Phase space problem

[Tiago Vale]

Greetings,

When i try to generate this process listed above, the one with p p > tp tp~ > z t z t~ > ... is said to not have enough phase space available and tells me to check mass spectrum. But when I generate that process alone, no such problem arises. And at the event generation, the following problem is shown:
 IOError: [Errno 2] No such file or directory: '/home/tdvale/VLQ/LH_RH/MG5xProtos/TT/TT_1TeV/SubProcesses/P0_gg_ztztx_z_ll_t_wpb_wp_lvl_z_ll_tx_wmbx_wm_lvl/G1/results.dat'

This error does not happen when the process is generated alone.

The whole process list is the following:

generate p p > tp tp~ > h t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~,\
w- > l- vl~)
add process p p > tp tp~ > h t z t~, (t > w+ b, w+ > l+ vl),( t~ > w- \
b~, w- > l- vl~), z > l+ l-
add process p p > tp tp~ > h t w- b~, (t > w+ b, w+ > l+ vl), w- > l- \
vl~
add process p p > tp tp~ > z t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
b~, w- > l- vl~), z > l+ l-
add process p p > tp tp~ > z t z t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
b~, w- > l- vl~), z > l+ l-
add process p p > tp tp~ > z t w- b~, (t > w+ b, w+ > l+ vl), z > l+ l\
-, w- > l- vl~
add process p p > tp tp~ > w+ b h t~, (t~ > w- b~, w- > l- vl~), w+ >\
l+ vl
add process p p > tp tp~ > w+ b z t~, (t~ > w- b~, w- > l- vl~), w+ >\
l+ vl, z > l+ l-
add process p p > tp tp~ > w+ b w- b~, w+ > l+ vl, w- > l- vl~

The UFO Model I am using is the one in this link:
http://feynrules.irmp.ucl.ac.be/wiki/VLQ

Thank you.

[Olivier Mattelaer (olivier-mattelaer)]

Dear Tiago,

I have try with the latest version and did not face any problem. Did you run with the latest version as well?

Cheers,

Olivier

[Tiago Vale (tiagval)]

Hi Olivier,

Yes, I tried with the latest and previous one and it still happens...
Well, in the previous one it says something about not being able to
reduce color indices and does not generate as well.

Cheers,

Tiago

On 09/15/2015 10:35 PM, Olivier Mattelaer wrote:
> Dear Tiago,
>
> I have try with the latest version and did not face any problem. Did you
> run with the latest version as well?
>
> Cheers,
>
> Olivier
>

[Olivier Mattelaer (olivier-mattelaer)]

Dear Tiago,

So clearly, I'm not able to reproduce your problem.
Since you did not mention anything about the benchmark point, I guess that you are using the one associate to the model.
Is this correct? (same for the cut I guess)

Cheers,

Olivier

PS: It can help if you upload to launchpad th debug file created automatically.

[Tiago Vale (tiagval)]

Yes, that is correct.

On 09/16/2015 04:40 PM, Olivier Mattelaer wrote:
> Dear Tiago,
>
> So clearly, I'm not able to reproduce your problem.
> Since you did not mention anything about the benchmark point, I guess that you are using the one associate to the model.
> Is this correct? (same for the cut I guess)
>
> Cheers,
>
> Olivier
>
>
> PS: It can help if you upload to launchpad th debug file created automatically.
>

[Olivier Mattelaer (olivier-mattelaer)]

Hi Tiago,

I succeed to reproduce the problem.
The reason of the "no phase-space" is due that you run in multi-core.
Some core have already started the computation when some other are still compiling the code.
(So far this is normal)
the problem is that the node started the computation crashed since the color associated to one of the events can not be generated.
This create an error file which is wrongly interpret by the node of the compilation like a no phase-space.

So looks like they are still one case where the epsilon_ijk structure is not handle in full generality.
I will look at that.

Cheers,

Olivier

[Tiago Vale (tiagval)]

Hello Olivier,

Thank you very much for your answer and your help. I appreciate it.

