When the original PDB file contains ligand hetero atoms, the script from FirstGlance using JSmol 14.32.64c does not work in JSmol 14.31.8. I think this is because the older version does not load the ligand heteroatoms, while the newer version does. I suspect this causes the atomindices to be different so that the select commands don't work, but I haven't verified that.
When the original PDB file happens to have NO ligand hetero atoms, the FirstGlance script works in the old JSmol. I have experimented with such a case, 6bp8.
FIRST TRY
This scene was saved after dropping into the SAT JSmol a script from FirstGlance modified with this command:
load "" FILTER "*.CA;biomolecule 1;bmchains;/=3"
The scene renders perfectly in the SAT JSmol, rendering 20,358 alpha carbons. But when saved and played back from this green link
only the asymmetric unit (6,157 nonH atoms, 782 alpha carbons) loads. This is because the script saved by the SAT has removed the FILTER "..." from the original load command:
[:https://proteopedia.org/wiki/scripts/91/918475/6bp8/1.spt].