[QE-users] qe v7.5 self test metaGGA-spin.in failed. Cause for concern?

[Michael Dayan]

Okay thank you for the insight. I figured it was not a show stopper. Good
to have the additional context regarding it being a change from the
historical norm, and that it is associated with libxc which wasn't clear to
me.

On Sun, Nov 9, 2025 at 2:36 AM Pietro Davide Delugas <pdelugas at sissa.it>
wrote:

> No, at least not for what concerns correctness of the results.
> With the new versions of libxc that test is regularly  taking 35
> iterations to converge instead of the 25 in the benchmark, but converges to
> the same ground state as in the benchmark.
> Pietro
>
> Il 9 nov 2025 1:22 AM, Michael Dayan via users <
> users at lists.quantum-espresso.org> ha scritto:
> Hello,
>
> I am new to QE and I finally got v7.5 installed w/flags for libxc v7.0.0
> This is in a virtual machine Ubuntu 24.04.3 LTS
> When I ran the QE self test it passed 251 out of 252 tests.
>
> The only one that failed had the following console output.
>
> pw_metaGGA - metaGGA-spin.in: **FAILED**.
> n1
>     ERROR: absolute error 1.00e+01 greater than 8.00e+00. (Test: 25.0.
> Benchmark: 35.0.)
>
> Is this cause for any concern? and if so is there any further diagnostics
> to pinpoint the root cause?
> Sincerely,
> Michael Dayan
> --
> Master of Nanoengineering 2024 - Present
> Bachelor of Science in Electrical Engineering 2021
>
> *Michael Dayan*
>
> Pack Motorsports FSAE - Mentor Hardware Engineering
>
> College of Engineering - Materials Science and Engineering Dept.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20251109/725d4d6d/attachment.html>