💾 Archived View for spam.works › mirrors › textfiles › computers › fangrev3a500 captured on 2023-12-28 at 17:16:45.
⬅️ Previous capture (2023-06-14)
-=-=-=-=-=-=-
From luth!sunic!mcsun!hp4nl!telmail!neabbs!johans Wed Nov 1 12:30:45 MET 1989
Article 36064 of comp.sys.amiga:
Path: luth!sunic!mcsun!hp4nl!telmail!neabbs!johans
>From: johans@neabbs.UUCP (HANS VAN DER WERF)
Newsgroups: comp.sys.amiga
Subject: Fatter Agnus
Message-ID: <237122@neabbs.UUCP>
Date: 30 Oct 89 18:28:14 GMT
Organization: NEABBS multi-line BBS +31-20-717666 (13x), Amsterdam, Holland
Lines: 49
Hi there superfat Agus fans,
here is the story of upgrading my A500 to 1MB Agnus.
After buying the Fat Lady (dfl 150,-) i opened up my A500 (for the
first time). I assumed that i had a REV 5 board, so the transition
should be fairly easy. What i found after removing all screws and
and shielding defies description. First of all there was no JP2
between the ROM and the CPU (oink), second there was no REV number on
the board (what ???), thirdly the _ROMEN pin on the Kickstart ROM was
torn out of its socket. Soldered to it was a blue wire wich on the
other end connected to a 3D mounted transistor (shiver). After uttering
some strong words i also removed the bottom shielding. This revealed
some more manual modifications, and last but not least the REV number.
The REV number said 3. After sitting down and taking a deep breath i
tried to think of a sunny side of all this. All i could come up with
was that this board might be a collectors item by now (anybody
interested???).
Now how to put Mrs. Obese in there? Since there was no JP2 i made my
own. What JP2 does on REV 5 boards is connecting A19 on Agnus to A23
or A19 on the CPU address bus. My Agnus A19 was hardwired to A23 on
the CPU address bus. So i carefully cut this connection on the bottom
side of the board between the CPU pins. Then i soldered in a new
connection from the CPU A19 (pin 47) to where the previously cut
connection runs through the board from bottom side to top side, thus
effectuating a connection between CPU A19 and Agnus A19. The rest is
fairly straight forward. The _EXRAM signal on GARY must be kept high
thus preventing it to detect the RAM expansion. I did this by cutting
the connection near the GARY socket (pin 32) on the top side of the
board. The connection to a pullup which keeps pin 32 at VCC is still
intact since it runs under the socket in the opposite direction.
Now the only thing left to do was replacing HER. Ahh, wait a minute
wasn't there something about Agnus pin 41 and PAL/NTSC. Yes, Agnus
pin 41 should be kept at a high level to force a PAL display. Because
i live in Europe i'm kinda attached to PAL. So i bent the pin 41
inside the Agnus socket outward (careful now), preventing it from
forcing the PAL/NTSC input to GND level. After putting Humpty Dumpty
back together again, i crossed my fingers and turned on the power ....
It worked. Avail reported somewhat over 1040K chipmem, hmm nice.
I'am still in the process of trying out what software works, and what
doesn't anymore. Up till now only ACT OUT refused work altogether,
actually it GURUed. Furthermore, it is a pity that you must have fast
ram to use DPAINT-III's animation features. By the way, is there somebody
out there who has a list of programms that freak out when a 8272(A) is
installed ???