17 May
1997
17 May
'97
8:41 a.m.
On Sat, 17 May 1997, Tim Shoppa wrote:
I thought the WD-17XX and WD-19XX chips were programed to automatically
get the first track/sector on reset. I seem to remember that the other
controller chips from that time also did this. I always thought this was
the ONLY way to do it 8-) I have often thought that if all the
peripherials (did I spell that right?) did the same thing (at reset) and
made thier *internal drivers* available at boot up - it would be a much
easier world with IO. You would just have to set a switch on the IO
card/device to tell it what CPU it is working with and then be done with
it.
I've
heard incredible things about the Apple ][ Disk drive. One, that it
took only 5 ICs for the hardware, and now only 256 bytes of code to read
from it? I'd really like to see both the schematics and the code. That's
just incredible.
The 256-byte-long bootstrap for the Disk ][ is really quite long,
as the very simple hardware needs extensive help from the software
to read anything from the disk. The very complex hardware/software
interaction is well-documented in Worth's and Lechner's _Beneath
Apple DOS_, still available new from Quality Computers. The
schematics are published in the back of both the Apple DOS 3.2
and 3.3 manuals, though the schematics themselves don't do you
a lot of good unless you know the programming of the bipolar
PROM's used in the controller's state machine.
For comparison, my S-100 boxes with WD1771-derived controllers have
bootstraps that are just over a dozen bytes long. For many PDP-11 disk devices, the bootstrap is only a
couple words.
When you have to toggle the bootstrap in through the front panel every
time you boot, a short bootstrap is extremely desirable.
Of course, in these cases, there is substantially more intelligence
in the device and controller than there is in the Disk ][.
Oh I don't know 8-) Some of them seem pretty stupid when I try to talk
to them. They don't seem to understand that they have to work the way
I WANT them to 8-)
Tim. (shoppa(a)triumf.ca)
BC