Re: 8088 card and bank 15

From: Steve Gray <sjgray_at_rogers.com>
Date: Thu, 26 Oct 2017 18:58:53 +0000 (UTC)
Message-ID: <524962274.4083696.1509044333081@mail.yahoo.com>
I doubt it. The CBM-II Cartridge port was designed for ROM, not RAM. Even the area from $1000-1FFF is for ROM for the "Low Cost Disk Port". In other words, BANK 15 is for system use (KERNAL, BASIC, Cartridge ROM, Disk ROM, I/O, and Screen RAM). Normal RAM would go in BANKS $0 to $E. Bank 0 is not used in the B-Series I believe due to maybe a holdover from the P500's special VIC-II access.
Steve

      From: william degnan <billdegnan@gmail.com>
 To: "cbm-hackers@musoftware.de" <cbm-hackers@musoftware.de> 
 Sent: Thursday, October 26, 2017 1:40 PM
 Subject: Re: 8088 card and bank 15
   
It's possible that it was expected that one install a RAM cart in the user port to so something related to the 8088 card within bank 15 ($2000-2FFF).  Many people with the 8088 coprocessor would also have had the RAM cart
I have a newer one that goes all of the way up to 7FFF for use with the FastBus that allowed for connection to a serial drive like the 1571http://www.vintagecomputer.net/browse_thread.cfm?id=513


Bill

On Thu, Oct 26, 2017 at 12:31 PM, David Wood <jbevren@gmail.com> wrote:

Do you suppose the adder in the 8088 card is partly the reason RAM starts in bank 1 instead of bank 0 on the CBM2 series then?  I had always wondered about this.

On Thu, Oct 26, 2017 at 12:24 PM, <Ruud@baltissen.org> wrote:

Hallo alemaal,


When studying the schematic I noticed a 283 adder. It genererates
the lines BP0..3 that are used to select the needed 64 KB DRAM bank
using the PLA.
The adder adds one to the original address. Why? When the 8088
addresses the the 0Fxxxxh area, say bank 15, it addresses the ROM.
But when addressing 0Fxxxxh, BP0..3 are all zero. But what happens
if the 8088 addresses the 0Exxxxh area, BP0..3 are all one. Or in
other words: bank 15!

I studied the schematic and so far I don't see any reason why it
wouldn't work. So those who have a working CBM-II/8088 card
combination, please try it out.




How did I get to the above:
Michał and I are researching if it is possible to create an improved
8088 card. Why? There won't be that many customers for a remake of
the orignal one. And even that remake will be improved a bit at
least. For example: it will use the better available 27xxx series as
ROM. So the idea was that if we had to offer a card at all, why not
one with extra features?

Before anything else : the improved card will be 100% compatible
with the old one.

One idea was adding an ISA slot. So far it looks that I can do it
with adding just one extra IC, 74LS00.

One of my ideas was adding 512 KB of SRAM to the board. It can be
used as addition to the original RAM of the CBM-II. Having a 256 KB
machine it would mean we have a surplus of 128 KB of RAM. This can
be used as UMB RAM in the 0Dxxxh and 0Exxxxh area. And maybe better,
64 KB in the 0Axxxxh area so the machine can have 704 KB of RAM for
DOS. The other 64 KB can be used in one of areas mentioned before.
At this point I started studying the schematics and noticed the 283
for what it was, raising the question: why?
A 74LS138 plus some jumpers could do the trick.

As said the improved will be 100% compatible with the original one.
But that means we have to give in on some things. One example: IR7
of the 8259 Programmable Interrupt Controller (PIC) is already in
use. So I decided to connect IRQ7 (for the LPT port) to IR3. Having
no IR3 anymore, I decided to connect IRQ3 and IRQ4 together to IR4.
IRQ5 has been connected to IR5 and IRQ6 to IR6.
IRQ6 is used by the floppy disk. A 360 KB floppy can be read not
using DMA. And maybe a 720 KB one as well. (for Michał, I think I
was wrong in a previous email)
I have seen a Z80 and a 6809 system using the 765 W/O DMA so why
not?


--

Kind regards / Met vriendelijke groet, Ruud Baltissen
www.Baltissen.org







       Message was sent through the cbm-hackers mailing list






   

       Message was sent through the cbm-hackers mailing list
Received on 2017-10-26 19:00:18

Archive generated by hypermail 2.2.0.