Mine does work for sys3-4 as well, just covers up one of the diagnostic buttons a bit. Can still reach it, just not as convenient.

Only waiting to get licensing sorted out again... I have 100 PCB's, including 40 already built. Can sell them any time with blank EPROMs if you guys are able to sort out doing the code combinations.

Sorry for the short and choppy posts, on the road for MGC right now.

One thing to keep in mind with any version of this solution, it's not a 100% solution. Similar to the different versions of the 2732 modifications that have been floating around for a few years. Something odd went on with Williams interchanging parts with different specs or from different suppliers, and it affects the addressing of the ROM. There are times this mod won't work with some CPU boards. (Which is why there are two versions of the 2732 mod floating around.).

I haven't been able to nail down what the source of this is, wish I knew, but I haven't found it yet.

Very possible the LS's could be part of the issue. Particularly if they suddenly have open circuits on the outputs where There used to be rom chips.

I just haven't revisited the issue in quite a while. Since I don't do repair work anymore, I don't have a variety of boards coming through anymore to test out theories.

On system 6, something else gets gated along with the VMA, I just don't remember off the top of my head. Still not home yet. Almost exactly the same as my schematic other than that one additional line, and I feed it into a 74ls20. (Or is it an 02? the quad input gate).

The jumpering shouldn't matter, as it only affects addressing for the ROM chips, which would be removed already anyway. I don't recall it altering the addressing for any other ic's.

Yep, those are the two methods. I just never found any explanation why the Zig only works sometimes, while the Duncan is so reliable. What changed in the CPU boards that causes the difference?

They don't seem to mind at all, to be honest.