Can't decay top quark in mu mu > vv tt~ z{0} using Madspin

Asked by Jacob Chen

I was trying to generate mu+ mu- > vm vm~ t t~ z{0} in 30 TeV in the sm-no_widths mode, and decay finals states with Madspin as following :

t > w+ b, w+ > l+ vl
t~ > w- b~, w- > l- vl~
z > l+ l-

The results I get is the following:

"Error detected in "generate_events run_02"
write debug file /Users/****/Documents/Software-Codes/MG5_aMC_v3_2_0/mumu2vvttz0/run_02_tag_1_debug.log
If you need help with this issue please contact us on
str : A compilation Error occurs when trying to compile /Users/*****/Documents/Software-Codes/MG5_aMC_v3_2_0/mumu2vvttz0/full_me/SubProcesses/P1_mummup_vmvmxttxz0_t_bwp_wp_epve_tx_bxwm_wm_emvex_z_epem.
 The compilation fails with the following output message:
     gfortran -w -fPIC -ffixed-line-length-132 -c -o driver.o driver.f

         9 | PRWIDTH(-2)=ZERO
           | 1
     Error: Symbol 'zero' at (1) has no IMPLICIT type
     make: *** [driver.o] Error 1

 Please try to fix this compilations issue and retry.
 Help might be found at
 If you think that this is a bug, you can report this at "

I couldn't make sense of it. Would appreciate some help! The version I'm using is MG5_aMC_v3_2_0.

PS1, you might wonder why I'm using no-widths mode. That's because longitudinal polarization for w,z in the final states can cause unphysical energy increase due to unclean gauge cancellation. One solution is set all width to be zero, and let Madspin decay the final states. But this strategy doesn't seem to work here.

Question information

English Edit question
MadGraph5_aMC@NLO Edit question
No assignee Edit question
Solved by:
Jacob Chen
Last query:
Last reply:

This question was reopened

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said :


This is conceptually problematic since MadSpin includes offshell effect and therefore having the width set to zero is problematic.
Independently of that MadSpin use a re-weighting approach and if you set the width to zero the matrix-element will be not bounded and MadSpin will take an infinite amount of time to run.

Actually I do not have a real solution for you with keeping the width to zero.
Probably the OSP (onshell projection) might be a solution for you --knowing that gauge invariance is not really preserve due to the arbitrary reshuffling--. This is only alpha (and we do not have any plan to release it) but you can find it here:
However this is only implemented for W decay (since the W mass is hardcoded inside the code to do the special treatment)
You can try to edit the W-mass to the Z mass to have something working.

Note that you might want to merge this first with the latest version to include all the bug fixed that have been released since the development of that test code.



Revision history for this message
Jacob Chen (rainbowend) said (last edit ):

Hi Olivier,

Thanks the answer. I think I found a way to bypass the problem by asking other friends. I didn't import sm-no_widths, instead I just import sm model,and set the width to 0 in the parameter card after launching. It worked.

Still I appreciate the help. Maybe will try your solution some time.



Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said :

I'm pretty sure that MadSpin do not use the 0 width that you request in that case.
Did you check the invariant mass of the Z?



Revision history for this message
Jacob Chen (rainbowend) said (last edit ):

Setting the width to 0 in the parameter card doesn't affect the Madspin operation, since Madspin will import the model again using the default setting of sm model file. I think that's the trick that makes it work. (See below)


J. C.

* *
* W E L C O M E to M A D S P I N *
* *
INFO: Extracting the banner ...
INFO: process: mu+ mu- > vm vm~ w+ w- z
INFO: options:
INFO: detected model: sm. Loading...
lhapdf-config does not seem to correspond to a valid lhapdf-config executable.
Please set the 'lhapdf' variable to the (absolute) /PATH/TO/lhapdf-config (including lhapdf-config).
Note that you can still compile and run aMC@NLO with the built-in PDFs
 MG5_aMC> set lhapdf /PATH/TO/lhapdf-config

Set group_subprocesses to Auto
Note that you need to regenerate all processes
set max_weight_ps_point 400 # number of PS to estimate the maximum for each event
decay w+ > l+ vl
decay w- > j j
decay z > b b~
INFO: Will use seed 897088290
INFO: We need to recalculate the branching fractions for z,w-,w+
INFO: using the FeynRules formula present in the model (arXiv:1402.1178)
INFO: Restrict model /Users/****/Documents/Software-Codes/MG5_aMC_v3_2_0/models/sm with file models/sm/restrict_default.dat .
INFO: Run "set stdout_level DEBUG" before import for more information.


Revision history for this message
Jacob Chen (rainbowend) said :