I am glad to see that they finally moved the project to PCB, but, man, is breadboard ever a terrible platform for microprocessor development.
Doesn't anybody do wirewrap anymore? I, too, struggled with bad connections and crosstalk in my breadboard projects in my first and second year at university, and then I discovered wirewrap, and never looked back.
Apart from having to think in reverse, wirewrap is almost as easy as breadboarding (arguably easier when things get crowded), and it is robust enough to be used for production projects.
There's a book called 'The 8088 Project Book' that takes you step-by-step through building an 8088 breadboard computer starting focusing on a 'get the minimum working, then add to it' approach (I guess we call that MVP now). I enjoyed building/learning that one; it's maybe one of the few Tab books that wasn't rubbish.
There's a similar breadboard computer video that builds on an mc68008.
This was a well done and engaging series, kudos. I remember creating projects like this - still amazed that you can step a CPU like that one cycle at a time.
Rochester Electronics (REI) does. They're in the product lifespan extension space and are licensed by the big guys (Intel, AMD, etc) to produce popular products after the OEM has moved on to newer stuff, usually to satisfy contract obligations (e.g. "Must provide 20 years of spare parts", often for military apps). They aren't cheap.
Innovasic used to, but I dunno what they're up to since Analog Devices picked them up. Similar biz plan to REI.
There's probably an embedded ASIC being sold someplace that incorporates an 8086/8088 somewhere deep down. You probably won't find '8086' mentioned in the datasheet, nor will you likely have heard of the company.
These days there are numerous synthesizeable models you can cram into an FPGA. And there's still bins full of NOS chips around.
GH: https://github.com/Mati365/ts-c-compiler