Reply 20 of 26, by snorg
To the OP: while I think your idea is interesting and it is something I've been toying around with myself, I don't think you can make a viable business around this if that is your thinking.
There just isn't a big enough market for it. If the target is a hobbyist engineer, then Frankensteining a homebrew x86 PC might be appealing for some sort of FPGA-based kit or motherboard based around the system-on-chip that Hightreason posted. An FPGA based system would allow people to try other architectures and not be limited to just a 386/486 system. I'm not sure that most hobbyists are going to be able to homebrew a motherboard capable of running a 386/486 class system, so then you're looking at a custom run of these things, and unless your're talking several hundred or thousands of parts it will be very expensive to make. I checked that link to the 486 motherboard but didn't see any prices, so that's not encouraging.
My guess is that to someone very interested in the underlying tech and wants to sort of design their own system is going to be more
interested in the FPGA route, or some other sort of kit.
If the goal is just to play games and not build/design a custom computer, then there are much cheaper ways to do it.