Shortest-proof-game challenge

I thought #6690 would be my final extension of the oscillating queen, but I just came up with this position. Is this a record for "most oscillations by a queen"?

Unique SPG in 31.0.

Why can't the bRa2 be on a1?

Because the rook could swap a move with the bishop going to a7. But here's the unique 32.5 move version of the position.

Adding those moves stops the Q oscillation, and thus detracts from the main theme.

"Longer" is not always "better".

Beauty is in the eye of the beholder. But if you insist on ending with queen moves, you could at least add 32. Qc2, getting a longer proof game and an extra half-cycle of oscillation. You could even add 32. Qc2 Ng2+ without detracting from the theme.

I don't insist on anything, as it's not my composition.

These are just my opinions on artistic merit.

I prefer the final position with Q back on d1. It would be fine to up it to 33.0 also, etc. and get more full oscillations.

Is this saying that the SPG for this position is cooked? What about the PG (not SPG!) in 32.5 without 33…Ng2+? I initially thought about posting this instead, but there might be Rg8-g1-a1 cooks to save a move for black.

Like @BigDoggProblem, I also like having the queen on her home square. Having complete oscillations seems nicer in my opinion.

The 33.0 PG is cooked. That Rg8-g1-a1 cook works.

It doesn’t work with the 32.5 PG (with Ng2 —> e3) though, so it might be unique. Maybe.

short unique spg?

SPGs in 13.5.

I was hoping the white queen couldn't go on the back rank

Unique SPG in 32.5. (Stelvio 2.1, Solving time: 5:11:18 hours)

Non-unique SPG in 12.0.

switchback, but also literally

I'd like someone other than @n9531l1 to post the solution to this one.