Using hexarms to get more done at once is the usual idea, but this one was a pain to find. I kept thinking that using more bonders would be a bad thing.

Before making this solve, most of my solutions had length 1 arms for the main work since I needed to get the qs to either side of the bonder in one cycle. It was tinkering with this that made me realize that you don't need length 1 arms for that, which let me optimize a few lower G solves (90, 130) too.