Jmol 11.4.RC2 (and also Revision 8973 retrieved from SVN a few minutes ago) doesn't display the annotated secondary structure of the asymmetric unit of PDB entry 3BKD correctly.
The asymmetric unit PDB file (attached together with the file of biological unit 1 in a ZIP file) contains several "HELIX" records. After loading it Jmol displays the whole protein as unstructured. If the PDB file of the biological unit 1 is loaded instead, the helices are displayed as annotated. Since the PDB just copies (some) of the header lines from the asymmetric unit files to the biological unit files the 'HELIX' records are identical in both files.
I havn't found other cases yet. There are at least cases where it works correctly like entry 1DEH.
The problem was introduced somewhere between 11.3.47 and 11.3.60 because the display is correct with 11.3.47.