I've been thinking about the best arrangement for bitfields plus flags. Unfortunately, the answer has not come to me. It took me several weeks to think of the best arrangement for a bitfield chart. Other matters may take considerably longer. For example, it took me two years to think of a feasible method to structure a hydroponics diagnostic system - and I have yet to implement it. The trick here is to structure a diagnostic tree
expert system so that it promotes homoeostasis.
I've been flummoxed on simpler tasks. For example, I struggled with a family tree Motorola derived processors focused on 6502 derivatives. However, if you look at Wikipedia's family trees of programming languages or web browsers, you'll find a common style which also arranges them in a proportionate timeline. That's just showing off.
With the luxury of interactive scripting within a web page, it might be desirable to have some similar to a Digikey or Mouser component picker. Specifically, it would be useful to select the union or intersection of opcodes available on a subset of different 6502 variants and then match by addressing mode, partial bitfield, flag (in)sensitivity (such as decimal or carry in) or flag clobbering while allowing successful matches to be sorted alphabetically, by addressing mode, by cycle count or other criteria. It may also be useful to show partial matches.
For static layout, I considered adding a row within each bitfield cluster to denote affected flags. This may require a more fine-grained arrangement of bitfields. However, a significantly better technique may become obvious within the next two years.
Login
Register
FAQ
Search
Let me second the recommendation for the Eyes & Lichty programming manual. Its size might be scary; but it starts out small, and builds, so you don't have to take the whole thing in one gulp, but rather stop and digest new material before moving on. Then there are large reference sections which I still refer back to frequently now even after 25+ years of having it.