1 GROMACS 2018.3 release notes
2 ----------------------------
4 This version was released on TODO, 2018. These release notes document
5 the changes that have taken place in GROMACS since version 2018.2, to fix known
6 issues. It also incorporates all fixes made in version TODO and
7 earlier, which you can find described in the :ref:`release-notes`.
9 Fixes where mdrun could behave incorrectly
10 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
12 Fix Conjugate Gradient assertion failure at end of minimization
13 """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
15 When the final step coincided with a coordinate output step,
16 conjugate gradient minimization would exit with an assertion failure
17 instead of writing confout.gro.
21 Multi-domain Conjugate Gradient minimimization no longer segfaults.
22 """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
26 Fix pairlist buffer with Brownian Dynamics
27 """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
29 With Brownian Dynamics and bd-fric > 0, the Verlet pairlist buffer would
30 be determined with incorrect masses for constrained atoms and virtual
31 sites. This would lead to a too small buffer for typical atomistic
32 systems with constraints.
36 Avoid "atom moved to far" errors
37 """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
39 The introduction of the dual pair list has led to larger nstlist values,
40 which leads to larger atom displacements between domain decomposition
41 steps. This has made it more likely that the errors
42 "An atom moved too far between two domain decomposition steps" and
43 "N particles communicated to PME rank M are more than 2/3 times the cut-off
44 out of the domain decomposition cell ..." appear for stable systems.
45 Now atom displacements are correctly taken into account when determining
46 the minimum cell size, so these errors should only appear for unstable systems.
50 grompp now checks that pull groups are not close to half the box size
51 """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
53 Pull groups that use a reference atom for periodic boundary treatment
54 should have all their atoms well within half the box size of this reference.
55 When this is not the case, grompp will issue a warning.
59 Fixed segmentation fault in mdrun with QM/MM ONIOM scheme
60 """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
64 Fixes for ``gmx`` tools
65 ^^^^^^^^^^^^^^^^^^^^^^^
67 Fix bfactor output error caused by fix for :issue:`2511`
68 """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
69 The fix for the PQR file output broke the output of bfactors from other tools.
73 Made sure that gmx rms can skip values
74 """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
75 When requested to skip values, gmx rms would still output all values despite
76 the option. Now it only outputs values that are requested to be processed.
80 Fixes to improve portability
81 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
86 Improve OpenCL kernel performance on AMD Vega GPUs
87 """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
88 The OpenCL kernel optimization flags did not explicitly turn off denorm handling
89 which could lead to performance loss. The optimization is now explicitly turned
90 on both for consistency with CUDA and performance reasons.
91 On AMD Vega GPUs (with ROCm) kernel performance improves by up to 30%.