OK, it's now obvious to me that dynamics works differently in DS than it does in Poser.
I don't recall ever seeing a freshly loaded up cape standing out in the back like that in PS. We don't "pose" the cape, only the character that's going to be wearing it.
For instance, I would load up the character, and the cape, in T-Pose in frame 1, and then pose the character in the final frame, in this case about 30 to 35 frames would be necessary. Then I set up the cape to be simulated while I'm in the last frame, and when I run the simulation, the character, and the cape, go from T-Pose in frame 1 to the final pose in the last simulation frame, 1-step-at-a-time.