Drive capacity names (Was: WTB: HP-85 16k RAM Module and HPIB Floppy Drive

Fred Cisin cisin at xenosoft.com
Thu Nov 16 17:27:53 CST 2017


>>> No, the 9122C has two high-density, two-sided 80 cylinder drives. A drive 
>>> has no capacity, this is the function of the on-disk format.
>>> ;-)
>> 
>> "high-density" is even more meaningless than referring to them by their 
>> capacity in a given format.  It is a BOGUS marketing term!

On Thu, 16 Nov 2017, Christian Corti via cctalk wrote:
> Fred, you should know by now that you don't need to tell *me* the correct 
> definitions and terms.

I know that, but I was addressing the entire group with my rant, and not 
everybody is as closely familiar with these details as you are.

> And with "high-density", I didn't mean the media capacity but the analog 
> recording aspects like coercivity, write current, frequency and so on.

Actually, when speaking about the MEDIA, it is much easier to create a 
name that is both accurate and unambiguous.
For example, with 5.25" disks, we have "5.25 inch with 300 Oersted" and 
"5.25 inch with 600 Oersted".
Of course, if somebody wants to be difficult, there are still variant 
forms, including both 10 and 16 sector hard-sectored, Amlyn 600 Oersted 
with special cutouts for the disk changer, Twiggy, no-notch disks for some 
minor tamper resistance in software distribution, etc.

>> Unformatted capacity would be a more correct nomenclature, although ...
> Unformatted capacity doesn't tell you much without reference to the recording 
> layout, i.e. no. of tracks, modulation, frequency and so on.

True.

>> Some specifications:
>> 5.25" MFM "High Density" was 360 RPM at 500,000 bits per second. (about 1M 
>> unformatted per side)
>
> What about 5¼" FM "High Density" at 360 RPM?

By "Some specifications", I meant specifications of SOME examples of the 
most common form of each size.  I was absolutely not intending it to be 
an exhaustive, comprehensive list of all possibilities.


> The Amiga (more exactly, the "HD" Chinon FZ-357A drives used in Amigas) 
> switched to 150 RPM to keep the raw bit rate at 250kbits/s.

THAT is exactly what I was including as examples in my later "exceptions" 
list.  Although a different disk size, that is the same engineering kludge 
as the Weltec 5.25" 180RPM drive.

>> 3.5" MFM "ED" (vertical recording?/barrium ferrite) were 300 RPM at 
>> 1,000,000 bits per second.  (2M unformatted per side)  NeXT referred to 
>> theirs by the unformatted capacity: 4M, further confusing their users.
> What about FM?

Again, just listing examples of most common, NOT intending it as a list of 
all possibilities that were theoretically possible.  I have never seen an 
ED disk recorded FM, and do not believe that there was ever a commercial 
system that used that.  If you know of one, please give us the details!


>> Can you name another 20 exceptions?   (Chuck and Tony probably can)
>
> Do you want me to start with things like 100tpi drives, GCR, M²FM, 
> hard-sectored and other crazy formats?

It can be a very long list.  I was trying to stick with ones that were 
very close to the main branch of our "current" evolutionary tree, but 
there isn't a clear boundary.  I estimate that there were approximately 
2500 different microcomputer floppy disk formats, with a large portion of 
those being variant forms, not just different choices of number and size 
of sectors, directory location and structure, etc. 
I implemented just over 400 formats in XenoCopy that were straight-forward 
to handle with IBM PC hardware.  Those are not all that could have been 
implemented, nor does it deny the existence of many variants, or 
completely different ones that are not feasable with PC.

> Just accept that I am not as dumb as you may think.

I have NEVER thought that you were dumb.  Everything that I have seen 
of your posts has been competent and well-informed.  But, I don't think 
that you follow what I was attempting to convey.

I wanted to:
1) rant about marketing creating terminology, including "double density" 
and "high density".  And creating a new definition of Megabyte (1,024,000) 
for the "1.44M" format (1,474,560 bytes/1.40625Mebibytes)

2) state my opinion that using the specific one that comprises at least 
75%? of the use of a given configuration as the name for that 
configuration creates a name that is admittedly inaccurate, and fraught 
with exceptions, nevertheless relatively unambiguous, at least to the 
extent that purchases will usually be usable.

If I buy "360K diskette", it will usually be the 300 Oersted 5.25 inch, 
and be the closest of what is available to buy for 87.5K TRS80, 
Apple2, PET, Osborne, PC 160K/180K/320K/360K, DEC Rainbow, Canon AS100, 
Elcompco, Eagle, Otrona, etc. 
Yes, there were people who used 41 or 42 tracks of a 40 track drive, but I 
consider those to be "corner cases", to be considered as alterations, not 
as the main form.
Admittedly, there were differences in testing between SSSD, DSSD, DSSD, 
DSDD, and 48tpi v 96tpi marketing of disks with the same chmical 
formulation.  Purchasing diskettes now for something such as a DEC 
Rainbow, I would settle for the 360K testing.

If I buy "720K 3.5 inch diskette", I expect to receive 600 Oersted 3.5"

If I buy "1.44M Diskette", I expect to receive a "HD" 3.5 inch diskette, 
with about 720 to 780 Oersted.


BUT, as you've pointed out, when we refer to the DRIVE, we can't really be 
certain that it won't be misinterpreted unless we list every spec that we 
expect it to conform to.  Or order by manufacturers model number.
Shugart/Matsushita 455/465/475
Tandon TM100-2/TM100-4
Teac 55B, 55F, 55G, 55FG, etc.
(EXAMPLES.  NOT TO BE CONSTRUED AS A "complete" LIST)

BTW, Tandon made a 100tpi drive (TM100-4M) for Micropolis compatability, 
but many/most? of those are mislabelled "TM100-4" (missing that critical 
'M' modifier!)

--
Grumpy Ol' Fred     		cisin at xenosoft.com


More information about the cctalk mailing list