Re: Did Commodore cheat with the quad density floppies?

----- Original Message ----- 
From: "Francesco Messineo" <francesco.messineo@gmail.com>
To: <cbm-hackers@musoftware.de>
Sent: Tuesday, January 08, 2019 11:53 AM
Subject: Re: Did Commodore cheat with the quad density floppies?


> On Tue, Jan 8, 2019 at 5:39 PM Gerrit Heitsch <gerrit@laosinh.s.bawue.de> wrote:
>>
>> On 1/8/19 5:07 PM, Mike Stein wrote:
>> >
>> > I think it was a clever hack; if you insist on seeing it as evidence of the incompetence of the designers and engineers who had to build it within the usual constraints, so be it.
>>
>> It was a clever hack. But it also meant you were limited to only 2
>> drives per Controller where the standard back then allowed up to 4
>> drives on a cable.
> 
> I don't really have a strong opinion on this, but I'd like to have a
> better background. How did the various CP/M machines 'solve' this
> issue? Did both motors start together or did they have separate cables
> for each drive?
> I have used a few CP/M systems in the early '80s and some Olivetti
> '70s computers with 8" drives always spinning (all of them, I remember
> they seemed strange) and of course CBM, Apple II computers and a few
> early MS-DOS compatible systems. I remember an NCR computer with 2 x 5
> 1/4 drives, running a custom version of MS-DOS 1.14 I think. Every
> standard newer MS-DOS disk I tried on it, never booted.
> 
> I had at one point a linux box (back in the 1990's) with 3 floppy
> drives: 5 1/4" HD drives can't really write DD 48 TPI media, in the
> sense the saddle (or tunnel) erase head is too narrow and can't
> completely erase a previous track written by a proper 48 TPI head,
> this isn't an issue if you read the DD disk in another HD drive, but
> usually you wanted to read on an older DD drive that would get noise
> from the previous-not-completely-erased track data, so I had 2 x 5 1/4
> drives, one HD, the other DD and one HD 3 1/2 drive. I ended up using
> two floppy controllers for that, even if I remember there was a way to
> actually jumper one drive to have three drives on the same controller.
> I have nowadays two linux box, one with HD drives, one with DD drive
> on it, just in case :)
> 
> (I guess this thread's gone too far away).
> 
> Frank
------------------------------------------------------------
Actually, I think you've brought it back to reality ;-)

I can't speak for all systems, but the CP/M systems I own and am familiar with do indeed run all drive motors simultaneously and you insert and remove disks or just open and close the door as required; that's why many pictures of the day show the operator with one hand on the drive door ;-)

To avoid unnecessary wear on the diskettes if you left the door closed some drives had a head load solenoid normally activated by the drive select signal.

But as I said, these were the wild west days of different, often incompatible 'standards' and ways of doing things, so it would not surprise me if some systems did use separate cables.

Later 8" and 5 1/4" drives did have jumpers that optionally allowed separate control of the motor, usually using the drive select signal and a delayed turn-off to avoid unnecessary starting and stopping.

As to mixing drive types, most normal PC controllers were either low density (DD/QD) or high density, but rarely both; the original PC controllers could use four drives, as time went on most controllers & BIOSes could only handle two, then only one, and finally none at all today.

But aftermarket controllers were available from DTC and others that could control all four drive types, either replacing the main controller or in addition to it using the secondary address; using the secondary address theoretically gives you eight possible drives, which was sometimes actually used for bulk duplication of diskettes.

Here's a good overview of the various configurations in the PC/DOS world:

https://www.seasip.info/VintagePC/floppies.html