The inspiration for this solve comes from two sources: our hexarm-driven P3 cycles solve iterations, and Fiesta's dual-pipeline Sum4 solution to the 2024 pre-tournament puzzle, Simulacrum.  The gist of things is that hexarms are nice for middling-to-fast Rate, because they don't need to reset and can operate faster than P4; and Simulacrum showed us that parallel processing can be a very effective Sum strategy, even if it makes for Cost-heavy, Cycles-light solves.

In fact, there are only two reasons these pipelines are not perfectly symmetrical.  First, *someone* needs to output, and neither constructor arm can do that from where it naturally sits.  Second, the input hexarm cannot easily spin backwards as the copper one does, which necessitates the sad, barely-used simple arm on the top left.

We went with 1P or simplicity.  We briefly considered 1/4P, pulling 2 inputs per loop but causing one grab to miss every fourth loop, but this first idea looked so clean that we decided not to bother. It's just a weekly, after all.