Good work.
Cheers,
Tiago

On Sat, Sep 26, 2015, 12:35 AM Olivier Mattelaer <
<email address hidden>> wrote:

> Hi Tiago,
>
> I succeed to reproduce the problem.
> The reason of the "no phase-space" is due that you run in multi-core.
> Some core have already started the computation when some other are still
> compiling the code.
> (So far this is normal)
> the problem is that the node started the computation crashed since the
> color associated to one of the events can not be generated.
> This create an error file which is wrongly interpret by the node of the
> compilation like a no phase-space.
>
> So looks like they are still one case where the epsilon_ijk structure is
> not handle in full generality.
> I will look at that.
>
> Cheers,
>
> Olivier
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1493832
>
> Title:
> Phase space problem
>
> Status in MadGraph5_aMC@NLO:
> New
>
> Bug description:
> Greetings,
>
> When i try to generate this process listed above, the one with p p > tp
> tp~ > z t z t~ > ... is said to not have enough phase space available and
> tells me to check mass spectrum. But when I generate that process alone, no
> such problem arises. And at the event generation, the following problem is
> shown:
> IOError: [Errno 2] No such file or directory:
> '/home/tdvale/VLQ/LH_RH/MG5xProtos/TT/TT_1TeV/SubProcesses/P0_gg_ztztx_z_ll_t_wpb_wp_lvl_z_ll_tx_wmbx_wm_lvl/G1/results.dat'
>
> This error does not happen when the process is generated alone.
>
> The whole process list is the following:
>
> generate p p > tp tp~ > h t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~,\
> w- > l- vl~)
> add process p p > tp tp~ > h t z t~, (t > w+ b, w+ > l+ vl),( t~ > w- \
> b~, w- > l- vl~), z > l+ l-
> add process p p > tp tp~ > h t w- b~, (t > w+ b, w+ > l+ vl), w- > l- \
> vl~
> add process p p > tp tp~ > z t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
> b~, w- > l- vl~), z > l+ l-
> add process p p > tp tp~ > z t z t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
> b~, w- > l- vl~), z > l+ l-
> add process p p > tp tp~ > z t w- b~, (t > w+ b, w+ > l+ vl), z > l+ l\
> -, w- > l- vl~
> add process p p > tp tp~ > w+ b h t~, (t~ > w- b~, w- > l- vl~), w+ >\
> l+ vl
> add process p p > tp tp~ > w+ b z t~, (t~ > w- b~, w- > l- vl~), w+ >\
> l+ vl, z > l+ l-
> add process p p > tp tp~ > w+ b w- b~, w+ > l+ vl, w- > l- vl~
>
> The UFO Model I am using is the one in this link:
> http://feynrules.irmp.ucl.ac.be/wiki/VLQ
>
> Thank you.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mg5amcnlo/+bug/1493832/+subscriptions
>

[Olivier Mattelaer (olivier-mattelaer)]

Hi Tiago,

A workaround for this bug is to generate with the following syntax:

generate p p > tp tp~ > h t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~,\
w- > l- vl~) @1
add process p p > tp tp~ > h t z t~, (t > w+ b, w+ > l+ vl),( t~ > w- \
b~, w- > l- vl~), z > l+ l- @2
add process p p > tp tp~ > h t w- b~, (t > w+ b, w+ > l+ vl), w- > l- \
vl~ @3
add process p p > tp tp~ > z t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
b~, w- > l- vl~), z > l+ l- @4
add process p p > tp tp~ > z t z t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
b~, w- > l- vl~), z > l+ l- @5
add process p p > tp tp~ > z t w- b~, (t > w+ b, w+ > l+ vl), z > l+ l\
-, w- > l- vl~ @6
add process p p > tp tp~ > w+ b h t~, (t~ > w- b~, w- > l- vl~), w+ > \
l+ vl @7
add process p p > tp tp~ > w+ b z t~, (t~ > w- b~, w- > l- vl~), w+ > \
l+ vl, z > l+ l- @8
add process p p > tp tp~ > w+ b w- b~, w+ > l+ vl, w- > l- vl~ @9

Cheers,

Olivier

On 26 Sep 2015, at 00:37, Tiago Vale <email address hidden> wrote:

