired criteria. It had a rigid plastic shell, an automatically closing shutter which protected the media, and a metallic hub not subject to plastic wear. Most other designs were simply smaller variations of the existing 5.25-inch design, sharing some of the same problems.

Two representatives from IBM announced that they would adopt whatever design the X3B8 committee decided on, but from the data they presented it was apparent that IBM had investigated the MIC/Sony offering in great detail and preferred it over their own design. After a short break for individual discussions, the meeting reconvened and the Sony design was overwhelmingly adopted. That 3.5-inch drive design became the FDD standard, establishing the form factor for all subsequent FDDs using that media … but more importantly set the form factor standard for the 3.5-inch HDD (had to fit in the chassis hole) which became the largest selling and most prominant HDD in the history of computing.

The FDD was a new device, so the concepts and ‘rules’ of use had to be developed such that user data interchange (a key ANSI objective) was maintained. Since disk storage is random access, tape handling concepts had to be abandoned. New protocalls (some borrowed from drums) included tracks or cylinders of data instead of lengths on a tape, etc.

The various rules adopted evolved to a DOS (disk operating system) for FDD, which was later adopted in large part by the first HDD devices. The bottom line is that FDD was the direct ancestor of desktop HDDs, in size and data organization. “If it looks like a duck …

Provenance note: This article was originally written by Bill Carlson. Revision 5 of the article was reviewed at the June 20, 2012, meeting of the Computer History Museum’s Storage SIG and it was decided that the article was not suitable for inclusion on the main timeline but should be evolved into an drive packaging survey article with links to and from such events on the main timeline.

Articles_bottom
AIC
ATTO
OPEN-E