Amiga 1200
Additional pictures. Note that loading the page could take a while...
(pictures copyright by M.A.Grundke)
 |
serial number:
A1200 GR
8510693365067002008585 (upside down sticker)C= COMMODORE
MODEL A1200 ID.NO.530311 |
 |
A1200 REV 1D.1 board overview:
first thing you see, moving from left to right, is the PCMCIA-slot, left below the internal IDE harddisk connector and the MITSUMI RF modulator. Right from PCMCIA and four smaller DM74LS bus transceivers, are three bigger chips, LISA (manufactured by HP, topmost chip at U4), AA-GAYLE (middle, at U5) and 8374-ALICE (U2). Moving right from LISA is the small keyboard MPU (U13), with BUDGIE (the large VLSI-chip, located at U20) below it and the (white) internal keyboard connector above.
Right-below BUDGIE, next to a 28.38MHz crystal are the two Kickstart-ROMs (U6A, U6B), with the (spare) FPU-socket (U0) left-, and the small MC68020FG16 CPU (U1) right-below them. Moving right again from there, you can see the board silkscreening and the 150 pin expansion slot. Above the Kickstart-ROMs, futhermore, are the 4 pcs. 256Kx16 RAMs (U16-U19), and, still above, is one of the two little CIAs. Further right follows Paula (U3), the soundchip, and the second CIA (beneath the internal floppy drive connector, U8). On the rightmost part of case, normally covered by the diskdrive, is some empty space with a blind at the case's back, which was intended for 3rd party
expansion - there were, for example, SCSI-boards connecting to expansion-slot accelerators, that provided an external SCSI connector for the machine's rear. Note that the A1200's board is normally completely covered by a PCB shield, except for Kickstart-ROMs, keyboard-, floppy- and harddisk-connectors.
Note also that only surface-mounted devices (SMD) are used (except for the Kickstart-ROMs) - this highly integrated design was introduced with the A600, to reduce production cost and failure due to socket oxidization. However, like in all SMD-technologies, replacing parts is not that easy, since all chips are soldered directly to the board. At least the Kickstart ROMs are socketed... |
 |
A1200 LED PCB (mounted into the case's upper part), prividing 'POWER' (LED 1), 'H. DISK' (harddisk, LED 2) and F.DISK (floppydisk, LED 3) LEDs. Silkscreening:| A1200 LED | COMMODORE | | REV 1 | (C) 1992 |
| | | GRR | | | BF/EG | | | | | NINES | |
|
 |
board silkscreening, part 1:
CHANNEL Z
A1200 REV 1D1
C= COMMODORE ELECTRONICS LTD.
(C) COPYRIGHT 1992 ALL RIGHTS RESERVED
GR/BF/EG/CG/TC/DF/DF/FISH NINES
HMC/LW/CW/GUAY/CW/DUNN/JOEP/JH
ICDES/ICLAYOUT/HK/P.A./Q.A./CATS/DOC./SOFTWARE
FCC ID BR9530311Before it, the 150 pin internal expansion slot, which can be accessed though a 'trapdoor' from below the machine. The expansion slot is not pin-compatible to the CD³²'s, but contains similar signals, i.e. almost all customchips are directly attached. It makes the A1200's
expansion bus much more versatile than e.g. the A500's, which could only be used for RAM expansion in the 1 MByte address range |
 |
board silkscreening, part 2:
SCH 364717
ASSY 364716
FAB 364719
AWK 364720-04 |
 |
board silkscreening, part 3:
Those were the days when the machines were tested before they left the fab! |
 |
| The Motorola MC68EC020FG16 is the 16MHz, flat-grid package version of the MC68020. The EC-version's address-bus is limited to 24 bit, so it can address a maximum of 16 MBytes RAM (it's still got a 32 bit data-bus, however). The very same CPU is also used in the Amiga CD³². If you needed more RAM or speed, there were lots of 3rd-party CPU expansion boards, with MC68030, MC68040 or even MC68060-CPUs and up to 128 MB 32bit RAM |
 |
| These spare pads could easily be used to add a PLCC FPU socket to the A1200. An 68881-FPU (which fits the 68020 perfectly) would greatly improve floating-point performance. However, there was (officially) never an A1200 that came pre-equipped with FPU, because of the cost - for a homecomputer, you didn't need a FPU these days. Interestingly, however, my german A1200 manual states that the machines comes with an FPU (so it was definitely planned)... By the way, see the small IC to the sockets left? I don't know what it is, but one of its pins is wired to BUDGIE - purpose unknown. Must be a 'hot fix' for the REV 1D.1 mainboard... |
 |
| The CSG 8374 ALICE is 32 bit-version of the CSG 8375 'Fat Agnus' used in the ECS-machines; it was of course adapted to work with the AA-chipset. Although many people believe it to be technically possible, ALICE (in the current state) cannot access more than 2 MBytes ChipRAM - those rumours were spread because of an '8 MByte ChipRAM jumper' in early Amiga 4000 schematics (which didn't have a function, by the way). But the later AAA-prototypes are said to be capable of addressing 8 Mbytes ChipRAM, finally... ChipRAM, by the way, is the type of RAM the Amiga-architecture uses for storing graphics- and sound-data,
and the coprocessors (ALICE, LISA, PAULA) simply cannot access FastRAM, in order not to disturb (and slow down) the CPU |
 |
| The A1200's RAM bank, consisting of 4 pcs. Hitachi HM514260AJ8 256Kx16-DRAMs. Left and right from the chips you can see soldering pads on the board. These were planned for internal memory expansion and battery-backed up clock (the three pads right below the RAMs). Also interesting is that the board could alternatively be fitted with 256Kx8 (=1 MB ChipRAM, e.g. if 256Kx16 DRAMs would've become expensive), and later be expanded to the full 2 MBytes (A1201 memory expansion). From what I've heard, this was only true for early production machines & prototypes (do final A1201 expansions exist?) |
 |
| These are the two 256KB Kickstart-ROMs (v39.106 = Kickstart 3.0) of the Amiga 1200, C391523-01 and C391524-01. Kickstart 3.0 was also available for the OCS- and ECS-Amigas, in one ROM (with the same content, of course). Note that the later A1200s (under AMIGA Technologies brand) came with Kickstart 3.1 (versions 40.68, 40.70), which fixed some bugs in the graphic libraries and allowed the use of AmigaOS 3.5 and later 3.9.
