4

I have lots of old IBM 5.25" diskettes.

I have 360 kB "Single sided", 360KB "Double Sided" and 1.2 MB diskette drives.

Are the diskettes THEMSELVES different for use in these drives? Do IBM branded diskettes have different labels or part numbers to be "rated" for use in 360 kB or 1.2 MB drives?

user3840170
  • 23,072
  • 4
  • 91
  • 150
Terry King
  • 41
  • 1

1 Answers1

10

If a disk is sold as double-sided, that means that the manufacturer has done some testing on the material that's on both sides of the disk, and confirmed that it meets specifications. If a disk was sold as single-sided, that means that the manufacturer only promises that one side meets specifications. Disks which are sold as single-sided may be any combination of the following:

  1. Disks made from material that was tested on both sides and is in fact good.

  2. Disks made from material that was tested on both sides, and where one side failed the test, and where only one side of the disk would be able to reliably hold data on all parts of it.

  3. Disks made from material that was tested on both sides, and where one side failed the test, but where all defects happen to be in places where a drive wouldn't try to store data anyway (meaning that all parts of the disk that would be used to hold data are just fine).

  4. Disks made from material that was tested on one side, and assembled into "single-sided" disks without testing the reverse side.

Before manufacturers refined their coating processes, a substantial number of "single-sided" disks may have been in the second and third categories. As processes improved, the amount of material failing testing would have gone down, to the point that most disks would fall into the first or fourth categories.

I'm not sure I've ever encountered a "single sided" disk that wouldn't be usable as a "flippy" on machines that didn't use a rotation index sensor, but I would expect that in the early days of 5.25" floppies that might have happened more often.

supercat
  • 35,993
  • 3
  • 63
  • 159
  • 1
    The diskettes IBM sold as "Part# 6023451 Single Sided Double Density 40 Track Soft Sectored" were widely used internally by IBM sites on double-sided drives on PC-XT's

    I was the IBM PC Coordinator for a site with 4000 PC users and I almost never saw diskettes other than those "Single Sided Double Density" versions.

    . I have only one or two "Part No. 6023450 Double Sided Double Density 40 track soft sectored".) Of 100s

    So I believe the IBM labelled "Single Side: diskettes are in fact Case (1) Above "Disks made from material that was tested on both sides and is in fact good."

     – Terry King Jun 11 '22 at 16:49
  • 1
    @TerryKing: Alternatively, it may be that the process that made the material was good enough that testing didn't really matter except for purposes of ensuring that the production equipment stayed in calibration. THough maybe testing might not be as hard as I had been thinking. If one were to use a head the width of a web to record a signal across the entire width of the media, and then a bunch of playback heads placed across the width of the media, each with a circuit to detect any difference between its received signal and a reference, I guess it might be possible to build such a circuit... – supercat Jun 11 '22 at 17:12
  • ...using only two transistors per playback head. If one had two sets of heads that were staggered, one could scan the entire surface of the media without any lateral gaps, and with longitudinal gaps being a small fraction of the wavelength of the recorded signal. – supercat Jun 11 '22 at 17:14
  • Just one query, supercat, regarding bullet point 3. If "all defects happen to be in places where a drive wouldn't try to store data anyway", wouldn't that be usable as double-sided? – paxdiablo Jun 13 '22 at 03:46
  • @paxdiablo: I may be mistaken, but I would expect that it would be easier to coat disks and test the quality of the coating on a continuous flat web than on punched-out disks, with a testing process that was agnostic as to where the boundaries of the disks would be. If a head that confirmed the quality of the coating were placed immediately before the puncher, it could simply feed material until it had seen a continuous section that was good, then punched out a disk, feed until it was another section that was good, punch, and so forth. If only one side was being tested and there were... – supercat Jun 13 '22 at 15:01
  • ...a defect on the other, that defect could end up being very near the outer or inner edge of the punched-out section, beyond what a drive head could reach in ordinary operation. – supercat Jun 13 '22 at 15:01
  • @paxdiablo: Rereading your comment, I think you misunderstood my point #3, so I'll tweak the wording. – supercat Jun 14 '22 at 17:01
  • @paxdiablo: Thinking further, I find myself curious as to whether each side of a roll of material were treated as pass/fail in its entirety, or whether machines would try to punch disks out of good parts of the material, and whether the punching machine would check the material, or whether some other machine would mark bad spots in some fashion (e.g. by punching a near the edge, which could be detected via optical scanner). – supercat Jun 14 '22 at 17:05