[Coco] ?PEEK(&HFF90) RETURNS 126 ON MY UNUPGRADED 26-3124?
mark at cloud9tech.com
Wed Apr 25 15:42:09 EDT 2007
Man yo are hitting the first for everything. I'll maqke sure in the
future that the IC is in more protected place. The Antistat foam and
bubble packaging usually is enough.
The MPI is simple. The equations are simple. The interface and design
are VERY problematic.
Quoting coco at yourdvd.net:
> got the gal replacement for the pal. it was squished by the post office
> - looked like they stepped on the sucker, but i straightened all those
> pins out plugged 'er in and voila - coco 3 mpi :-)
> While I had the 3024 MPI opened, I noticed just how different it is from
> the 3124. The 3024 should be easy to build an exact clone but using your
> gal instaed of the tandy pal - all other circuits at first glance appear
> to be standard ttl and some power supply stuff. i made one some time ago
> using 74ls138's and 74ls374 latches, etc. to clone the mpi - a lot of
> discrete circuitry. had 8 slots, but otherwise worked like a coco 3
> mpi. I had planned to replace the 138's with 154's and make 16 slots
> (major overkill), but then i popped by paul bartons site and he
> mentions the upper bit of both nybbles being used as an enable disable
> of the associated signal. I didn't think the multi-pak worked like that
> and didn't account for it in the thing i built, but it still worked :-)
> one day i'll dig all this junk i made out of storage and stick the
> schematics somewhere.... r
> p.s. i made the slots out of a short piece of ribbon cable and a 40 conn
> ids connector.
>> -------- Original Message --------
>> Subject: Re: [Coco] ?PEEK(&HFF90) RETURNS 126 ON MY UNUPGRADED 26-3124?
>> From: Mark Marlette <mark at cloud9tech.com>
>> Date: Tue, April 24, 2007 10:57 am
>> To: Robert Gault <robert.gault at worldnet.att.net>
>> Cc: coco at maltedmedia.com
>> I don't have the data right in front of me....hold on let me check
>> C-9's CVS for the mpak manual...
>> Nope...have to add it...
>> I looked at my source to the .abl code for the GAL, 26-3024 update.
>> The new range of the MPI is $FF40-$FF7F.
>> If I looked at the 3124 mod that I have already documented at home I
>> could quickly figure out what it's original range was. From my web
>> page it appears that the MPIs unmodified range went into the GIME's
>> address space($ff90).
>> The peek makes sure that it is not. This si the first I have heard
>> that it isn't working.
>> Done this MANY years ago so I don't recall.
>> Sorry for the slow response. Was gone over the weekend and have many
>> business/list emails to reply to.
>> Mark Marlette wrote:
>> What GIME is in your CoCo?
>> Test performed on the same CoCo?
>> I'm sure I have done this test on both MPIs. Been too long ago.
>> At 4/21/2007 02:12 PM, you wrote:
>> Marks test of ?peek(&hff90) returns 255 on all of my 3024 multi-paks,
>> however it returned 126 on my 3124. I opened the 3124 and there is no
>> satellite board nor any piggy backed chips nor any mod of any kind - i
>> can't see that this pak has been upgraded. Does this test only work for
>> the 3024's? thanks - Rob
>> I just tried a test with a Coco3 and an MPI 3024. Without the MPI the
>> PEEK returned 126 and the same with the MPI.
>> What are you trying to determine with this PEEK, whether the MPI was
>> upgraded for Coco3 use?
>> Coco mailing list
>> Coco at maltedmedia.com
> Coco mailing list
> Coco at maltedmedia.com
More information about the Coco