[gmx-developers] gmx_fft_2d_real() output size

[Erik Lindahl]

Hi Benson,

Interesting - I've actually looked into spiral before, but at the time their generator was not open source. We can certainly add support for it, but the more interesting aspect is that we might be able to use it as a basis for our own SIMD-flavor FFTs. We have not been able to borrow the FFTW schedules for that due to their strict GPL license (since gromacs is LGPL).

Cheers,

Erik

--
Erik Lindahl <erik.lindahl at scilifelab.se>
Professor of Biophysics
Science for Life Laboratory
Stockholm University & KTH
Office (SciLifeLab): +46 8 524 81567
Cell (Sweden): +46 73 4618050 
Cell (US): 1 267 307 8746


> On Aug 10, 2019, at 10:38 PM, Benson Muite <benson_muite at emailplus.org> wrote:
> 
> Hi,
> 
> Might it be worth trying Spiral?
> 
> https://users.ece.cmu.edu/~franzf/papers/fftx-ecp-poster2019_rows.pdf
> 
> http://users.ece.cmu.edu/~franzf/papers/08510983_Spiral_IEEE_Final.pdf
> 
> http://www.spiral.net/doc/usermanual/gettingstarted.html#installing-spiral
> 
> Benson
> 
>> On 8/10/19 6:56 PM, Erik Lindahl wrote:
>> Hi Peter,
>> 
>> All our FFTs employ the same standard output formats as used by e.g. FFTW and and MKL, so for the full documentation it's probably easiest to read there.
>> 
>> Short story: All values are stored as proper real+complex parts (i.e., 2 real values), so depending on input size the output array might be larger. For in-place R2C transforms that usually means padding (but not out-of-place ones).
>> 
>> Cheers,
>> 
>> Erik
>> 
>>> On Sat, Aug 10, 2019 at 10:36 AM Peter Kasson <kasson at virginia.edu> wrote:
>>> Hi developers,
>>>   The doxygen docs are a little ambiguous on gmx_fft_2d_real(), and the source doesn't offer much clarification because we're wrapping external libraries.  Given a real-to-complex 2D FFT on a grid of size x by y, is the output a complex grid of size x by y also?  There are comments about in-place FFT (which I'm not doing) taking a 5*4 real array and writing a 5*2 complex array.
>>> 
>>> (The context of this is trying to compute power spectra for an analysis routine.)
>>> 
>>> Thanks,
>>> --Peter
>>> 
>>> ----------------------------------------------------------------------
>>> Peter Kasson, MD, PhD
>>> Associate Professor
>>> Departments of Molecular Physiology and Biological Physics
>>> and of Biomedical Engineering
>>> University of Virginia
>>> -- 
>>> Gromacs Developers mailing list
>>> 
>>> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List before posting!
>>> 
>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>> 
>>> * For (un)subscribe requests visit
>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers or send a mail to gmx-developers-request at gromacs.org.
>> 
>> 
>> -- 
>> Erik Lindahl <erik.lindahl at dbb.su.se>
>> Professor of Biophysics, Dept. Biochemistry & Biophysics, Stockholm University
>> Science for Life Laboratory, Box 1031, 17121 Solna, Sweden
>> 
>> 
> -- 
> Gromacs Developers mailing list
> 
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List before posting!
> 
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> 
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers or send a mail to gmx-developers-request at gromacs.org.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20190812/e67a441d/attachment-0003.html>