Insertions refinement for shuffle-greedy growth sequences

The tensegrity growth sequences in the previous post show that leaving the order of insertion edits to chance can lead to unnecessary instabilities.

For example, it is not necessary to pass through a stage as unstable as the 2-hole ladder (below) in growing a cube.

2-hole ladder: an unstable stage in previous growth sequence of a cube.

By attempting to alternate polarity in insertions, we get this alternate growth sequence for a cube, which still has 8 stages:

Growth code:/n/.us/.ns/...uss

Growth sequence:

np (3-strut)

nudp (2-prism)

nupd (tetrahedron)

nnpupd (eared tetrahedron)

nnuppd (3-prism)

nnuudppd (camp chair)

nnuupdpd (boat dual)

nnuuppdd (cube)

The same strategy applied to the dodecahedron gives a growth sequence of the same number of stages (29) that successfully alternates insertion polarity all the way through.

/u/1ns/3us/2nss/5uss/3nsss/7uss/5nsss/9uss/7nsss

ud
unpd
undp
undudp
undupd
unnpdupd
unndpupd
unnduppd
unnduudppd
unnduupdpd
unnduuppdd
unnnpduuppdd
unnndpuuppdd
unnndupuppdd
unnnduupppdd
unnnduuudpppdd
unnnduuupdppdd
unnnduuuppdpdd
unnndnpuuuppdpdd
unnndnupuuppdpdd
unnndnuupuppdpdd
unnndnuuupppdpdd
unnndnuuuudpppdpdd
unnndnuuuupdppdpdd
unnndnuuuuppdpdpdd
unnndnunpuuuppdpdpdd
unnndnunupuuppdpdpdd
unnndnunuupuppdpdpdd
unnndnunuuupppdpdpdd