This has been a bit of an ongoing project for me over the last month or so.
I met a bloke on another forum who had some old miggies sequestered away in his garage.
I bought 2 A1000's off him.
Cosmetically they've both seen much better days. This is the first one I looked at:
The inside seems to be in good order
I was a bit surprised when I plugged in and was greeted with the 'workbench 1.3 disk' screen.
I thought the 1000's needed a kickstart disk first.
A closer look at the motherboard revealed this:
'Phoenix Enhanced Motherboard for the Amiga 1000'
This is in fact the A1000 Phoenix upgrade board of which only a limited quantity were made.... it's an original 1989/1990 model and not one of the later remakes.
I was also fortunate enough to pick up this sidecar expansion:
It has 1MB of Fast RAM which is quite useful for my 500 and 1000's. The disk controller seems to be a proprietary 26-pin interface, JVC made 26-pin drives at one point so I suspect that's what its for.
Here's the two keyboard I got with it. They are foul - I think one's had milk spilled in it at one point because it was absolutely rank when I pulled it apart. I've tried to clean it as best I could. Both of them had quite a few faulty keys.
here's some more shots of the board.
There are heaps of messages and greetings all over the underside:
Sorry for the poor quality. This last one says 'Sorry sheldon we did it anyway' and 'the various user groups who supported us'
This one says 'To the 540 depositors who made it all possible, my thanks and in particular, Margaret Wilson, Jonathon Potter, Mike Chow and the others scattered around the board'.
The phone number is an old 7 digit number, in the mid 90's Australian phone numbers were changed to 8 digits.
You might notice there are a LOT of empty sockets. Unfortunately all the cool features of this board required PAL (Programmable Array Logic) chips which were programmed at the factory, and included with the board or sold as an optional upgrade.
I have no idea why this board is missing them. Either it was raided for parts or it was purchased as an absolute barebones model.
I managed to track down Andrew Wilson, the guy who designed the board at Phoenix Technologies. I spoke to him on the phone but he isn't in a position to help me at the moment. He did point me to a1k.org where I uncovered some further information on it.
On the forums there they have the Phoenix CD for download which includes heaps of utils and information including the user manual and jumper settings.
I managed to deduce that I was missing these chips:
U59 - Kickstart switcher
U21 - FPU
U60 - 2MB RAM
U70 - B2000 expansion slot
SCSI - U27, U23, U25, U26, U24, Also requires EPROM in U31 and actual SCSI controller chip in U30
Thankfully one of the users was kind enough to dump some of the chips for me
So I now have the .JED files to program replacements and the scsi autoboot ROM. Here they are for anyone that's looking for them.
All the other SCSI controller chips etc can be found online.
So to that end I've bought a cheap and nasty G540 USB universal programmer off ebay and some rewritable GAL chips from futurlec. Waiting for them to arrive so I can try them out
Unfortunately the U60 chip, which allows a 1MB Agnus to interface to 2MB of RAM won't work in a rewritable GAL. It will only work in a write-once PAL for reasons unknown (Andrew Wilson himself said he never figured this out). It's impossible to find the required type of PAL that hasn't already been programmed, and even if I did you can't program them with a normal programmer.
Of course if I managed to track down a 2MB Agnus chip that fits, the docs say that would work too, but these are pretty rare. I might have to just make do with 1MB.
The board has 1MB of RAM installed which I've configured as all chip memory. You can reconfigure it for 512 chip and 512 slow (similar config to an A500 with the 512KB trapdoor expansion) while the sidecar expansion is 1MB fast. Funnily enough, if I configure for 512KB/512KB Chip/Slow, sysinfo reports about 3% faster speed than 1MB/1MB chip/fast? I thought the whole thing with slow memory was that it was like chip memory (controlled by agnus, which introduces a delay as the CPU can't access it every cycle) but processor had exclusive use of it, while fast mem was controlled by processor and could access it on every cycle so there was no delay. Strange.
Anyway, while I'm waiting for the chips and programmer to arrive I've been keeping busy. I've stripped the machine and cleaned it, and tried to retrobright it. I tried using cling wrap to prevent the retrobright from drying out unfortunately this just made it brighten the plastic unevenly (wherever there was a crease in the cling wrap). I work full time and I don't want to leave it out in the hot aussie sun from 8-5 without being able to check on it every hour or so... and I've been busy on weekends so its been slow progress on this front.
In the meantime I've managed to get a working keyboard.
The one that was missing the spacebar had only 4 or so dead keyswitches plus the missing switch for the spacebar. I managed to get the others working with a few squirts of WD40 to loosen them up. I managed to desolder 5 working ones off the other board (which had missing keys all over the place but funnily enough QWERTY at least worked fine) and solder them onto the new board.
I've also since replaced the dead clock battery with a coin cell holder so as to make replacements easier in the future
I've also replaced the corroded old RCA jacks with nice, clean new ones
Everything is humming along nicely
It definitely sounds a lot clearer with the new RCA jacks.
I also got the steel wool and some Autosol and polished up those metal brackets on the edge connectors, they are still dull but at least they aren't brown and rusty any more. I don't think these serve any purpose anyway? the sidecar doesn't even touch them when its installed.
I tried to polish up the RF shields but they were too far gone. After an hour of furious scrubbing and polishing they still looked crap so I've put them aside for now.
So now its basically just waiting for the chip and programmer to arrive, at which point I hope to get the FPU, kickstart switcher and SCSI working. Planning to get a 50 pin SCSI card reader so I can install workbench to it.
Most B2000 CPU expansion card don't look like they'll fit inside the standard A1000 case so I'll just forget about that for now.
