CalcHEP

bad precision of cross section with calchep !

Asked by chahrazed

Hi
I have a problem with some processes in Calchep which include the photon .
I get this error message ;

decay0: vv=-1.130318e-12 m1=0.000000e+00 m2=0.000000e+00
decay_0: negative incoming mass
mkmom infinite factor

and I obtain a bad precision in the value of crosse sections , is this problem comming from the Breit-Wigner propagator ??

Thanks.

Question information

Language:
English Edit question
Status:
Solved
For:
CalcHEP Edit question
Assignee:
No assignee Edit question
Solved by:
Alexander Pukhov
Solved:
Last query:
Last reply:

This question was reopened

Revision history for this message
Alexander Pukhov (pukhov) said : 		#1

Sorry, I was busy and could not reply you. Send me addition information about model, process and regularization you try
to implement. I hope to reply fast.

Best
  Alexander Pukhov

Revision history for this message
chahrazed (chahra-guella) said : 		#3

Thanks for your reply,

I work in SM model and I try this process ;

p,p -> e,E,E,ne

Regularizationn ;

34 MZ wZ 2
56 MW wW 2

Regards
Chahrazed

Revision history for this message
Alexander Pukhov (pukhov) said : 		#4

OK, you need regularization

34 MZ wZ 2
35 MZ wZ 2
46 mw

Revision history for this message
Best Alexander Pukhov (pukhov) said : 		#5

OK, you need regularization

34 MZ wZ 2
35 MZ wZ 2
46 MW wW 2
56 MW wW 2

Also you need a cut on
   M(e,E) | 10
Otherwise we'll have divergence for A->e,E

At such conditions one can reach a good precision.
But I met a problem with points with negative cross section.
It seems precision is lost. One can improve it asking CalcHEP to work with
'long double' numerical type. It a little bit slow down calculation, may be 20%, not more.
To force CalcHEP to work with long double one can open the file
include/nType.h and uncomment the first line
#define _LONG_.
After that we need recompile the package by

make clean
make

It result negative points still appear, but their contribution is about 10^{-7)%.

Best
    Alexander Pukhov

Revision history for this message
chahrazed (chahra-guella) said : 		#6

Thanks Alexander Pukhov, that solved my question.

Revision history for this message
chahrazed (chahra-guella) said : 		#7

Thank's so so much . My problem was solved.

Best regards