Lisa floppy problems

From: Eric Smith <eric_at_brouhaha.com>
Date: Tue Sep 15 16:03:30 1998

I wrote about Lisa vs. Macintosh 400K disk drives:

> No, they are in fact identical, and have auto-eject. The eject when the
> white "power" button on the Lisa is pressed is done by software; the button
> doesn't really have anything to do with the power supply.

Marion.Bates_at_Dartmouth.EDU (Marion Bates) wrote:

> I asked David Craig about this, and here's what he had to say:
>
> "Concerning your recent question about the differences between the Mac
> and Lisa 400k drives as pertaining to auto eject at machine shutdown, my
> understanding is the Lisa drive has extra circuitry to detect when its
> power is turned off and it then auto ejects the floppy. The Mac drive
> does not have this extra circuitry and therefore does not auto eject a
> floppy by itself. The Mac floppy ejection is controlled by the system
> software.
>
> See Sun's Lisa repair manual which talks about this feature.

I don't have Sun's Lisa repair manual (though I'd like to get one);
I have Apple's. The replacement part numbers for Macintosh and Lisa 400K
drive mechanisms are identical. No big surprise, since the drive was
originally designed for the Macintosh, not for the Lisa.

I've replaced broken Lisa drives with Macintosh drives, and they still
eject when the Lisa is powered down.

Furthermore, it can easily be demonstrated that if you pull the power plug
on a running Lisa, the disk doesn't eject.

And they would have needed a whopping big capacitor to store enough energy
to operate the eject mechanism once the power has failed, since there's no
signal from the computer to the disk drive that provides any advance notice
of imminent power loss.

I stand by my claim that the Lisa's power-down eject is done by software.

This was true of the Lisa's original Twiggy drive as well. The schematics
show no sign of any power-fail detection.

> FWIW, the Lisa drive being smarter does not suprise me since the Lisa
> hardware philosophy was to have specific components handle specific
> tasks. That's why the Lisa twiggy drives were smart and did stuff like
> formatting whereas the Mac handles everything about the disk.

That's also why the Lisa was so f'ing expensive. Putting in special
hardware to handle infrequent operations like formatting disks is a
waste of money. Anyhow, they could have used an industry-standard disk
controller chip and saved money over either the Mac or Lisa implementation,
but they didn't because that had an extremely severe case of NIH Syndrome.

> The Lisa's keyboard was also smart and had its own chip to control it --
> COPS, control oriented processing system."

The Lisa's keyboard is no smarter (and arguably dumber) than a standard
PC keyboard. The Lisa keyboard contains a 4-bit National Semiconductor
COPS microcontroller, which almost all PC keyboards contain 8-bit Intel
8048 or 8051-family microcontrollers.

The Lisa I/O board contains another COPS for communicating with the keyboard,
which was overkill. It would have been completely unjustified, except that
they used it for the RTC also. IBM added another Intel 8-bitter (8042) to
talk to the keyboard in the PC/AT, which was only actually needed because
they needed a way for the 286 processor to make some piece of external
hardware generate a processor reset in order to return from protected mode
to real mode.

Don't get me wrong; I think the Lisa is a neat machine, and I have several of
them. But the hardware engineering tradeoffs that were made were completely
absurd, as was proven by the disappointing sales.

Eric
Received on Tue Sep 15 1998 - 16:03:30 BST