I felt silly for not joining after my (incomplete) candidates for TI snow amputation would have ended up way more competitive than I expected. So here I am.

I figured out that 4-length blocks strike the best balance of extra speed from being a block and having time to exploit that speed. Except my spreadsheet was probably 90% off-by-one errors so who knows. Its result never lined up with the machine, but every version returned 2x 4-length blocks, so hopefully that is still optimal.

I'm sure it's possible to do this a bit better - the handoff to the stick bonder looks like a weakness to me. But I'm not planning on optimizing that unless I figure out why the spreadsheet estimate is wrong.

The title was more true of V1.