Hi, I have (very) high resolution data, for which direct summation method seems to give significantly better map than FFT (i.e. visible hydrogen vs not). My main problem is that refinement got from quite slow, to barely not really possible to handle slow. I have 2 question from that : Am I looking at an artifact ? Meaning should at this resolution (0.84A for CC1/2 at 30% and good overall statistics) see extreme difference between map quality with direct summation vs FFT ? And if yes, can I speed the process compiling phenix with openMP on our cluster ? Or shoul I just become patient ? Many thanks Clément
Dear Clement, as far as I understand, SHELXL is compiled against the Intel FFT library. Before, SHELXL used direct summation. If there was any noticeable difference, I am sure George Sheldrich would not have taken this step, as he would have had to expect a flood of complaints. My guess is that the difference is not because of the difference between FFT and SHELXL. Best regards, Tim On 11/02/2018 05:01 PM, Clement Degut wrote:
Hi, I have (very) high resolution data, for which direct summation method seems to give significantly better map than FFT (i.e. visible hydrogen vs not). My main problem is that refinement got from quite slow, to barely not really possible to handle slow. I have 2 question from that : Am I looking at an artifact ? Meaning should at this resolution (0.84A for CC1/2 at 30% and good overall statistics) see extreme difference between map quality with direct summation vs FFT ? And if yes, can I speed the process compiling phenix with openMP on our cluster ? Or shoul I just become patient ?
Many thanks
Clément
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]
-- -- Paul Scherrer Institut Tim Gruene - persoenlich - OSUA/204 Forschungsstrasse 111 CH-5232 Villigen PSI phone: +41 (0)56 310 5297 GPG Key ID = A46BEE1A
I agree, the difference is likely something else than "FFT vs direct summation"! Pavel On 11/3/18 00:19, Tim Gruene wrote:
Dear Clement,
as far as I understand, SHELXL is compiled against the Intel FFT library. Before, SHELXL used direct summation. If there was any noticeable difference, I am sure George Sheldrich would not have taken this step, as he would have had to expect a flood of complaints. My guess is that the difference is not because of the difference between FFT and SHELXL.
Best regards, Tim
On 11/02/2018 05:01 PM, Clement Degut wrote:
Hi, I have (very) high resolution data, for which direct summation method seems to give significantly better map than FFT (i.e. visible hydrogen vs not). My main problem is that refinement got from quite slow, to barely not really possible to handle slow. I have 2 question from that : Am I looking at an artifact ? Meaning should at this resolution (0.84A for CC1/2 at 30% and good overall statistics) see extreme difference between map quality with direct summation vs FFT ? And if yes, can I speed the process compiling phenix with openMP on our cluster ? Or shoul I just become patient ?
Many thanks
Clément
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]
Hi Clement,
I have (very) high resolution data, for which direct summation method seems to give significantly better map than FFT (i.e. visible hydrogen vs not).
at some point we spent a lot of time experimenting with this: Acta Cryst. (2004). A60, 19-32 On a fast calculation of structure factors at a subatomic resolution P. V. Afonine and A. Urzhumtsev https://doi.org/10.1107/S0108767303022062 and I can't remember seeing any significant difference in R-factors/maps. So.. there must be something special going on in your case. If you send me files I may be able to tell you what exactly is going on. Otherwise, if you'd like to tackle this yourself, you can start with changing sampling finess (resolution_factor) to something like 1./5 or 1/6., which will still be much faster than direct summation yet produce even greater match between approximations. Pavel
participants (3)
-
Clement Degut -
Pavel Afonine -
Tim Gruene