ttbar_multi jets matching problem

Hi Huanian,

Is the 103 the cross-section before or after pythia?
Could you provide both number?

Cheers,

Olivier

On Mar 15, 2014, at 6:51 AM, Huanian Zhang <email address hidden> wrote:

> New question #245532 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/245532
>
> Hi
>
> I had some questions about the match problem. I generate the process below:
>
> generate p p > t t~, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
> add process p p > t t~ j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
> add process p p > t t~ j j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
> output ttbar_j_jj
>
> The corss section is about 73 pb using the default run_card.cat. However, the cross section is about 103 pb if I modify the run_card.cat according to the link you provided me several days agolike below:
>
> 20 = ptj ! minimum pt for the jets
> 5 = etaj ! max rap for the jets
> 0.001 = drjj ! min distance between jets
> 20 = xqcut ! minimum kt jet measure between partons
>
> But the cross section should be much smaller for the modified run_card.cat than the default one, right?
>
> Why I get a larger number? Thank you very much.
>
> Huanian
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Hi Olivier,

Yes, the cross section is the one after pythia. I was so confused when I got those two strange results.
Thank you very much for your help.

Sorry now I'm confuse.

if you set xqcut to 0, which cross-section do you have: 73pb?
if you set the xqcut to 20, what are the value that you get. after pythia 103 and what is the number before pythia?

Or should I consider your problem as solved?

Cheers,

Olivier

Sorry, I clicked "Solved" by accident. I still have questions.

Yes, the cross section is 73 for "xqcut = 0,"

And the value is 103 for both before and after pythia. I am also a little bit confused by the results that the cross section is the same for before and after pythia. Thank you very much.

Huanian

Did you set ickkw on 1?

Olivier

On Mar 16, 2014, at 5:11 AM, Huanian Zhang <email address hidden> wrote:

> Question #245532 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/245532
>
> Status: Answered => Open
>
> Huanian Zhang is still having a problem:
> Sorry, I clicked "Solved" by accident. I still have questions.
>
> Yes, the cross section is 73 for "xqcut = 0,"
>
> And the value is 103 for both before and after pythia. I am also a
> little bit confused by the results that the cross section is the same
> for before and after pythia. Thank you very much.
>
> Huanian
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Yes. I do set ICKKW to 1.

Could you run the following set of command (write them in a file and run it like ./bin/mg5 PATH)

install pythia-pgs
generate p p > t t~, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
add process p p > t t~ j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
output
launch -f
launch
pythia=ON
set ickkw 1
set xqcut 20
set etaj 5
set cut_decay F

Those set of command returns in my case:
RUN1 (no matching/merging) -> 31.74
RUN2 (before pythia) -> 43.23 ± 0.15
RUN2 (after pythia) -> 20.26 ± 0.23

Could you check that you reproduce those number. If yes then compare those generation with the one that you did.
To understand the difference.

Cheers,

Olivier

Thank you very much for your help, Olivier. I am out of town these days, but I will run it as soon when I am back on campus. And provide the feedback to you.

Hi Olivier,

I copy your commonds into a file, and run it, but it encounters some error, like below:

Command "import /data2/fantasyzhn/MadGraph5_v1_5_13/ttbar_j_jj.sh" interrupted in sub-command:
"set ickkw 1" with error:
InvalidCmd : Possible options for set are ['group_subprocesses', 'ignore_six_quark_processes', 'stdout_level', 'fortran_compiler', 'gauge', 'complex_mass_scheme']

But the cross section for run 1 is 41.41 pb.

Thank you very much.

Huanian

Ok the error is because I use a syntax which is only valid for 2.0.0 (and later).

The following should have work for 1.4.0 and later:

install pythia-pgs
generate p p > t t~, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
add process p p > t t~ j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
output
launch -f
launch
pythia
set ickkw 1
set xqcut 20
set etaj 5
set cut_decay F

If you still have problem, please tell me.

Cheers,

Olivier

The results I got are:

RUN1: 41.11 pb

RUN2 (before match and merging): 51.56 pb
RUN2 (after match and merging): 23.34 pb

This one looks reasonable.