However, I am missing a front chipram cover for the second A1000 case. I was going to post a wanted to buy thread here but someone's already got one open and hasn't had any offers, so it would be bad form for me to start my own thread too. Looks like replacement plastics will be very hard to come by
I met a bloke on another forum who had some old miggies sequestered away in his garage.
I bought 2 A1000's off him.
Cosmetically they've both seen much better days. This is the first one I looked at:
The inside seems to be in good order
I was a bit surprised when I plugged in and was greeted with the 'workbench 1.3 disk' screen.
I thought the 1000's needed a kickstart disk first.
A closer look at the motherboard revealed this:
'Phoenix Enhanced Motherboard for the Amiga 1000'
This is in fact the A1000 Phoenix upgrade board of which only a limited quantity were made.... it's an original 1989/1990 model and not one of the later remakes.
I was also fortunate enough to pick up this sidecar expansion:
It has 1MB of Fast RAM which is quite useful for my 500 and 1000's. The disk controller seems to be a proprietary 26-pin interface, JVC made 26-pin drives at one point so I suspect that's what its for.
Here's the two keyboard I got with it. They are foul - I think one's had milk spilled in it at one point because it was absolutely rank when I pulled it apart. I've tried to clean it as best I could. Both of them had quite a few faulty keys.
here's some more shots of the board.
There are heaps of messages and greetings all over the underside:
Sorry for the poor quality. This last one says 'Sorry sheldon we did it anyway' and 'the various user groups who supported us'
This one says 'To the 540 depositors who made it all possible, my thanks and in particular, Margaret Wilson, Jonathon Potter, Mike Chow and the others scattered around the board'.
The phone number is an old 7 digit number, in the mid 90's Australian phone numbers were changed to 8 digits.
You might notice there are a LOT of empty sockets. Unfortunately all the cool features of this board required PAL (Programmable Array Logic) chips which were programmed at the factory, and included with the board or sold as an optional upgrade.
I have no idea why this board is missing them. Either it was raided for parts or it was purchased as an absolute barebones model.
I managed to track down Andrew Wilson, the guy who designed the board at Phoenix Technologies. I spoke to him on the phone but he isn't in a position to help me at the moment. He did point me to a1k.org where I uncovered some further information on it.
On the forums there they have the Phoenix CD for download which includes heaps of utils and information including the user manual and jumper settings.
I managed to deduce that I was missing these chips:
U59 - Kickstart switcher
U21 - FPU
U60 - 2MB RAM
U70 - B2000 expansion slot
SCSI - U27, U23, U25, U26, U24, Also requires EPROM in U31 and actual SCSI controller chip in U30
Thankfully one of the users was kind enough to dump some of the chips for me
All the other SCSI controller chips etc can be found online.
So to that end I've bought a cheap and nasty G540 USB universal programmer off ebay and some rewritable GAL chips from futurlec. Waiting for them to arrive so I can try them out
Unfortunately the U60 chip, which allows a 1MB Agnus to interface to 2MB of RAM won't work in a rewritable GAL. It will only work in a write-once PAL for reasons unknown (Andrew Wilson himself said he never figured this out). It's impossible to find the required type of PAL that hasn't already been programmed, and even if I did you can't program them with a normal programmer.
Of course if I managed to track down a 2MB Agnus chip that fits, the docs say that would work too, but these are pretty rare. I might have to just make do with 1MB.
The board has 1MB of RAM installed which I've configured as all chip memory. You can reconfigure it for 512 chip and 512 slow (similar config to an A500 with the 512KB trapdoor expansion) while the sidecar expansion is 1MB fast. Funnily enough, if I configure for 512KB/512KB Chip/Slow, sysinfo reports about 3% faster speed than 1MB/1MB chip/fast? I thought the whole thing with slow memory was that it was like chip memory (controlled by agnus, which introduces a delay as the CPU can't access it every cycle) but processor had exclusive use of it, while fast mem was controlled by processor and could access it on every cycle so there was no delay. Strange.
Anyway, while I'm waiting for the chips and programmer to arrive I've been keeping busy. I've stripped the machine and cleaned it, and tried to retrobright it. I tried using cling wrap to prevent the retrobright from drying out unfortunately this just made it brighten the plastic unevenly (wherever there was a crease in the cling wrap). I work full time and I don't want to leave it out in the hot aussie sun from 8-5 without being able to check on it every hour or so... and I've been busy on weekends so its been slow progress on this front.
In the meantime I've managed to get a working keyboard.
The one that was missing the spacebar had only 4 or so dead keyswitches plus the missing switch for the spacebar. I managed to get the others working with a few squirts of WD40 to loosen them up. I managed to desolder 5 working ones off the other board (which had missing keys all over the place but funnily enough QWERTY at least worked fine) and solder them onto the new board.
I've also since replaced the dead clock battery with a coin cell holder so as to make replacements easier in the future
I've also replaced the corroded old RCA jacks with nice, clean new ones
Everything is humming along nicely
I also got the steel wool and some Autosol and polished up those metal brackets on the edge connectors, they are still dull but at least they aren't brown and rusty any more. I don't think these serve any purpose anyway? the sidecar doesn't even touch them when its installed.
I tried to polish up the RF shields but they were too far gone. After an hour of furious scrubbing and polishing they still looked crap so I've put them aside for now.
So now its basically just waiting for the chip and programmer to arrive, at which point I hope to get the FPU, kickstart switcher and SCSI working. Planning to get a 50 pin SCSI card reader so I can install workbench to it.
Most B2000 CPU expansion card don't look like they'll fit inside the standard A1000 case so I'll just forget about that for now.
However, I am missing a front chipram cover for the second A1000 case. I was going to post a wanted to buy thread here but someone's already got one open and hasn't had any offers, so it would be bad form for me to start my own thread too. Looks like replacement plastics will be very hard to come by