Possibility of making clone MCA sound cards.

[TrekkiesUnite118]

So I've recently gotten my IBM PS/2 Model 50 working again and for the most part it's doing the job I intended for it quite nicely. That is to be an 80s and early 90's era DOS gaming rig. Most of what I've thrown at it runs fine such as Kings Quest, Silpheed, Monkey Island, Wolfenstein 3D, etc. But the one glaring flaw in this is the lack of a sound card, and as we all know MCA sound cards are rare and expensive. There are options like the OPL2LPT, but for my purposes at least I'd need to have individual games patched as a 286 can't use the adlipt application.

As I've been diving into this rabbit hole I noticed there's been quite a few clones made for Adlib, Soundblaster, and Roland MT-32 interface cards on the ISA side of things and it got me wondering, how hard would it be to make clones of existing MCA sound cards? As I've gathered looking at threads like this one about making an IDE MCA card it seems one of the big issues with new cards is not knowing enough about the MCA bus to properly make a new card.

However for something like this I'd imagine you could use the few existing MCA sound cards as a blueprint to reverse engineer and possibly make clones. I'm curious if there is some technical issue in the way of something like this or if it's just a simple lack of interest. I admit I have very little knowledge or experience in this kind of thing, so it's very likely I'm oversimplifying things. So I figured it might not be a bad idea to make a thread to discuss the idea and if there was interest see if there was any way to get the idea off the ground.

[Chuckster_in_Jax]

I'm curious if there is some technical issue in the way of something like this or if it's just a simple lack of interest.

Not a lack of interest. I would buy several. But that would be a lot of work for someone.

[Goodtofufriday]

Not a lack of interest. I would buy several. But that would be a lot of work for someone.

Im looking to create an adlib mca clone one. But without a reverse side image of the board its not feasible.

[TrekkiesUnite118]

But that would be a lot of work for someone.

True, though it might be something that could be a bit more manageable as a group effort. For example, those who have existing MCA sound cards could provide high resolution pictures of both sides of the boards, the back panel, etc. Then people who are knowledgeable could use those to start identifying components, figuring out the traces, and start to paint a clearer picture of what these cards do. When that's figured out it would then become an issue of sourcing components and coming up with a new PCB, which again people who have expertise in that area could step in to help there.

[kgober]

It would have been extremely uncommon for someone to use a Micro-Channel PS/2 as a gaming rig. These systems were mostly used by businesses or other institutions. This is why it's so hard to find MCA sound cards -- not very many people bought them because business applications didn't need sound at that time.

Making MCA cards now is probably technically possible, but difficult because there is little public documentation. IBM learned their lesson with the ISA bus and in order to avoid losing control of the platform again they set things up so that you would need to license IBM patents in order to make a Micro-Channel expansion card. The documentation you would need was only obtainable under NDA from IBM.

If you really want a DOS-era gaming rig you are better off with an ISA-bus PS/2 (like the Model 30 286) or something like a Compaq Deskpro.

[TrekkiesUnite118]

It would have been extremely uncommon for someone to use a Micro-Channel PS/2 as a gaming rig. These systems were mostly used by businesses or other institutions. This is why it's so hard to find MCA sound cards -- not very many people bought them because business applications didn't need sound at that time.

True, though they are a rather common machine to find. And if clone cards could be made, suddenly a lot of them would become viable gaming rigs.

Making MCA cards now is probably technically possible, but difficult because there is little public documentation. IBM learned their lesson with the ISA bus and in order to avoid losing control of the platform again they set things up so that you would need to license IBM patents in order to make a Micro-Channel expansion card. The documentation you would need was only obtainable under NDA from IBM.

I get that, which is why I was asking about reverse engineering existing cards. I'd imagine by doing that you might be able to figure out what you need to make a straight clone by examining the existing cards.

[Unknown_K]

There are so many machines of the era around that can be used for gaming I don't see why anyone would bother. There are sound cards (non gaming) for PS/2 machine if you like playing with audio and there are video editing cards (sound in and out). The common video cards (or built in video) are not great for gaming either.

[TrekkiesUnite118]

There are so many machines of the era around that can be used for gaming I don't see why anyone would bother. There are sound cards (non gaming) for PS/2 machine if you like playing with audio and there are video editing cards (sound in and out). The common video cards (or built in video) are not great for gaming either.

I don't think the video card would really be much of an issue. The built in VGA out on my Model 50 seems to be more than enough for the 2D DOS games I'm throwing at it. The few 3D ones of the era like Wolfenstein are CPU brute forced anyways. For the purpose of being a 80s to early 90s DOS gaming rig I think the video hardware available should be decent enough. If you hit the point where you need more, at that point you'd probably start looking at a later windows based machine anyways. Really the big thing that seems to hold these things back is the lack of easily available sound cards.

Yeah, you could get a generic ISA based clone system, but that doesn't really solve an ongoing problem. When you see what MCA sound cards go for I'd imagine there is some demand to justify at least exploring the idea.

[PeterNC]

There are many obstacles. ISA cards are easily designed / supported / produced because there were and are so many. Replicas are therefore straightforward for 8BIT cards (not a lot of components and readily available at that: relatively few traces: surface only). IBM PC bus (8BIT ISA) is extremely well documented. A MCA card would probably be hard to design, and ADF would need to be programmed / cloned and proprietary CT / Roland chips (CT5320 / CT5330 / MPU-IMC) would require serious retro identification / engineering / hacking.

We looked into a MCA and PCI MIF-IPC-A clone a few years ago but gave up on the idea: too complicated / time consuming / cost prohibitive. Check out www.vogons.org: AdLib, MPU-401 (HardMPU, Music Quest PC MIDI Card), GUS and other clones have been developed / produced there. Have yet to see an aftermarket MCA PCB though. With a MPU-IMC clone (with lots of components / traces) this is only half the equation: a MPU-IMC BOB is another requirement. A CT5320 has far fewer components / traces but good luck locating the ICs.



https://en.m.wikipedia.org/wiki/Micr...l_architecture

VCFED history: http://www.vcfed.org/forum/archive/i...p/t-40572.html.

[glitch]

Part of what makes MCA sound cards so expensive is just that they're rare. Who knows if the people buying them are really using them.

If anyone were to pursue a hobbyist-driven MCA interface, I think, as I've discussed elsewhere, that probably a MCA -> ISA bridge would be the smart way to go, with a MCA board that carried one or more PC/104 modules. That way, the expensive part only has to be made once, and any number of off-the-shelf interfaces could be plugged in, including SoundBlaster-compatible sound.