Micropolis

~ the last words ~

Micropolis started in Chatsworth, California in 1976, which puts it in the first generation of disk companies, a good decade ahead of the crowd that got wiped out in the shakeout. Stuart Mabon founded it and ran it for nineteen years. The first business was floppy drives and the subsystems around them, sold into the S-100 hobbyist market when that market still mattered, and Vector Graphic machines were among the ones that shipped with Micropolis parts inside.

The company moved into hard disks with an 8-inch drive and worked downward from there, reaching a 3.5 inch line in 1991. Manufacturing went to Singapore in 1986, a decade before a Singaporean buyer turned up for the whole business. Toward the end there was a recognizable specialty in drives sold for audio and video work, where sustained transfer rate mattered more than average seek time and the buyers would pay for the difference. That was a good niche and it was not big enough to carry a company through what happened next.

By 1995 revenue had dropped hard and the losses were large. Mabon handed the chief executive job to Larry Smart that year. In March 1996 the disk drive business went to Singapore Technologies, an arm of Temasek Holdings, for about $47 million, and the transaction closed on the 29th. Whatever remained of the old company, the subsystems side, renamed itself StreamLogic.

Micropolis was eight years old by the time I was born and finished the year before I got my first modem. There is no version of my life where the name could have meant anything to me, which is true of most of the first disk generation. What they built arrived in my hands as a default, in a machine somebody else specified, with a different company's sticker on the drive.

~ the record ~

~ the quad density years ~

The MetaFloppy line is the part worth remembering on its own terms. Shugart had set the 5.25 inch standard at 48 tracks per inch, and Micropolis doubled both the track density and the recording density along each track, which multiplied the capacity of a diskette by four. It chose 100 tracks per inch mostly because a round hundred is a pleasant number to design to, and that worked out at 77 tracks on the disk against the 35 or 40 everyone else used. The 1043 and 1054 model numbers meant something to anyone buying S-100 boards in 1980.

Going denser than the standard was a genuine bet with an obvious cost attached. Media got more expensive, alignment tolerances got tighter, and your disks were not going to be readable in anybody else's drive. It worked well enough that the format acquired a name of its own, and Tandon built Micropolis-compatible 77-track drives to feed the same customers. Commodore shipped the density in the 8050 and the 8250. The company kept making that sort of bet for twenty years, right up to the point where customers were buying on price per megabyte and nothing else.

~ what nineteen months cost ~

Singapore Technologies did not buy a failing business by accident. It bought into a market that four companies were competing in seriously at the start of 1996, and by 1997 there were at least seven, all of them adding capacity into the same demand. That is the shape of the drive industry in one sentence, and it is why the sector produced more corporate deaths than almost any other part of computing.

The numbers on the way out are brutal. Revenue over the first three quarters of 1997 came to $140 million against a target of $45 million a month. Total cost to Singapore Technologies was around $120 million with debts above $45 million left at the end. Coopers and Lybrand were appointed as liquidators on 10 November 1997, nineteen months after the deal closed.

About 3,000 people lost their jobs, roughly 1,300 of them in Singapore on hard disk assembly and 700 in Thailand making head stack assemblies, with 300 more in the United States doing research and marketing. Seagate was rumored as a buyer of the remains and Maxtor is said to have been offered it and declined. Nobody took it. The brand stopped being marketed in 1998.

~ the founder's next move ~

Mabon started again. In 1996 he founded Medea, which built storage arrays for video editing, the same niche his old company had been growing into when it ran out of room. A founder who leaves and immediately rebuilds the profitable corner of what he lost is telling you exactly which part of the business he thought was worth keeping.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »