Don't `opt_rpitit_info` as a separate query ... another attempt to undo regressions r? `@ghost`
doc(hidden)