<div><br></div><div>Hello Juerg,</div><div><br></div><div>I have tried FFTW with NGRIDS=1 but it still keeps failing at pw_grid_assign step. Can you suggest more possible solutions? For your convenient, I attached my input, topology, psf and pdb files here. Thank you!</div><div><br></div><div>Best,</div><div>Yue</div><div><br></div><br><br>On Wednesday, August 1, 2012 1:36:24 AM UTC-7, jgh wrote:<blockquote class="gmail_quote" style="margin: 0;margin-left: 0.8ex;border-left: 1px #ccc solid;padding-left: 1ex;">Thanks for the analysis.
<br>
<br>Unfortunately, you cannot change the available radices for
<br>the FFTSG library.
<br>Using FFTW with extended lengths could be a solution.
<br>
<br>regards
<br>
<br>Juerg
<br>
<br>------------------------------<wbr>------------------------------<wbr>--
<br>Juerg Hutter Phone : ++41 44 635 4491
<br>Physical Chemistry Institute FAX : ++41 44 635 6838
<br>University of Zurich E-mail: <a href="mailto:hut...@pci.uzh.ch" target="_blank">hut...@pci.uzh.ch</a>
<br>Winterthurerstrasse 190
<br>CH-8057 Zurich, Switzerland
<br>------------------------------<wbr>------------------------------<wbr>---
<br>
<br>-----<a href="mailto:cp...@googlegroups.com" target="_blank">cp...@googlegroups.com</a> wrote: -----
<br>To: <a href="mailto:cp...@googlegroups.com" target="_blank">cp...@googlegroups.com</a>
<br>From: Noam Bernstein
<br>Sent by: <a href="mailto:cp...@googlegroups.com" target="_blank">cp...@googlegroups.com</a>
<br>Date: 07/31/2012 02:55PM
<br>Subject: Re: [CP2K:3949] pw_grid_assign error
<br>
<br>On Tue, Jul 31, 2012 at 7:37 AM, <<a href="mailto:hut...@pci.uzh.ch" target="_blank">hut...@pci.uzh.ch</a>> wrote:
<br>> Hi
<br>>
<br>> I have also to guess for this one. Could be an additional
<br>> constraint introduced by the QMMM solver that gets problematic
<br>> for parallel jobs.
<br>> Try
<br>> NGRIDS 1
<br>> or/and use FFTW. If the error still is there, we have
<br>> to take a closer look.
<br>
<br>Actually, I've tracked down the problem. I'd be happy to solve it, but I
<br>want to know how you want it solved. Basically, my finest grid has
<br>a size of 200, and I have 4 grid levels. That means 100, 50, and 25
<br>in addition. The fftsg library, which is used to select grid sizes,
<br>includes radixes of 200, 100, and 25, but not 50. But when the
<br>compatible grid size check happens (in pw_grid_find_n), it only checks
<br>the top and bottom levels, i.e. 200 and 25, which check out OK. Then,
<br>when it gets around to making all the subgrids, pw_grid_init_setup fails
<br>because 50 isn't a valid size.
<br>
<br>So, I can think of two things that should perhaps be done. One is
<br>to add 50 to the fftsg list of valid radixes. I don't understand enough
<br>about the fft algorithm to tell if this is acceptable. Also, the code in
<br>pw_grid_find_n should check all the subgrids, not just the coarsest
<br>one. Perhaps it should print a warning if one of the middle subgrids
<br>fails. I'd be happy to code up both those changes, but I don't want to
<br>do something (e.g. change the list of radixes) that'd break the code.
<br>
<br> Noam
<br>
<br>--
<br>You received this message because you are subscribed to the Google Groups "cp2k" group.
<br>To post to this group, send email to <a href="mailto:cp...@googlegroups.com" target="_blank">cp...@googlegroups.com</a>.
<br>To unsubscribe from this group, send email to <a href="mailto:cp2k%2Bun...@googlegroups.com" target="_blank">cp2k+unsubscribe@googlegroups.<wbr>com</a>.
<br>For more options, visit this group at <a href="http://groups.google.com/group/cp2k?hl=en" target="_blank">http://groups.google.com/<wbr>group/cp2k?hl=en</a>.
<br>
<br></blockquote>