 The magazine of the Melbourne PC User Group Media Mixing and Matching - Why It Fails! Tom Coleman |
 |
0nce upon a time there may have been Single Density diskettes. I don't know, I was not computing in those days. It was probably back when mainframes had 256K of RAM and used 8 inch diskettes. However most of us grew up with Double Density diskettes and as the technology improved High Density and Quad Density have become part of our environment.
Most of this article discusses 5 1/4 inch diskettes. I will toss a tit-bit to the 3 1/2 inch users at the end.
There is an intuitive feeling that High/Quad Density (H/4D) is in some way superior to the lowly Double Density (2D). This feeling is tied to the purpose these diskettes were manufactured to fulfil. As they were meant to store well over a megabyte of information they must be able to store the much stronger signal coming from the higher capacity diskette drives.
Holding more information means they must have a stronger signal and thus be more sensitive, doesn't it.
Wrong! NO! Sorry; try again. Back to square one and do it again. You have flunked. Failed.
The H/4D diskette is only HALF AS SENSITIVE as the old 2D diskette.
The way they get more information on the diskette is mainly by cramming the tracks closer together. The information is written as small magnetic disturbances in the diskette coating. Little magnets if you like. Now the closer together you place these magnets the more they interfere with each other until they start to corrupt each other and data is destroyed.
To overcome this effect they store the signal on a less sensitive media so that it is less responsive to the adjacent tracks.
This leads us to the moral of this story. You can save money by putting 360K 2D diskettes into your AT 1.2meg drive. Unfortunately the signals are written at full strength on a sensitive media and over the next few days/weeks/months they bleed across and corrupt the neighbouring tracks.
Your data magically vaporizes to condense as "Non System Disk" or "Abort, Retry, Fail" or "Track 0 bad" or some other encouraging communication from DOS.
That's not all. The 1.2 MB drive writes with a much stronger current so the signal you lay down is much stronger than the 360K signal. If you now write to the diskette in a 360K drive, the weaker signal is not enough to wipe out the existing strong signal so you get your new signal and bits of the old one as well when you come to read the diskette.
This usually causes the computer to sulk.
If you reverse the situation where you put even a new H/4D diskette in a 360K drive, perhaps in the mistaken belief that you are using a higher quality diskette, you write to a low sensitivity disk with a weaker current and the end result is a signal so weak that it cannot be read.
Even if it is readable, it is at best one quarter the strength of what you would get if you had used 2D diskettes. It is nothing like the strength that the read heads expect to find.
To complicate the issue further, the 360K drive lays down a track that is twice as wide as the 1.2 MB drive. This means that the 1.2 MB drive cannot completely overwrite a 360K track. Thus data overwritten by a 1.2 MB drive may scramble the new data when it is read by a 360K drive.
Generally none of the above applies to 3 1/2 inch diskettes which are slightly more consistent in their setup.
However, you need to be aware of what kind of drive you use if you are going to be swapping between 720K and 1.44 MB 3 1/2 inch drives.
This particularly applies to the IBM models 50, 60, 70 and 80. The drives in these machines do not have the switch that senses the media sensing hole found on the 1.44 MB 3 1/2 inch drives. This is almost exclusive to IBM and not many other manufacturers have drives that cannot sense the difference between the two kinds of disk.
The result of this is that some machines cannot tell the difference between 720K media and 1.44 MB media with the problems of writing with an inappropriate current.
The message for all of us that we should buy and use the diskettes our drives were designed for. If we are going to swap diskettes between different drive densities then we should anticipate some problems that can be avoided with a little thought. Even reformatting a really psychotic disk may not cure it, as formatting writes to the disk and the current or track width may be still be wrong on just some parts of the disk. You may have to bulk erase it, and even that has been known to fail.
Who was it that said "Standards stifle innovation" - it makes you yearn for the good old days when IBM set the standard and everyone followed.
Hang on! These drives are all IBM standard.
Reprinted from the April 1990 issue of PC Update, the magazine of Melbourne PC User Group, Australia
