Name: Vladyslav Shtabovenko (email_not_shown)
Date: 02/11/15-06:33:39 PM Z

The conversion of FeynArts output to FeynCalc input is a recurring
question so let me answer it here in some details.

In general, FeynArts developers do not really care about any
compatibility to FeynCalc. This is why some effort is always needed
when doing a conversion between the two.

The first problem is that FeynArts and FeynCalc use same names
for their functions which inevitably leads to shadowing. This is
why FeynArts must be patched before it can be used with FeynCalc
on the same kernel.


Another problem is that many objects that you get in the output of
FeynArts are either called differently in FeynCalc or not present
at all (e.g. MajoranaSpinor). For some time we were
trying to integrate certain FeynArts objects into FeynCalc in order to
achieve some sort of seamless conversion. However, this didn't really
work out. In some cases FeynArts and FeynCalc would give such objects
completely different properties and that would lead to issues or even bugs.

Recently we changed our approach in the sense that we are not trying
to integrate FeynArts in FeynCalc or to change the way it behaves or
even to rewrite its models (that would be too much work anyhow).
Instead, we attempt to provide a filter that takes the output of
FeynArts as it is and converts it to an expression that can be evaluated
in FeynCalc as good as possible. This is done by the FCPrepareFAAmp function


For QCD and QED this already works quite well. The user just has to
rename the momenta as he or she likes, make the polarization vectors
transverse (if needed) and drop SumOver objects.

Have a look at the description in the wiki


or at the computations here


for some real life examples.


This archive was generated by hypermail 2b29 : 02/16/19-06:40:01 PM Z CET