You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have successfully installed both tinker8 and tinker9. Using GPU in tinker9 really speeds up calculations, thank you! However, I run into a problem. When I tried to minimize a structure, tinker9 gives the results shown below in Snapshot 1. However, if after removing file with extension *xyz_2 (to ensure that starting point is the same), tinker 8 is run, results are different (see Snapshot 2 below),
Also if you stop tinker8 after several steps when function values drops below 10^10 (that's somewhat empirical value, do not take it as carved in the stone...) and continue from that point with tinker9, everything runs smoothly (see snapshot 3). The force field used in the calculations is charm22.prm (from your distribution) and tinker.key does not contain any additional info, only path to forcefield file.
It seems that if F value is large at first step (and it may happen if your model file has some clashes), the value overflow happens and somewhere '-' sign is erroneously assigned to the results of calculations. Could you fix that problem? Without that it is not possible to use tinker9 for large-scale calculations in automated regime.
Thank you.
Petras Kundrotas, PhD
Associate Research Professor
Center for Computational Biology
The University of Kansas
Lawrence, Kansas 66046
United States of America
tel. +1 785 864 1963
email: [email protected]
LinkedIn: se.linkedin.com/pub/petras-kundrotas/69/b23/251/
Compiled at: 10:26:23 Dec 11 2023
Commit Date: Thu Nov 9 12:11:35 2023 -0800
Commit: 70bd052
GPU Device : Setting Device ID to 0 from GPU utilization
Limited Memory BFGS Quasi-Newton Optimization :
QN Iter F Value G RMS F Move X Move Angle FG Call Comment
0 -0.4484D+09 0.1631D+09 1
LBFGS -- Normal Termination due to SmallFct
Final Function Value : -448433300.8223
Final RMS Gradient : 163068839.3241
Final Gradient Norm : 19517916398.3256
---------------------------------------- End Snapshot 1 ---------------------------------------------------------------------
---------------------------------------- Snapshot 2 ---------------------------------------------------------------------
I have successfully installed both tinker8 and tinker9. Using GPU in tinker9 really speeds up calculations, thank you! However, I run into a problem. When I tried to minimize a structure, tinker9 gives the results shown below in Snapshot 1. However, if after removing file with extension *xyz_2 (to ensure that starting point is the same), tinker 8 is run, results are different (see Snapshot 2 below),
Also if you stop tinker8 after several steps when function values drops below 10^10 (that's somewhat empirical value, do not take it as carved in the stone...) and continue from that point with tinker9, everything runs smoothly (see snapshot 3). The force field used in the calculations is charm22.prm (from your distribution) and tinker.key does not contain any additional info, only path to forcefield file.
It seems that if F value is large at first step (and it may happen if your model file has some clashes), the value overflow happens and somewhere '-' sign is erroneously assigned to the results of calculations. Could you fix that problem? Without that it is not possible to use tinker9 for large-scale calculations in automated regime.
Thank you.
Petras Kundrotas, PhD
Associate Research Professor
Center for Computational Biology
The University of Kansas
Lawrence, Kansas 66046
United States of America
tel. +1 785 864 1963
email: [email protected]
LinkedIn: se.linkedin.com/pub/petras-kundrotas/69/b23/251/
----------------------------------------------------- Snapshot 1 ------------------------------------------------------------------------------
##########################################################################
Tinker9 -- Software Tools for Molecular Design
Version 1.4.0 February 2023
Copyright (c) Zhi Wang & the Ponder Lab
All Rights Reserved
##########################################################################
######################################################################
Compiled at: 10:26:23 Dec 11 2023
Commit Date: Thu Nov 9 12:11:35 2023 -0800
Commit: 70bd052
GPU Device : Setting Device ID to 0 from GPU utilization
Limited Memory BFGS Quasi-Newton Optimization :
QN Iter F Value G RMS F Move X Move Angle FG Call Comment
LBFGS -- Normal Termination due to SmallFct
Final Function Value : -448433300.8223
Final RMS Gradient : 163068839.3241
Final Gradient Norm : 19517916398.3256
---------------------------------------- End Snapshot 1 ---------------------------------------------------------------------
---------------------------------------- Snapshot 2 ---------------------------------------------------------------------
##########################################################################
Tinker --- Software Tools for Molecular Design
Version 8.10.6 October 2023
Copyright (c) Jay William Ponder 1990-2023
All Rights Reserved
##########################################################################
######################################################################
Limited Memory BFGS Quasi-Newton Optimization :
QN Iter F Value G RMS F Move X Move Angle FG Call Comment
---------------------------------------- End snapshot 2 ---------------------------------------------------------------------
--------------------------------------------- Snapshot 3 ---------------------------------------------------------------------
##########################################################################
Tinker --- Software Tools for Molecular Design
Version 8.10.6 October 2023
Copyright (c) Jay William Ponder 1990-2023
All Rights Reserved
##########################################################################
######################################################################
Limited Memory BFGS Quasi-Newton Optimization :
QN Iter F Value G RMS F Move X Move Angle FG Call Comment
--------------------------------------- End snapshot 3 ---------------------------------------------------------------------
The text was updated successfully, but these errors were encountered: