Tripartite Palindromic Partition of Integer (HP 50g) Challenge

[3298]

That's indeed the correct way to interpret it - a number has a certain value, regardless of the base used to print it out. ZINTs may be stored in BCD and shown to the user in decimal as well, but as soon as you use integer division to chop it into digits of a given base, nobody cares about that anymore. The BCD encoding even gets lost in the conversion to BINTs, because inside the program the input number and under-construction palindromes are represented as a series of BINTs such that one BINT stands for one digit.
I considered using this list-of-digits format as user-facing format as well, but it's a pain to error-check as input, unlike an all-in-one number. The output side wouldn't have that problem, but it's ordinary numbers as well for consistency with the input.

I planned for arbitrary bases from the start, which I knew would be simpler than retrofitting them afterwards: it prevented me from randomly missing a corner case where a previously assumed base would stay in and mess up the results. Apart from that, it wasn't very hard ... just follow the proof's instructions, including the use of \(g\), the base. Mildly annoying at times, because I couldn't simplify 1GETLAM #1- (which stands for \(g-1\)) into just BINT9, I'll grant you that, but not hard.
Development included a lot of tests (with the lack of automatic error checks in SysRPL I pretty much had to if I wanted to avoid potential crashes). That means I would be surprised if something was wrong - and if it works for decimal, it should work for any base, since there are no special cases for that. Some of my tests also specifically checked other bases, but all those were good for was speedier exhaustive tests (with \(g=5\)) for the short-number algorithms, or for the longer ones of those, speedier exhaustive tests for certain ranges corresponding to a group of cases in the algorithms.

In other news, the big post about the internals of my program and its optimization is pretty much done, I'll post it shortly.