Hi Olivier,

When I run the command (which is written in a file), it gives reasonable results:

install pythia-pgs
generate p p > t t~, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
add process p p > t t~ j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
add process p p > t t~ j j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
output
launch -f
launch
pythia
set ickkw 1
set xqcut 20
set etaj 5
set cut_decay F

Cross section before mathcing: 82.04
Cross section after matching: 24.31

But when I run the command seperately, like below:

generate p p > t t~, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
add process p p > t t~ j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
add process p p > t t~ j j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
output ttbar_j_jj
exit

then I go to the directory ttbar_j_jj, change the run_card.cat to the sets you provided, and run, it gives very wierd result: 103 pb.

Would you have a look at this case? Thank you very much.

Huanian

How did you launch it locally?

./bin/generate_events [if so what did you reply to the question?]
./bin/generate_events -f
something else?

Are you sure that you are asking for running pythia? (i.e. do you have both unmerged and merged cross-section on the html page?)
If not did you copy the default pythia card into the valid one? (cp Cards/pythia_card_default.dat Cards/pythia_card.dat) [This might be necessary depending of the options on which you run the executable]

I’ve run locally (only +1j) and I didn’t face any trouble at all.

Cheers,

Olivier

On Mar 22, 2014, at 7:26 PM, Huanian Zhang <email address hidden> wrote:

> Question #245532 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/245532
>
> Status: Solved => Open
>
> Huanian Zhang is still having a problem:
> Hi Olivier,
>
> When I run the command (which is written in a file), it gives reasonable
> results:
>
> install pythia-pgs
> generate p p > t t~, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
> add process p p > t t~ j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
> add process p p > t t~ j j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
> output
> launch -f
> launch
> pythia
> set ickkw 1
> set xqcut 20
> set etaj 5
> set cut_decay F
>
> Cross section before mathcing: 82.04
> Cross section after matching: 24.31
>
> But when I run the command seperately, like below:
>
> generate p p > t t~, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
> add process p p > t t~ j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
> add process p p > t t~ j j, (t > b w+, w+ > l+ vl), (t~ > b~ w-, w- > l- vl~)
> output ttbar_j_jj
> exit
>
> then I go to the directory ttbar_j_jj, change the run_card.cat to the
> sets you provided, and run, it gives very wierd result: 103 pb.
>
> Would you have a look at this case? Thank you very much.
>
> Huanian
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

I launch it by ./bin/generate_events.

Yes, it is running pythia. Only one cross section is returned.

And I use Cards/pythia_card.dat.

For the 1j case, launching locally also returns two cross sections, but the values are a little larger than the cross sections of the method you provided.

Or if I write all commands in a file, how do I replace the delphes_card.dat with a new one? And how do I run multi_run?

Thank you so much for all your help.

Hi huanian,

If only one cross-section is returned it means that you run_card doesn't have icckw=1 in it.
Since it means that the matching/merging is not apply correctly.

>For the 1j case, launching locally also returns two cross sections, but the values are a little larger than the cross sections of the >method you provided.

The two run method just differ by the way to use the interface, so the difference might either be
a) different cut/ parameter value
b) statistical fluctuation

>Or if I write all commands in a file, how do I replace the delphes_card.dat with a new one? And how do I run multi_run?

Both method are equivalent, I like using scripting, since this limit user intervention in files and limit the possibility of error.
But you are free to use any method that you like for editing the file/ run the code.

For Delphes, you can specify the path to the Delphes_card that you want:
so for the above example:
launch PROCESS_PATH
pythia
set ickkw 1
set xqcut 20
set etaj 5
set cut_decay F
DELPHES_PATH
PYTHIA_CARD

>And how do I run multi_run?

you need to use "launch -i" which is the equivalent of "./bin/madevent"
i.e.:
launch PROCESS_PATH -i
pythia
set ickkw 1
set xqcut 20
set etaj 5
set cut_decay F
DELPHES_PATH
PYTHIA_CARD

Cheers,

Olivier