> Hello Olivier,
>
> Thank you very much for your answer and your help. I appreciate it.
>
> Good work.
> Cheers,
> Tiago
>
> On Sat, Sep 26, 2015, 12:35 AM Olivier Mattelaer <
> <email address hidden>> wrote:
>
>> Hi Tiago,
>>
>> I succeed to reproduce the problem.
>> The reason of the "no phase-space" is due that you run in multi-core.
>> Some core have already started the computation when some other are still
>> compiling the code.
>> (So far this is normal)
>> the problem is that the node started the computation crashed since the
>> color associated to one of the events can not be generated.
>> This create an error file which is wrongly interpret by the node of the
>> compilation like a no phase-space.
>>
>> So looks like they are still one case where the epsilon_ijk structure is
>> not handle in full generality.
>> I will look at that.
>>
>> Cheers,
>>
>> Olivier
>>
>> --
>> You received this bug notification because you are subscribed to the bug
>> report.
>> https://bugs.launchpad.net/bugs/1493832
>>
>> Title:
>> Phase space problem
>>
>> Status in MadGraph5_aMC@NLO:
>> New
>>
>> Bug description:
>> Greetings,
>>
>> When i try to generate this process listed above, the one with p p > tp
>> tp~ > z t z t~ > ... is said to not have enough phase space available and
>> tells me to check mass spectrum. But when I generate that process alone, no
>> such problem arises. And at the event generation, the following problem is
>> shown:
>> IOError: [Errno 2] No such file or directory:
>> '/home/tdvale/VLQ/LH_RH/MG5xProtos/TT/TT_1TeV/SubProcesses/P0_gg_ztztx_z_ll_t_wpb_wp_lvl_z_ll_tx_wmbx_wm_lvl/G1/results.dat'
>>
>> This error does not happen when the process is generated alone.
>>
>> The whole process list is the following:
>>
>> generate p p > tp tp~ > h t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~,\
>> w- > l- vl~)
>> add process p p > tp tp~ > h t z t~, (t > w+ b, w+ > l+ vl),( t~ > w- \
>> b~, w- > l- vl~), z > l+ l-
>> add process p p > tp tp~ > h t w- b~, (t > w+ b, w+ > l+ vl), w- > l- \
>> vl~
>> add process p p > tp tp~ > z t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
>> b~, w- > l- vl~), z > l+ l-
>> add process p p > tp tp~ > z t z t~, (t > w+ b...

[Tiago Vale (tiagval)]

Thank you very much

On Sat, Sep 26, 2015, 6:25 PM Olivier Mattelaer <
<email address hidden>> wrote:

> Hi Tiago,
>
>
> A workaround for this bug is to generate with the following syntax:
>
> generate p p > tp tp~ > h t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~,\
> w- > l- vl~) @1
> add process p p > tp tp~ > h t z t~, (t > w+ b, w+ > l+ vl),( t~ > w- \
> b~, w- > l- vl~), z > l+ l- @2
> add process p p > tp tp~ > h t w- b~, (t > w+ b, w+ > l+ vl), w- > l- \
> vl~ @3
> add process p p > tp tp~ > z t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
> b~, w- > l- vl~), z > l+ l- @4
> add process p p > tp tp~ > z t z t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
> b~, w- > l- vl~), z > l+ l- @5
> add process p p > tp tp~ > z t w- b~, (t > w+ b, w+ > l+ vl), z > l+ l\
> -, w- > l- vl~ @6
> add process p p > tp tp~ > w+ b h t~, (t~ > w- b~, w- > l- vl~), w+ > \
> l+ vl @7
> add process p p > tp tp~ > w+ b z t~, (t~ > w- b~, w- > l- vl~), w+ > \
> l+ vl, z > l+ l- @8
> add process p p > tp tp~ > w+ b w- b~, w+ > l+ vl, w- > l- vl~ @9
>
> Cheers,
>
> Olivier
>
>
> On 26 Sep 2015, at 00:37, Tiago Vale <email address hidden> wrote:
>
> > Hello Olivier,
> >
> > Thank you very much for your answer and your help. I appreciate it.
> >
> > Good work.
> > Cheers,
> > Tiago
> >
> > On Sat, Sep 26, 2015, 12:35 AM Olivier Mattelaer <
> > <email address hidden>> wrote:
> >
> >> Hi Tiago,
> >>
> >> I succeed to reproduce the problem.
> >> The reason of the "no phase-space" is due that you run in multi-core.
> >> Some core have already started the computation when some other are still
> >> compiling the code.
> >> (So far this is normal)
> >> the problem is that the node started the computation crashed since the
> >> color associated to one of the events can not be generated.
> >> This create an error file which is wrongly interpret by the node of the
> >> compilation like a no phase-space.
> >>
> >> So looks like they are still one case where the epsilon_ijk structure is
> >> not handle in full generality.
> >> I will look at that.
> >>
> >> Cheers,
> >>
> >> Olivier
> >>
> >> --
> >> You received this bug notification because you are subscribed to the bug
> >> report.
> >> https://bugs.launchpad.net/bugs/1493832
> >>
> >> Title:
> >> Phase space problem
> >>
> >> Status in MadGraph5_aMC@NLO:
> >> New
> >>
> >> Bug description:
> >> Greetings,
> >>
> >> When i try to generate this process listed above, the one with p p > tp
> >> tp~ > z t z t~ > ... is said to not have enough phase space available
> and
> >> tells me to check mass spectrum. But when I generate that process
> alone, no
> >> such problem arises. And at the event generation, the following problem
> is
> >> shown:
> >> IOError: [Errno 2] No such file or directory:
> >>
> '/home/tdvale/VLQ/LH_RH/MG5xProtos/TT/TT_1TeV/SubProcesses/P0_gg_ztztx_z_ll_t_wpb_wp_lvl_z_ll_tx_wmbx_wm_lvl/G1/results.dat'
> >>
> >> This error does not happen when the process is generated alone.
> >>
> >> The whole process list is the following:
> >>
> >> generate p p > tp tp~ > h t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~,\
> >> w- > l- vl~)
> >> add process p p > tp tp~ > h t z t~, (t > w+ b, w+ > l+ vl),( t~ > w- \
>...

