• Please review our updated Terms and Rules here

The mysterious C1000 MCA to IDE card (alternative HDD solution for 55SX)

PS2Fan

New Member
Joined
Jun 21, 2022
Messages
5
Hi,

On my quest to find an alternative hard drive setup for a PS/2 55 SX I'm reviving (alongside a 30-286 which I posted about the other day), I've stumbled upon references to a Custom Computer Systems' C1000 card, which should be an MCA to IDE adapter.

Great, right?

However... the Internet is almost blank.

1) Does anyone know if these cards are possible to find at all?

2) Would anyone happen to know if schematics are available so one could build one?

Whilst I'm at it, I've also been looking into going the SCSI route as well. 85F0002 cards seem to be available... has anyone tried that in the 55SX? What puzzles me slightly is the card seems to be 32-bit, but apparently it's supposed to work in 16-bit slots as well. I guess leaving the last part of the connector dangling just disables something-something.
 

IBMMuseum

Veteran Member
Joined
Aug 28, 2006
Messages
1,449
Hi,

On my quest to find an alternative hard drive setup for a PS/2 55 SX I'm reviving (alongside a 30-286 which I posted about the other day), I've stumbled upon references to a Custom Computer Systems' C1000 card, which should be an MCA to IDE adapter.

Great, right?

However... the Internet is almost blank.

1) Does anyone know if these cards are possible to find at all?

2) Would anyone happen to know if schematics are available so one could build one?

Whilst I'm at it, I've also been looking into going the SCSI route as well. 85F0002 cards seem to be available... has anyone tried that in the 55SX? What puzzles me slightly is the card seems to be 32-bit, but apparently it's supposed to work in 16-bit slots as well. I guess leaving the last part of the connector dangling just disables something-something.

The 'Tribble' SCSI adapter is actually 16-bit microchannel - and any 32-bit microchannel adapter 'notched' for 16-bit has an application on a 16-bit microchannel system - somewhere. FWIW, the 65SX had a 'Tribble' as a stock adapter (which is why the ADF is there on a 55SX/65SX Reference Diskette).
 

mR_Slug

Veteran Member
Joined
Nov 28, 2006
Messages
941
Location
UK
I've been collecting MCA stuff on and off for 20 years and have NEVER seen MCA IDE cards.
 

Eudimorphodon

Veteran Member
Joined
May 9, 2011
Messages
5,116
Location
Upper Triassic
And if someone would actually trace the schematic of something like the AC-1070 or Sigma IDE it might help some motivated soul to do something about the shortage of drive options for these machines.
 

IBMMuseum

Veteran Member
Joined
Aug 28, 2006
Messages
1,449
There are actually a couple of projects that have advanced beyond the limitations of these adapters - typically that IDE is done in a non-standard way and has ROM table entries just for particular (small, and just as unavailable now) capacity IDE drives.
 

Eudimorphodon

Veteran Member
Joined
May 9, 2011
Messages
5,116
Location
Upper Triassic
There are actually a couple of projects that have advanced beyond the limitations of these adapters - typically that IDE is done in a non-standard way and has ROM table entries just for particular (small, and just as unavailable now) capacity IDE drives.

Link?
 

IBMMuseum

Veteran Member
Joined
Aug 28, 2006
Messages
1,449
Clarification: Never seen them in the wild. Have you ever actually seen one? I've got two of those 16bit IDE 8bit ISA silicone Vally (or is it Image) controllers. They used to turn up occasionally as well as the XTA ones. Never MCA (regular) IDE though.
Well - for the C1000, I scanned the "manual" with mine and imaged the ROM. I thought that I took photos of the adapter, but the picture at the link isn't mine and is missing the ROM. There was a diskette with it as well, but it may just be that Tomas hasn't updated the page yet or I am mistaken and need to look again.

"Manual" was in quotes because my copy is a copy of a copy - there probably was a level of production since the adapter looks and functions well, but I don't know the company history or how many were turned out. But yes, everything I linked, I have. Typically singles that I've had for a decade or two, which is why they won't get loaned out.
 