Also interesting is that the ROMs are DIP40 packaged, but the sockets are DIP42 - appearently, the board could also take differently organized Kickstart ROMs (the same versatility as for RAM) |
 |
| The machine's keyboard controller (or 'keyboard MPU'), CBM 391508-01 REV 0. It replaces the CBM 6570 and some logic, which older machines contained on their keyboard PCB. In the A1200, however, the keyboard controller is now part of the mainboard. Compared to its ancestor, the new chip also reduces power consumption. Should you think 'man, that chip's label looks strange', don't blame me - the letters really look that weird. Maybe the creators were drunken? |
 |
| LISA (CBM 391227-01) is a 32 bit-version of MOS 8373 ('Denise'), modified for the use with AA, and, an interesting detail, manufactured by Hewlett Packard in 1.5 micron CMOS process. The new chipset's higher bandwith allows resolutions up to 1280x960 in HAM8-mode (262.144 colors at once), and refresh rates up to 73 Hz. It is also used in Amiga 4000 (T) and CD³² |
 |
| Paula (here CSG 8364R7PL in plastic leaded chip carrier-package, PLCC) has never been changed in Amiga-history since it was first used in the Amiga 1000 (a great complaint among Amiga-users, by the way). It's still got 4 independant hardware sound-channels (8 bit mono), which can be combined to 2 sound-channels in 16 bit stereo. Note that the chips soldering pads look a little worn - seems that it gets quite hot there upon operation of the machine (and reminds me of early Intel P5-60 that soldered themselves out...) |
|
| Don't recognize old friends when you see 'em? Yes, these are the A1200's CIAs. Comparing the CSG 8520PL to its ancestor, the DIP-packaged MOS 8520, not much has changed. It's now SMD (PLCC-package, that must be what the 'PL' stands for) and smaller, but remains unchanged in any other way (well, most likely some adoptions to the AA-chipset have been added).Speaking of ancestors, the 8520 and its derivates are direct successors to the MOS 6526 used in Commodore's 8 bit designs |
 |
| AA-GAYLE doesn't only control the PCMCIA-slot and contains an IDE controller logic, it also controls the floppy-motor ('stepping'-capability etc.), and does address decoding. It includes all functionality of the OCS-'Gary', and is basically an AA-adapted version of the A600's GAYLE. And again, if you look closely, you can see heat traces on the board, indicating that it's getting very hot near GAYLE |
 |
| BUDGIE (CBM 391425-01) is a VLSI-logic-chip responsible mainly for controlling transfers and I/O access on the chipset's 32 bit-bus, and is interface between CPU- and chipset buses. It's also where that green wire from a chip near the FPU socket ends (or begins, whatever). As mentioned earlier, I've not yet found out about it, but it seems to be some kind of 'hotfix' for the revision 1D.1 board |
 |
| The MITSUMI RF-modulator (left) and the ADV101KP30 video digital/analogue converter (small chip to the right). Between them the composite jack (yellow). The RF modulator outputs a signal suiting your TV's 'antenna in'. The D/A converter converts LISA's digital video signals to an analogue RGB-signal for a monitor |
 |
| Also well-known from all other Amiga-designs are the MC1488D/ MC1489D serial line driver & receiver (in that order). Like in the other designs, they're used for creating RS-232 compliant voltage levels for input and output. The difference to the older machine is just the package - instead of the older DIP-package, these chips come in an even smaller surface-mountable fashion, SOIC-packaged ('small outline integrated circuit') |
 |
| Standard 16 bit PCMCIA-cards are usable in the A600 and A1200's PCMCIA-slot, which is PCMCIA 2.0-compliant. Thus many cheap ethernet- or memory cards became available for the Amiga-user. However, these memory-cards only provided slow 16-bit FastRAM, and due to a bug in GAYLE, most ethernet cards were only recognized while the machine was running (not when turning it on again!). Notable is that A600 and A1200 were the only Amigas with PCMCIA ever. Before it, SOIC-packaged, are four DM74LS; since they're directly connected to the PCMCIA-signals, they must be bus-transceivers to handle bi-directional communication with the A1200's system bus |
 |
| This is the A1200 case's spare room already mentioned earlier, lying directly under the integrated diskdrive. It is ideal for SCSI-controllers and flickerfixers, because of the opening to the machine's back (here still covered by the plastic blind). Although there were numerous 3rd party devices fitting there (or devices that just used the case opening to add an external connector), there has never been an official Commodore product of that kind. Probably it was never planned? |
Also see the Amiga CD³²'s board design!
