[gpaw-users] Failed test iron_sf_gssALDA

[Jens Jørgen Mortensen]

Den 12.06.2020 kl. 16.28 skrev Christopher Patrick via gpaw-users:
> Dear all,
> 
> I have just installed gpaw on a new box.
> 
> All the tests either run fine or skip, except one:

Thank you for reporting this.  You can ignore that one.  I can see that 
the tolerance has already been increased in the development version of GPAW.

Jens Jørgen

> response/iron_sf_gssALDA.py                  13.764  FAILED!
> #############################################################################
> Traceback (most recent call last):
>    File "/home/patrick/codes/gpaw-20.1.0/gpaw/test/__init__.py", line 
> 672, in run_one
>      exec(compile(fd.read(), filename, 'exec'), loc)
>    File 
> "/home/patrick/codes/gpaw-20.1.0/gpaw/test/response/iron_sf_gssALDA.py", 
> line 106, in <module>
>      equal(mw2, test_mw2, eta * 250)
>    File "/home/patrick/codes/gpaw-20.1.0/gpaw/test/__init__.py", line 
> 29, in equal
>      raise AssertionError(msg)
> AssertionError:  174.2664865364513 != 176.91 (error: 
> |-2.643513463548686| > 2.5)
> #############################################################################
> 
> Should I be worried?  |-2.6| is nearly < 2.5, after all...
> 
> I am linked to openblas, running on an AMD epyc 7542 with
> 
> python-3.6.9
> gpaw-20.1.0
> ase-3.19.1
> numpy-1.18.5
> scipy-1.4.1
> libxc-4.2.3
> 
> Thanks very much,
> 
> Chris
> 
> _______________________________________________
> gpaw-users mailing list
> gpaw-users at listserv.fysik.dtu.dk
> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>