[QE-users] FFT order too large
Lorenzo Paulatto
paulatz at gmail.com
Fri Jan 22 09:39:16 CET 2021
Hello Aaron,
for this version of QE, it is what it is, and the only choice you have is to modify and recompile the code. You can ask your computer guys to do it for everybody, I guess, since it has really no side effect. On the other hand, for the next version of QE, and considering that the CPU power keeps increasing, we may think about increasing this limit. I would recommend opening an issue in the gitlab tracker https://gitlab.com/QEF/q-e/-/issues if you want.
regards
--
Lorenzo Paulatto - Paris
On Jan 21 2021, at 11:35 pm, Aaron Friesz <friesz at usc.edu> wrote:
> Hello,
>
> I am receiving this error message:
> Error in routine good_fft_order (5076):
> fft order too large
>
> I have found this post regarding the error: https://lists.quantum-espresso.org/pipermail/users/2019-April/042636.html
>
> I have a few more questions.
>
> 1 ) In order to fix this it is stated that a change can be made in FFTXlib/fft_param.f90. This implies that it must then be recompiled. Is this correct? (I'd rather avoid this as it will make me responsible for maintaining the software, instead of our computing resources division.)
>
> 2 ) Is the number in parentheses, (5076), the number of terms in the fft? Are there any heuristics to reducing this number short of reducing the number of atoms in the system?
>
> 3 ) Why is this check made? Is the FFT computed in every loop of the algorithm, thus incurring a large computational cost? (From what I remember FFTW should be very fast on arrays of this size, but if it is calculated often that could be an issue.)
>
> Thank you for your help!
>
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20210122/ba89b9c5/attachment.html>
More information about the users
mailing list