Re: Software for MS-DOS 1.25

From: Mia Magnusson <mia_at_plea.se>
Date: Tue, 17 Oct 2017 05:39:30 +0200
Message-ID: <20171017053930.0000398c@plea.se>
Den Sun, 15 Oct 2017 18:10:20 +0200 skrev Michał Pleban
<lists@michau.name>:
> Mia Magnusson wrote:
> 
> > A compromise could be to have a character set based on what's
> > actually used in real life. Very few if any programs use for
> > example the smileys, and block graphics might not need inverse
> > mode, but it would be good to have inverse mode for atleast ascii
> > code 32-127 and preferable a subset of national character (which
> > each user should be able to select when generating a font eprom
> > image with some "generate" utility).
> 
> I would definitely like to have a character set that at least has the
> backlash character (used in MS-DOS as path separator), pipe and curly
> brackets. These seem quite important to me.

We seem to agree on this thing :)

> > As Ruud already stated, a XT class machine can use 1.2MB drives if
> > you provide appropriate driver (usually with a small extra rom).
> > But that is just a limitation of the XT bios. DOS will happily use
> > whatever it gets provided with.
> 
> But what does this driver actually do? How does it tell DOS "hey, now
> you have a 1.2M drive instead of 360k"?

The driver does two things. One is being able to select high density on
the FDC hardware. The other is as you expect tell DOS that the drive
has more sectors (and more tracks if it's 5,25"). I'm not sure how that
is done, but it's probably done the same way as in an IBM AT.

> > True. Maybe a great mod would be to have a daughterboard that both
> > moves the CBM ram and also adds more RAM? I don't know if it would
> > be even possible to add ram so it would be visible for the CBM too 
> 
> No, because there are chips on the CBM-II mainboard which multiplex
> the address bus between 6509 and 8088.

Oh, I thought that all those buffers/latches on the 8088 board did half
of this, and the stuff on the mainboard did the other half, i.e. the
8088 board connected directly to the RAM IC:s. Too bad that it isn't.

Some other expansion to get 1MB (or actually almost 1MB) would then be
a good thing.

Does all known CBM-II have the same layout of the mainboard? If so you
might be able to make a card that sits in the sockets for dram and some
other circuits, and has larger ram ICs (or more ram ICs) and the
required logic.

> But creating an 8088-only RAM extension should be straightforward, if
> we tap into the RAM extension connector _and_ the PLA which
> fortunately should be socketed.

It seems easier to do something with the F****-decoder so it disables
access to CBM-II mainboard for all adresses above 3FFFF and then just
add a 512k static ram IC decoded from 40000 to BFFFF. Such an
add-on-board could sit in the BIOS socket with some wires to the 8088
board.

Maybe it would be good to do something with the adress bus so local ram
and shared ram end up at other places, for example shared ram at
B0000-BFFFF?

> > But how does disk i/o work? I would have thought that the 8088
> > would be shut down completely giving memory access back to the 6509
> > while it does it's thing, and then starts up the 8088 again. 
> 
> That's exactly what happens.

Oh. If memory is added locally to the 8088 then atleast some I/O needs
to be double buffered.
 
> > Or is it done
> > sending one sector at a time via the 6526 interfaces while
> > both CPU's running at the same time?
> 
> Only control bytes (sector number, buffer address) are sent via 6526,
> the 6509 places the disk sector contents directly in the DRAM.

Ok.

> > Yeah, as I understand it there is 2k of ram for the 6509 on the 8088
> > board, that's a bit tight to have any code not strictly neccesary.
> 
> But you also have existing 1k at $0400 which, as I understand, is used
> for bootstrap code. If we move more bootstrap code to the 8088, we can
> re-use this RAM.

Yes, that's an amazing 3k :)

> > But I assume that writing a new BIOS on the 8088 side would also
> > mean at least changing the 6509 code somehow. Then IRQ0 and IRQ7
> > can be used freely.
> 
> No, you don't have to change 6509 code if you don't want to provide
> new IO functions to the 8088. The extension could theoretically work
> with the 6509 code unmodified.

Yes, I understand that it's not strictly neccesary, but it might seem
like a good idea anyway.

-- 
(\_/) Copy the bunny to your mails to help
(O.o) him achieve world domination.
(> <) Come join the dark side.
/_|_\ We have cookies.

       Message was sent through the cbm-hackers mailing list
Received on 2017-10-17 04:04:03

Archive generated by hypermail 2.2.0.