MadGraph5_aMC@NLO

Error in gg > hh jj [QCD] process

Asked by Harshal Kulkarni

While launching the process g g > h h j j [QCD] , I encountered the following error :

Command "launch auto " interrupted with error:
MadGraph5Error : A compilation Error occurs when trying to compile /mnt/c/Users/Harshal.Kulkarni/MG/MG5_aMC_v2_9_3/gg_hhjj/Source.
        The compilation fails with the following output message:
            rm -f ../lib/libdhelas.a
            cd DHELAS; make
            make[1]: Entering directory '/mnt/c/Users/Harshal.Kulkarni/MG/MG5_aMC_v2_9_3/gg_hhjj/Source/DHELAS'
            gfortran -O -fno-automatic -ffixed-line-length-132 -c -o aloha_functions.o aloha_functions.f
            gfortran -O -fno-automatic -ffixed-line-length-132 -c -o FFS1L1_2.o FFS1L1_2.f
            FFS1L1_2.f:27:26:

               27 | COEFF(3,1,1)= COUP*-CI * S3(5)
                  | 1
            Warning: Extension: Unary operator following arithmetic operator (use parentheses) at (1)
            FFS1L1_2.f:32:26:

               32 | COEFF(4,2,1)= COUP*-CI * S3(5)
                  | 1
              ...
              ...
              ...

            gfortran -O -fno-automatic -ffixed-line-length-132 -c -o VVVV4P0_1.o VVVV4P0_1.f
            ar cru ../../lib/libdhelas.a aloha_functions.o FFS1L1_2.o FFS1L2_1.o FFS1_0.o FFS1_1.o FFS1_2.o FFV1L1P0_3.o FFV1L1_2.o FFV1L2P0_3.o FFV1L2_1.o FFV1L3_1.o FFV1L3_2.o FFV1P0_3.o FFV1_0.o FFV1_1.o FFV1_2.o GHGHGL1_2.o GHGHGL2_1.o MP_FFS1L1_2.o MP_FFS1L2_1.o MP_FFS1_0.o MP_FFS1_1.o MP_FFS1_2.o MP_FFV1L1P0_3.o MP_FFV1L1_2.o MP_FFV1L2P0_3.o MP_FFV1L2_1.o MP_FFV1L3_1.o MP_FFV1L3_2.o MP_FFV1P0_3.o MP_FFV1_0.o MP_FFV1_1.o MP_FFV1_2.o MP_GHGHGL1_2.o MP_GHGHGL2_1.o MP_R2_GGHH_0.o MP_R2_GG_1_0.o MP_R2_GG_1_R2_GG_2_0.o MP_R2_GG_1_R2_GG_3_0.o MP_R2_GG_2_0.o MP_R2_GG_3_0.o MP_R2_QQ_1_0.o MP_R2_QQ_1_R2_QQ_2_0.o MP_R2_QQ_2_0.o MP_SSS1_1.o MP_VVS1_0.o MP_VVV1L2P0_1.o MP_VVV1P0_1.o MP_VVV1_0.o MP_VVVV1L2P0_1.o MP_VVVV1P0_1.o MP_VVVV3L2P0_1.o MP_VVVV3P0_1.o MP_VVVV4L2P0_1.o MP_VVVV4P0_1.o R2_GGHH_0.o R2_GG_1_0.o R2_GG_1_R2_GG_2_0.o R2_GG_1_R2_GG_3_0.o R2_GG_2_0.o R2_GG_3_0.o R2_QQ_1_0.o R2_QQ_1_R2_QQ_2_0.o R2_QQ_2_0.o SSS1_1.o VVS1_0.o VVV1L2P0_1.o VVV1P0_1.o VVV1_0.o VVVV1L2P0_1.o VVVV1P0_1.o VVVV3L2P0_1.o VVVV3P0_1.o VVVV4L2P0_1.o VVVV4P0_1.o
            ar: `u' modifier ignored since `D' is the default (see `U')
            ranlib ../../lib/libdhelas.a
            make[1]: Leaving directory '/mnt/c/Users/Harshal.Kulkarni/MG/MG5_aMC_v2_9_3/gg_hhjj/Source/DHELAS'
            gfortran -O -fno-automatic -ffixed-line-length-132 -c alfas_functions.f
            alfas_functions.f:78: Error: Can't open included file 'timing_variables.inc'
            make: *** [makefile:14: alfas_functions.o] Error 1

Can anyone please tell me what this error means and how I can fix it?

Thanks,
Harshal.

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
Last query:
Last reply:
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#1

First,

> g g > h h j j [QCD]

is a really weird syntax. Are you sure that what you try to compute make sense for QCD point of view.
Including quark in the final state but not in the initial state does not make that much sense.

> alfas_functions.f:78: Error: Can't open included file 'timing_variables.inc'

So looks like you are missing the file timing_variables.inc for your code to compile correctly.

But the main reason is why? This file is needed for NLO computation while on my laptop your syntax is schedulle for LO computation, did you play with the Template files or something like that?

I would suggest to fully re-install the code to be sure that you did not corrupt anything.

