[AI-assisted reply: drafted with Claude and checked by me]
This is a common issue with clinical MRI, and the explanation in this thread applies to knees too:
In short, each series is high-resolution only in the plane it was acquired in, typically 0.3–0.5 mm pixels with 3–4 mm slices. The other two views are reconstructed across those thick slices, so they look blurry. They also seem to have “more slices” because the slider steps through the fine in-plane spacing, not through new data. To see each plane sharply, show the series acquired in that plane in the matching view. Combining the separate series into one high-resolution volume is a hard problem with no standard solution. See Combining volumes - what am I missing? - #2 by lassoan.