[Tiago Vale (tiagval)]

Hello Olivier,

I just felt like I should tell you that this solution solved my problem.
Once again I am very thankful for your help.
Keep up the good work.

Cheers,
Tiago

On 09/26/2015 06:16 PM, Olivier Mattelaer wrote:
> Hi Tiago,
>
>
> A workaround for this bug is to generate with the following syntax:
>
> generate p p > tp tp~ > h t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~,\
> w- > l- vl~) @1
> add process p p > tp tp~ > h t z t~, (t > w+ b, w+ > l+ vl),( t~ > w- \
> b~, w- > l- vl~), z > l+ l- @2
> add process p p > tp tp~ > h t w- b~, (t > w+ b, w+ > l+ vl), w- > l- \
> vl~ @3
> add process p p > tp tp~ > z t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
> b~, w- > l- vl~), z > l+ l- @4
> add process p p > tp tp~ > z t z t~, (t > w+ b, w+ > l+ vl), (t~ > w- \
> b~, w- > l- vl~), z > l+ l- @5
> add process p p > tp tp~ > z t w- b~, (t > w+ b, w+ > l+ vl), z > l+ l\
> -, w- > l- vl~ @6
> add process p p > tp tp~ > w+ b h t~, (t~ > w- b~, w- > l- vl~), w+ > \
> l+ vl @7
> add process p p > tp tp~ > w+ b z t~, (t~ > w- b~, w- > l- vl~), w+ > \
> l+ vl, z > l+ l- @8
> add process p p > tp tp~ > w+ b w- b~, w+ > l+ vl, w- > l- vl~ @9
>
> Cheers,
>
> Olivier
>
>
> On 26 Sep 2015, at 00:37, Tiago Vale <email address hidden> wrote:
>
>> Hello Olivier,
>>
>> Thank you very much for your answer and your help. I appreciate it.
>>
>> Good work.
>> Cheers,
>> Tiago
>>
>> On Sat, Sep 26, 2015, 12:35 AM Olivier Mattelaer <
>> <email address hidden>> wrote:
>>
>>> Hi Tiago,
>>>
>>> I succeed to reproduce the problem.
>>> The reason of the "no phase-space" is due that you run in multi-core.
>>> Some core have already started the computation when some other are still
>>> compiling the code.
>>> (So far this is normal)
>>> the problem is that the node started the computation crashed since the
>>> color associated to one of the events can not be generated.
>>> This create an error file which is wrongly interpret by the node of the
>>> compilation like a no phase-space.
>>>
>>> So looks like they are still one case where the epsilon_ijk structure is
>>> not handle in full generality.
>>> I will look at that.
>>>
>>> Cheers,
>>>
>>> Olivier
>>>
>>> --
>>> You received this bug notification because you are subscribed to the bug
>>> report.
>>> https://bugs.launchpad.net/bugs/1493832
>>>
>>> Title:
>>> Phase space problem
>>>
>>> Status in MadGraph5_aMC@NLO:
>>> New
>>>
>>> Bug description:
>>> Greetings,
>>>
>>> When i try to generate this process listed above, the one with p p > tp
>>> tp~ > z t z t~ > ... is said to not have enough phase space available and
>>> tells me to check mass spectrum. But when I generate that process alone, no
>>> such problem arises. And at the event generation, the following problem is
>>> shown:
>>> IOError: [Errno 2] No such file or directory:
>>> '/home/tdvale/VLQ/LH_RH/MG5xProtos/TT/TT_1TeV/SubProcesses/P0_gg_ztztx_z_ll_t_wpb_wp_lvl_z_ll_tx_wmbx_wm_lvl/G1/results.dat'
>>>
>>> This error does not happen when the process is generated alone.
>>>
>>> The whole process list is the following:
>>>
>>> generate p p > tp tp~ > h t h t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~,\
>>> w- > l- vl~)
>>> add process p p >...

[Olivier Mattelaer (olivier-mattelaer)]

Thanks,

I have implemeted the equivalent of that fix automatically,

Thanks,

Olivier