Verify ILC 2->6 tests

[Juergen Reuter]

different test schemes, particularly the DEFAULT_CHECKS logic. This should be done in Siegen, Dec 3-5 201[7-9];).

[Juergen Reuter]

This unfortunately we missed, will have to do this at the next phone meeting. But most of the clean-up happened.

[Juergen Reuter]

Once all features and commands are reimplemented, we can enable the missing self tests.

[Juergen Reuter]

model_test reinstated in r5041. There are now only a few open things.

[Juergen Reuter]

This can only be finalized at the very end.

[Juergen Reuter]

This is a list of missing (not yet re-implemented) self tests:

[ 1. libs.sin (#583) eliminated in r5602 (obsolete) ]

2. user_cuts.sin (#529)
3. user_strfun.sin (#529)

[ 4. ilc.sin (#500) Done meanwhile.]

5. mlm_matching_fsr.sin (#44) Preparations.
6. isr_shower.sin (#44)

[ 7. jets_xsec.sin (#44, #545) Done in r5644.]

8. sm_ext.sin

[Juergen Reuter]

MSSM extended self tests: mssm_ext-aa.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-bb.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-bt.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-dd.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-dd2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ddckm.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-dg.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ee.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ee2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-en.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ga.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-gg.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-gw.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-gz.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-tn.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-tt.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ug.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-uu.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-uu2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-uuckm.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-wa.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ww.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-wz.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-za.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-zz.run --- works 64bit (dbl+quad) + 32bit (dbl+quad)

So, apparently all MSSM tests work on 64bit.

[Juergen Reuter]

NMSSM extended self tests: nmssm_ext-aa.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-bb1.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-bb2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-bt.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-dd1.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-dd2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-ee1.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-ee2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-en.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-ga.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-gg.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-gw.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-gz.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-qg.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-tn.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-tt1.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-tt2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-uu1.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-uu2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-wa.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-ww1.run --- works 64bit (dbl+quad) + 32bit (dbl+qquad) nmssm_ext-ww2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-wz.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-za.run --- works 64bit (dbl+quad) + 32bit (dbl+quad)

[Juergen Reuter]

nmssm_ext-tt2.run leads to 1 failure in the process e3, E3 => h01 h01:

************************************************************************
* Checking e3,E3 => h01,h01	 @ sqrt(s) = 5000 GeV
************************************************************************
[user variable] error_sum =  2.769089683983E-07
Expecting 4.9290295E-06
integral(nmssm_ext_tt2_50) =  3.960746804051E-06
[user variable] pull =  3.496754552769E+00
WARNING: pull > 3 in e3,E3 => h01,h01	 @ sqrt(s) = 5000 GeV
tolerance =  8.307269051948E-07
******************************************************************************
*** ERROR: expect: failure
******************************************************************************

Will have to look into this.

[Juergen Reuter]

Replying to jr_reuter:

nmssm_ext-tt2.run leads to 1 failure in the process e3, E3 => h01 h01:

************************************************************************
* Checking e3,E3 => h01,h01	 @ sqrt(s) = 5000 GeV
************************************************************************
[user variable] error_sum =  2.769089683983E-07
Expecting 4.9290295E-06
integral(nmssm_ext_tt2_50) =  3.960746804051E-06
[user variable] pull =  3.496754552769E+00
WARNING: pull > 3 in e3,E3 => h01,h01	 @ sqrt(s) = 5000 GeV
tolerance =  8.307269051948E-07
******************************************************************************
*** ERROR: expect: failure
******************************************************************************

Will have to look into this.

It turns out, this has difficulties to find all channels. Given enough calls, we get the desired result. Maybe I try a different seed. But it's nothing to worry about.

[Juergen Reuter]

Replying to jr_reuter:

Replying to jr_reuter:

nmssm_ext-tt2.run leads to 1 failure in the process e3, E3 => h01 h01:

************************************************************************
* Checking e3,E3 => h01,h01	 @ sqrt(s) = 5000 GeV
************************************************************************
[user variable] error_sum =  2.769089683983E-07
Expecting 4.9290295E-06
integral(nmssm_ext_tt2_50) =  3.960746804051E-06
[user variable] pull =  3.496754552769E+00
WARNING: pull > 3 in e3,E3 => h01,h01	 @ sqrt(s) = 5000 GeV
tolerance =  8.307269051948E-07
******************************************************************************
*** ERROR: expect: failure
******************************************************************************

Will have to look into this.

It turns out, this has difficulties to find all channels. Given enough calls, we get the desired result. Maybe I try a different seed. But it's nothing to worry about.

In r5653 this has been fixed by enhancing the number of iterations and calls for this particular part of the NMSSM test suite.

[Juergen Reuter]

Replying to jr_reuter:

NMSSM extended self tests: nmssm_ext-aa.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-bb1.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-bb2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-bt.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-dd1.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-dd2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-ee1.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-ee2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-en.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-ga.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-gg.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-gw.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-gz.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-qg.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-tn.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-tt1.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-tt2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-uu1.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-uu2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-wa.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-ww1.run --- works 64bit (dbl+quad) + 32bit (dbl+qquad) nmssm_ext-ww2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-wz.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) nmssm_ext-za.run --- works 64bit (dbl+quad) + 32bit (dbl+quad)

So, the NMSSM tests run through on 32 and 64 bit with 8 and 10 byte precision.

[Juergen Reuter]

Replying to jr_reuter:

MSSM extended self tests: mssm_ext-aa.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-bb.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-bt.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-dd.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-dd2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ddckm.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-dg.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ee.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ee2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-en.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ga.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-gg.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-gw.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-gz.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-tn.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-tt.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ug.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-uu.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-uu2.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-uuckm.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-wa.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-ww.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-wz.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-za.run --- works 64bit (dbl+quad) + 32bit (dbl+quad) mssm_ext-zz.run --- works 64bit (dbl+quad) + 32bit (dbl+quad)

So, apparently all MSSM tests work on 64bit and 32bit with 8 and 10 byte precision.

[Juergen Reuter]

User cut and user structure function tests are implemented as XFAIL and will be deferred to #529. This ticket remains as ticket for the ILC 2->6 precision tests (shifted to a later version)

[Juergen Reuter]

Milestone renamed

[Juergen Reuter]

Milestone renamed

[Juergen Reuter]

Milestone renamed

[Juergen Reuter]

Milestone renamed

[Juergen Reuter]

Transferred to Gitlab issue 309.