The $\phi^4$ theory is often given by $\mathcal{L}_{\text{int}} = - \frac{\lambda}{4!} \phi^4$ (see, e.g., Peskin–Schroeder). So it is natural to name the coupling constant lambda (not lambda^2).
It would be helpful if the reference manual explained why the following vertex definition results in an illegal coupling const error:
Model PHI4;
Particle phi,1;
Vertex phi,phi,phi,phi:lambda;
EndModel;
.end
FORM 5.0.0 (Jan 27 2026, v5.0.0) Run: Wed Feb 18 18:43:10 2026
Model PHI4;
Particle phi,1;
Vertex phi,phi,phi,phi:lambda;
EndModel;
*** illegal coupling const : nlegs - 2 + 2*loop: 2*loop = -1
*** Error : illegal value of c-constants
Program terminating at test.frm Line 4 -->
Terminate called from /home/tueda/work/form/sources/grcc.cc:10612 (erEnd)
Backtrace:
# 0: TerminateImpl at startup.c:1936:16
# 1: nextPart at grcc.cc:10618:1
# 2: _ZN5Model14addInteractionEP6IInput at grcc.cc:2324:25
# 3: LoadModel at diawrap.cc:161:2
# 4: CoEndModel at model.c:465:1
# 5: CompileStatement at compiler.c:697:5
# 6: PreProcessor at pre.c:1131:10
# 7: main at startup.c:1815:2
0.00 sec out of 0.04 sec
By the way, is eu-addr2line supposed to demangle Model::addInteraction without -C in the output above?
The$\phi^4$ theory is often given by $\mathcal{L}_{\text{int}} = - \frac{\lambda}{4!} \phi^4$ (see, e.g., Peskin–Schroeder). So it is natural to name the coupling constant
lambda(notlambda^2).It would be helpful if the reference manual explained why the following vertex definition results in an
illegal coupling consterror:By the way, is
eu-addr2linesupposed to demangleModel::addInteractionwithout-Cin the output above?