Name: V. Shtabovenko Date: 08/08/19-01:55:49 PM Z
Hi,
FCFAConvert simply renames the momenta in the FeynArts output like
IncomingMomentum1 -> p1. It doesn’t perform any analytic continuation
from
ingoing to outgoing states or vice versa.
So either you generate the diagrams with everything outgoing from the
very
beginning, or you need to do the analytic continuation by hand,
including the
mappings between the external states if you want to work on the
amplitude level.
For amplitudes squared it is somewhat simpler, since you would only
need
to remap
the momenta and pay attention to some imaginary parts.
BTW, it might be simpler to do this with QGRAF, where the vertices and
wave functions
can be inserted at the very end of the amplitude manipulations. It
might
also make
sense performance-wise, since FeynArts is likely to become very slow
for
6 external
states already at 1-loop level. But then you are probably better off
with QGRAF+FORM
anyhow.
Cheers,
Vladyslav
Am 08.08.19 um 01:58 schrieb ythang:
> Hi again,
>
> what is the right way to compute an amplitude with the convention of
all outgoing momenta? Suppose I have a 4pt process (event 5pt or 6pt).
Naively, I would pass the following method to FCFAConvert
>
> IncomingMomenta -> {-p1, -p2}, OutgoingMomenta -> {p3, p4}
> IncomingMomenta -> {-p1, -p2,-p3}, OutgoingMomenta -> {p4,p5}
> IncomingMomenta -> {-p1, -p2,-p3}, OutgoingMomenta ->
{p4,p5,p6}
>
> for 4pt,5pt and 6pt respectively. Is that right?
>
> Thanks.
>