The AGARD 445.6 configuration is the most popular validation test case for transonic flutter predictions, but the actual extent of truly nonlinear transonic flow for this case is unclear, due to the sparsity of the experimental data and the thin profile of the wing. This work utilizes a combination of mesh adaptation and the linearized frequency-domain method to obtain high-quality viscous and inviscid flutter predictions; these solutions show a double flutter dip through the transonic Mach range driven by complex shock growth across the wing. A single experimental flutter point lies in this flutter dip area, which is not enough to assess the accuracy of these transonic flutter predictions. Modeling the boundary layer of the wind tunnel wall (as opposed to the commonly assumed symmetry wall assumption) has a very large impact on the predicted flutter boundary, though the size of the boundary layer during the original experiment is uncertain.