IBMMuseum

Veteran Member
Joined
Aug 28, 2006
Messages
1,449

I'm guessing this means about the development of the microchannel IDE adapters, not about the older implementations of limited drive tables - there isn't much online. There was someone with an ASIC design already prototyped on the PS/2 Facebook group, but I don't know how it is progressing. It has been some time since the last update (this is from almost two years ago).

McIDE_1.jpgMcIDE_2.jpgMcIDE_3.jpgMcIDE_4.jpg
 

Eudimorphodon

Veteran Member
Joined
May 9, 2011
Messages
5,116
Location
Upper Triassic
I'm guessing this means about the development of the microchannel IDE adapters, not about the older implementations of limited drive tables - there isn't much online.

Yes. And, well, I guess that's what I was getting at. If nobody is actually working on it publicly in a collaborative mode (Or, if they want to keep it proprietary so they can sell it just by themselves, but they never actually get around to the selling part) then it's not a ton of help to someone that needs a solution now.

I would wager a guess from the parts on that picture that they might be using the same CPLD code, or a close copy thereof, that the Snark Barker MCA card uses. And if their card is actually decoding the "standard" IDE primary/secondary/tertiary ports then it's possible that the software on that ROM is just a slightly hacked (or maybe not hacked at all) version of the XTIDE BIOS. But again, if they just disappeared after posting this we might never know.

Why it would be interesting to have the schematics of the older cards that just use plain TTL logic and PALs is it might generally make it more comprehensible how to interface something to MCA without the need of a CPLD or an unobtainium bridge chip, but I suppose one could just work backwards from what the Snark Barker's CPLD does.

(IDE ports in and of themselves are very simple, so the question is just how much glue it takes to get one talking on MCA after you've done the address decoding, and I guess I just don't know off the top of my head how hard that would be because it's been a long time since I did any reading on it. I vaguely remember that interrupts are slightly different and I don't know what you need to do with the arbitration lines if you want to just drive a PIO device like IDE in one of the "simpler" modes.)
 

IBMMuseum

Veteran Member
Joined
Aug 28, 2006
Messages
1,449
Yes. And, well, I guess that's what I was getting at. If nobody is actually working on it publicly in a collaborative mode (Or, if they want to keep it proprietary so they can sell it just by themselves, but they never actually get around to the selling part) then it's not a ton of help to someone that needs a solution now.

I would wager a guess from the parts on that picture that they might be using the same CPLD code, or a close copy thereof, that the Snark Barker MCA card uses. And if their card is actually decoding the "standard" IDE primary/secondary/tertiary ports then it's possible that the software on that ROM is just a slightly hacked (or maybe not hacked at all) version of the XTIDE BIOS. But again, if they just disappeared after posting this we might never know.

Why it would be interesting to have the schematics of the older cards that just use plain TTL logic and PALs is it might generally make it more comprehensible how to interface something to MCA without the need of a CPLD or an unobtainium bridge chip, but I suppose one could just work backwards from what the Snark Barker's CPLD does.

(IDE ports in and of themselves are very simple, so the question is just how much glue it takes to get one talking on MCA after you've done the address decoding, and I guess I just don't know off the top of my head how hard that would be because it's been a long time since I did any reading on it. I vaguely remember that interrupts are slightly different and I don't know what you need to do with the arbitration lines if you want to just drive a PIO device like IDE in one of the "simpler" modes.)
I was going to bring up Eric as someone that is more proactive in building microchannel adapters if it is not in any group collaboration. Existing IDE microchannel adapters are drive-table implementations (the adapters were sold with just a few pre-defined drives, no LBA or auto-detecting geometry support - and non-standard ports) - Later versions need to ditch those limitations.
 

Krille

Veteran Member
Joined
Aug 14, 2010
Messages
1,011
Location
Sweden
This guy on facebook (I don't recall his name but I wouldn't be surprised if he is also a member here) said he was using an Arco Computer Products AC-1075 controller with XUB and he claimed that he had not done any modifications to the code - he was just using the standard 16-bit IDE controller device type with interrupts disabled. So I added auto-detection of this controller to XTIDECFG in r609. As far as I know it has not been tested yet.

Anyway, if someone were to make a clone of an existing IDE MCA controller then this might be a good candidate.
 
Top