Cheers,

Olivier

> On 6 Jun 2021, at 15:00, Harshal Kulkarni <email address hidden> wrote:
>
> g g > h h j j [QCD]

Revision history for this message
Harshal Kulkarni (hck18ms056) said : 		#2

Thanks for your reply.

Originally, I was analyzing p p > h h j j, but I just wanted to see the effect of the ggh and gghh triangle and box diagrams, so I considered g g > h h j j. Actually, the process p p > h h j j at the tree level gave me a cross section which was much different from the theoretical/expected cross section, so I was wondering what the reason might be. It would be great if you have any ideas hinting towards the reason.

Thanks, I will re-install everything. I hadn't made any changes in the template files after installing.

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#3

Hi,

Did you check the Feynman Diagram? When you ask for
p p > h h j j you will have some kind of vector-boson fusion diagram
which will be of type:
q q~ > h h q q~
If you add the [QCD] you will just compute NLO correction for that contribution.

So looking at the loop-induced contribution in presence of two jet is not something that you can do easyly in mG5aMC.
One solution is likely to use a model where the W boson do not couple to light quark, or to define a loop filter to select the loop amplitude that you want to include in the computation.

Cheers,

Olivier

> On 6 Jun 2021, at 16:35, Harshal Kulkarni <email address hidden> wrote:
>
> Question #697424 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/697424
>
> Harshal Kulkarni posted a new comment:
> Thanks for your reply.
>
> Originally, I was analyzing p p > h h j j, but I just wanted to see the
> effect of the ggh and gghh triangle and box diagrams, so I considered g
> g > h h j j. Actually, the process p p > h h j j at the tree level gave
> me a cross section which was much different from the
> theoretical/expected cross section, so I was wondering what the reason
> might be. It would be great if you have any ideas hinting towards the
> reason.
>
> Thanks, I will re-install everything. I hadn't made any changes in the
> template files after installing.
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Harshal Kulkarni (hck18ms056) said : 		#4

Yes, I checked the Feynman diagrams, but I wasn't able to get the Feynman diagram corresponding to VBF which is mediated by W/Z bosons. I don't know what the issue is, but ideally there should be a VBF diagram in the p p > h h j j process I guess. Actually none of the Feynman diagrams that I have got have a W/Z boson as a mediator. Is this what is expected?

Thanks for the additional information.

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#5

Hi,

I have just done
Import model sm
generate p p > h h j j
display diagrams

And see a lot of diagram with W/Z mediator.
Of course if you do not use the SM then since the above syntax do not specify which term in the alphas/alphaew expansion you want to keep it is possible
That the code do not select automatically the constraint that you want. But in the SM I do not have any issue to have those VBF processes

Cheers,

Olivier

> On 6 Jun 2021, at 17:10, Harshal Kulkarni <email address hidden> wrote:
>
> Question #697424 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/697424
>
> Harshal Kulkarni posted a new comment:
> Yes, I checked the Feynman diagrams, but I wasn't able to get the
> Feynman diagram corresponding to VBF which is mediated by W/Z bosons. I
> don't know what the issue is, but ideally there should be a VBF diagram
> in the p p > h h j j process I guess. Actually none of the Feynman
> diagrams that I have got have a W/Z boson as a mediator. Is this what is
> expected?
>
> Thanks for the additional information.
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Harshal Kulkarni (hck18ms056) said : 		#6

Thank you so much. I think exactly that happened. Definitely, I used the SM, however I had modified the p and j definitions to include b and b~ quarks too. So, the p p > h h j j process didn't pick the W/Z mediator diagrams, how do I ensure that, even with these definitions of p and j changed, the process will consider those including the W/Z mediator and the additional ones involving b b~ too?

Thanks,
Harshal

Revision history for this message
Best Olivier Mattelaer (olivier-mattelaer) said : 		#7

The default is to keep the Feynman Diagram with the highest number of QCD interaction.
(as written in our paper and as describe as first exercise in all our tutorial)

So if you want to keep sub-leading term you have to ask for them like the following one:
generate p p > h h j j QED<=4

Note that you are using a massive b with a PDF which assume massless b.
This is of course a source of inconsistency (or theoretical error).
I typically discourage to do that but if you understand exactly what you are doing.
In this case, you might need to use a dedicated model where the b is massless but the yukawa mass not equal to zero.

You can define such type of model following the instruction from here:
https://answers.launchpad.net/mg5amcnlo/+faq/2312 <https://answers.launchpad.net/mg5amcnlo/+faq/2312>

Cheers,

Olivier

> On 6 Jun 2021, at 17:30, Harshal Kulkarni <email address hidden> wrote:
>
> Question #697424 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/697424
>
> Harshal Kulkarni posted a new comment:
> Thank you so much. I think exactly that happened. Definitely, I used the
> SM, however I had modified the p and j definitions to include b and b~
> quarks too. So, the p p > h h j j process didn't pick the W/Z mediator
> diagrams, how do I ensure that, even with these definitions of p and j
> changed, the process will consider those including the W/Z mediator and
> the additional ones involving b b~ too?
>
> Thanks,
> Harshal
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Harshal Kulkarni (hck18ms056) said : 		#8

Thanks Olivier Mattelaer, that solved my question.