Opt=TS actually requires ModRedundant

Recently, I tried optimizing a transition state of rotation around a single bond – supposedly a simple task. However, Gaussian 16 gave me the following error:

 #N wB97XD/Def2TZVP Opt=TS SCRF(Solvent=Chloroform)
 --------------------------------------------------
 Opt=TS requires force constants or ModRedundant.

This can be solved (apparently) by changing

Opt=TS

to

Opt=(TS,ModRedundant)

and adding the desired dihedral definition in the additional input section after the geometry definition. Like this:

C 4.454656 2.343534 4.638545

5 6 7 8

Alas, I am still waiting for results to see if Gaussian did understand my directives :)