1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
|
\documentclass{article}
\usepackage[a4paper]{geometry}
\newcommand\siesta{\textsc{SIESTA}}
\newcommand\tsiesta{\textsc{TranSIESTA}}
\newcommand{\fdf}[1]{\textbf{#1}}
\newcommand{\opt}[1]{\textbf{#1}}
\newcommand{\code}[1]{\texttt{#1}}
\title{Compatibility issues across versions}
\author{\siesta}
\begin{document}
\maketitle
\clearpage
This list describes the compatibility issues when using different versions of \siesta.
\begin{description}
\setlength\itemsep{1pt}
\setlength\topsep{1pt}
\item[4.1]%
\begin{itemize}
\item %
\tsiesta\ has been fully redeveloped. The entire algorithm has been revised and
substantial improvements on performance and numerical precision has been achieved. One
cannot compare, directly, the results although the transmissions should be (roughly)
equivalent.
\item %
\fdf{DM.MixSCF1} is now defaulted to \fdf{true}.
\item %
Nearly all flags regarding the mixing options have changed name. However, the old
flags are still read (and used). In future versions they may be fully removed.
\begin{itemize}
\item All mixing routines are re-written. This means that one cannot compare with
prior versions in terms of the convergence path. Sometimes they will differ due to numerics.
\end{itemize}
\item \fdf{NumberOfSpecies} and \fdf{NumberOfAtoms} are now defaulted to \# of lines
in their respective blocks. If they are specified, however, they will be honoured.
\end{itemize}
\item[0 --- 4.0-b2] The following compatibility issues should be remarked when
comparing with any later version of \siesta.
\begin{itemize}
\item \emph{Electric field and dipole correction for slab calculations}
Older versions applied an incorrect dipole correction when also using an
electric field (old behaviour may be recovered by forcing \code{SlabDipoleCorrection .false.}).
Older versions over-estimated the energy contribution from the dipole correction by a
factor of 2 (old behaviour cannot be recovered).
\item\emph{Changes in the default precision from grid magnitudes}
The grid functions (charge densities, potentials, etc) were in
single precision by default in the 3.X versions, but are in double
precision by default for post-3.X versions. The \code{phi} array that
holds the values of the basis orbitals on the real-space grid is kept in
single precision. Please take this into account if you compare the
results with those of siesta-3.X runs. See the manual in both
versions for more information.
\item\emph{Changes in the geometry used for the analysis of the
electronic structure}
The update of the atomic coordinates after computing the forces on
the atoms no longer affects the analysis of the electronic
structure at the end of a run. To do the analysis using the actual
coordinates used in the computation of the electronic structure, the
coordinates are now reset to those values in \code{siesta\_analysis}. The old
behavior can be recovered by using by setting the backward
compatibility switch \opt{compat-pre-v4-dynamics} to 'true'.
This change will slightly alter the output of most calculations and the
detailed results of any post-processing. Keep this in mind if you
need to maintain coherency within a project.
\item\emph{Changes in the handling of the density-matrix and
hamiltonian after convergence of the scf cycle}
To guarantee that the DM and H used in the electronic-structure
analysis correspond to the latest electronic structure computed,
there should be no mixing at the end of the scf loop. This
behavior, controlled by the value of \opt{scf-mix-after-convergence}, is
now the default. The use of DM\_out instead of DM\_mixed provides
also a truly variational total-energy and more consistent
forces. If the scf cycle is carried out mixing H, H(DM\_in) is
recovered by re-setting H to Hold in \code{siesta\_forces} at the end of
the cycle.
Also, to preserve the original H for analysis it is necessary to
avoid overwriting it in \code{post\_scf\_work}. This is now done by
default, except if the option \opt{scf-recompute-h-after-scf} is
'true'.
There is another switch for pre-v4.0 compatibility in this
context: If \opt{compat-pre-v4-dm-h} is 'true', the
\opt{scf-mix-after-convergence} and \opt{scf-recompute-h-after-scf}
options are activated too.
This change will slightly alter the output of most calculations
and the detailed results of any post-processing. Keep this in mind
if you need to maintain coherency within a project.
\end{itemize}
\end{description}
\end{document}
|