Hi,
thanks for the bug report. ChangeDimension received some modification
during the addition of Cartesian tensors, but the checks indeed seem to
be to restrictive. It should only care about the special case (_Symbol -
4), while the rest should work as before. I pushed a fix to both
repositories that should recover the old behavior.
In general, for various reasons I would try to avoid having an explicit
Epsilon in the dimensions of Dirac matrices and tensors, but anyhow I
think that you know what you are doing.
Regarding PropagatorDenominator, its first argument is actually supposed
to be a linear combination of Momentum symbols. So the fact that
FeynAmpDenominator[PropagatorDenominator[a, b]] // FCI
used to work in the previous versions is rather undefined behavior than
something intentional, especially since PropagatorDenominator is an FCI
symbol that may appear
only inside FeynAmpDenominator.
It also was necessary to clean up the definitions of
PropagatorDenominator to make sure that new things like
DataType[z, FCVariable] = True;
FAD[l, {l + z p, m}] // FCI // StandardForm
(*FeynAmpDenominator[PropagatorDenominator[Momentum[l, D], 0],
PropagatorDenominator[Momentum[l, D] + z Momentum[p, D], m]]*)
can work properly.
Cheers,
Vladyslav
Am 11.02.20 um 17:12 schrieb D. Azevedo:
> Hello,
>
> I am having problems running a notebook I have written in FC v9.2 (Mathematica 10) within a PC with FC v9.3 (Mathematica 12).
>
> I have generated tadpole diagrams using FeynArts and I want to have the amplitude expression in FeynCalc form, for that I use:
>
> amps = FCFAConvert[CreateFeynAmp[DiagTadpolesBigH],
> IncomingMomenta -> {p}, LoopMomenta -> {q},
> UndoChiralSplittings -> True, SMP -> True, DropSumOver -> True,
> ChangeDimension -> 4 - \[Epsilon]] /.
> PropagatorDenominator[x_, y_] :>
> FeynAmpDenominator[PropagatorDenominator[x, y]] /. paramsubs //
> Contract
>
> But it doesn't accept the dimension change, it says
>
> ChangeDimension::failmsg: Error! ChangeDimension has encountered a fatal problem and must abort the computation. The problem reads: Unsupported choice of dimension!
>
> If I choose just \[Epsilon] then it doesn't give me any error but it seems the dimention becomes undefined. Also I need to change
>
> PropagatorDenominator[x_, y_] :>
> FeynAmpDenominator[PropagatorDenominator[x, y]]
>
> to
>
> PropagatorDenominator[x_, y_] :>
> FeynAmpDenominator[PropagatorDenominator[Momentum[x,4-\[Epsilon]], y]]
>
> for the propagator to behave "normally".
>
> Is there any change in this regard with the change of version of FC?
>
> Cheers,
> Duarte
>
>
This archive was generated by hypermail 2b29 : 09/04/20-12:55:05 AM Z CEST