I wonder why you'd solder ANYTHING in an environment like this one. This kind of project is PERFECT for wire-wrap. True, you didn't have the necessary parts, and, of course another day would have killed your desire to do to this, but if you'd waited a day or two and ordered the correct parts, starting, if you're point-to-point soldering, with a board that has pads to which to solder, and sockets for your active components, or with appropriate wire-wrap wire and sockets if you wire-wrap, you'd have had a lot more leeway in the debugging and repair process.

A part of building ANYTHING is planning. If you plan anything, you have to plan how to get from one end of the job to the other, and how to recover from a problem wherein the source of the fault isn't obvious is part of that planning. What did you plan to do if your MCU turned up "broken" one day?

There are several published 805x to IDE-channel interfaces on the www, and those are typically complete with code. You aren't building something "new." Would a day or two of delay waiting for the "parts that fit" have hurt anything? Why the hurry?

I'm also curious how you determined that the XRAM block isn't working with the XRAM. There's no external memory, right? Now, what, exactly, persuaded you that the XRAM wasn't working for you?

Now, I'm curious enough to plug this code into one of my circuits and try to determine whether it works there and, if not, why, or if so, why not in your application. I've got visiting grandkids right now, so it will be the weekend before I can give this a try, but I'll do that then.

The sad thing, of course, is that you've soldered wires to your MCU, hence, you'll have to unsolder each wire onto a socket if you're smart, or to another MCU if you're not, in order to determine whether there's a hardware problem, which I seriously doubt, and fix it.

RE