2R2 U2 2R2 U2 2R2 U2 But that alone doesn’t fix parity — it swaps front/back edges but leaves other issues. Let me give the tested, correct algorithm.
2R2 U2 2R2 Uw2 2R2 Uw2 Yes — this swaps the two opposite edge pairs on the last layer (UF↔UB and UR↔UL) without affecting corners. After this, your cube will have a solvable 3×3 PLL. Do a U or U' turn before and after the opposite swap algorithm to turn an adjacent swap into an opposite swap: 4x4 parity
