-
Notifications
You must be signed in to change notification settings - Fork 328
/
Copy pathReleaseNotes4.10.5.html
2812 lines (2728 loc) · 141 KB
/
ReleaseNotes4.10.5.html
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
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<HTML>
<HEAD>
<TITLE>Geant4 10.5 Release Notes</TITLE>
</HEAD>
<BODY BGCOLOR='F0F0F0'>
<P> </P>
<BR>
<P> </P>
<P ALIGN="Center">
<FONT SIZE="+4" COLOR="#238E23">
<B>Geant4 10.5 Release Notes</B>
</FONT>
<TABLE WIDTH="100%">
<TR><TD ALIGN="Right">
<FONT SIZE="-1" COLOR="#5C3317">
<B><I>December 7<SUP>th</SUP>, 2018</I></B>
</FONT>
</TD></TR>
</TABLE>
<BR><BR>
<HR ALIGN="Center" SIZE="7%">
<P>
The code and binary libraries for the supported systems are available
through our <A TARGET="ext" HREF="http://cern.ch/geant4/support/download">Source
Code Web page</A>.
</P>
<P>
<I>We are grateful for the efforts of Geant4 users who have provided
detailed feedback or comprehensive reports of issues.
We thank in particular those who have contributed corrections,
improvements or developments included in this release. </I>
</P>
<P>
Please refer to the
<A TARGET="ext" HREF="http://cern.ch/geant4/support/user_documentation">Geant4
User Documentation</A> for further information about using Geant4.
</P>
<H2><I>Contents</I></H2>
<OL>
<LI><A HREF="#1.">Supported and Tested Platforms</A></LI>
<LI><A HREF="#2.">Supported CLHEP version</A></LI>
<LI><A HREF="#3.">Items for migration of the user code</A></LI>
<LI><A HREF="#4.">New Developments and Capabilities</A></LI>
<LI><A HREF="#5.">Expected effects on physics and performance</A></LI>
<LI><A HREF="#6.">Known Run-Time Problems and Limitations</A></LI>
<LI><A HREF="#7.">Compilation Warnings</A></LI>
<LI><A HREF="#8.">Geant4 Software License</A></LI>
<LI><A HREF="#9.">Detailed list of changes and fixes</A></LI>
</OL>
<P> </P>
<A NAME="1."></a>
<HR>
<!-- ============================================== -->
<H2>1. Supported and Tested Platforms</H2>
Platforms:
<UL>
<LI>Linux, gcc-4.8.5.<BR>
Tested on 64 bit architectures (Intel or AMD) with
CERN CentOS Linux 7 (CC7) (based on CentOS Linux 7).</LI>
<LI>MacOSX 10.14 Mojave with llvm/clang-6.0.1 (Apple LLVM/Clang-10.0.0)</LI>
<LI>Windows-10 with Visual C++ 14.11 (Visual Studio 2017)</LI>
</UL>
More verified and tested configurations (64 bits):
<UL>
<LI>Linux, gcc-4.9.3/5.4.0/6.3.0/7.3.0/8.2.0, clang-3.9/5.0</LI>
<LI>Linux, Intel-icc 19.0</LI>
<LI>MacOSX 10.11, 12, 13 with clang-3.8/4.0/5.0</LI>
<LI>Windows-10 with Visual C++ 14.0 (Visual Studio 2015)</LI>
</UL>
<P> </P>
<A NAME="2."></A>
<HR>
<!-- ============================================== -->
<H2>2. Supported CLHEP version</H2>
This release of Geant4 requires and has been verified with
<A TARGET="ext" HREF="http://cern.ch/clhep/">CLHEP</A>, release <B>2.4.1.0</B>.
Use of a different CLHEP version may cause incorrect simulation results.<BR>
NOTE: an internal module of the relevant CLHEP classes is provided and can be
used as alternative to an external CLHEP library installation.
<P> </P>
<A NAME="3."></A>
<HR>
<!-- ============================================== -->
<H2>3. Items for migration of the user code</H2>
Listed here is some relevant information on developments included in this
release.
Note that for all users a full re-installation of libraries (or a full
re-compilation) and a re-compilation of user applications is required.
<P> </P>
<B>General</B>
<P> </P>
<UL>
<LI>Minimum required version of CMake to build Geant4 is 3.3.</LI>
<LI>Due to the migration from POSIX threading to C++11 threading, enabling
multi-threading on Windows for use with either static or dynamic
libraries, the following migration is required:
<UL>
<LI>Any usage of <TT>G4THREADCREATE</TT> should pass a reference to an
allocated <TT>G4Thread</TT> pointer.</LI>
<UL>
<LI><TT>G4THREADCREATE</TT> (previously a macro) is
now a variadic template function that accepts an unlimited
number of arguments to the function launched on the thread.</LI>
<LI><TT>G4Thread t; G4THREADCREATE(&t, myfunction, myarg);</TT>
is no longer valid</LI>
<LI><TT>G4Thread* t = new G4Thread();
G4THREADCREATE(t, myfunction, myarg);</TT> is required</LI>
</UL>
<LI><TT>G4MUTEX_INITIALIZER</TT>, <TT>G4MUTEXINIT</TT>, and
<TT>G4MUTEXDESTROY</TT> are now empty macros and their usage is
unnecessary and thus deprecated.</LI>
</UL></LI>
</UL>
<P> </P>
<B>Geometry</B>
<P> </P>
<UL>
<LI>The behaviour of G4Transportation and G4CoupledTransportation
concerning the treatment of looping particles has been reviewed:
<UL>
<LI>From this release on, only stable particles are killed if they
'loop', i.e. take more than the maximum (default 1000) integration
steps in one physics step; unstable particles are now propagated
indefinitely.</LI>
<LI>Warning exceptions with full details of the particle, location and
volume are now generated for stable looping particles which are killed
(as before, this only happens if their energy exceeds a threshold
'warning' energy).
In previous releases a short message with only energy and location
was generated, and only for installations which did not use
G4_NO_VERBOSE. Typically these warnings occur predominantly in vacuum
or other low density media.</LI>
</UL></LI>
<LI>As before the killing of loopers and potential warnings are controlled by
two energy thresholds and a step number parameter. Default settings imply:
<UL>
<LI><TT>100 MeV</TT> (warning energy): below this, tracks are killed
silently.</LI>
<LI><TT>250 MeV</TT> (important energy): above this, tracks are given
multiple chances (10 physics steps).</LI>
</UL>
These settings are fully under user control. The new methods in
G4PhysicsListHelper allow the user to select a set of low (or the above
high) values of the 'looper' energy thresholds.
Existing methods of G4[Coupled]Transportation can be used for fine-grained
control of the parameter values.
Additional information regarding how to reduce the incidence of looping
particles is provided in an expanded section on Transportation in the
Users Guide for Applications Developers.</LI>
<LI>G4MagneticField has been corrected to no longer inherit from
G4ElectroMagneticField; this anomaly used to allow a user to provide a
'full' EM field (mixed electric and magnetic), but obtain the force only
from the B-field when using the equation created in the G4ChordFinder
constructor. With this correction it is no longer possible to use the
equation of motion for a pure magnetic field for the case of a mixed
electromagnetic field. Instead the full chain of classes for equation
of full EM field, steppers and driver must be used for the case of the
'full' EM field.</LI>
</UL>
<P> </P>
<B>Electromagnetic and optical physics</B>
<P> </P>
<UL>
<LI>The new data set G4EMLOW-7.7 should be used.</LI>
</UL>
<P> </P>
<B>Hadronic physics</B>
<P> </P>
<UL>
<LI>The new data sets <TT>G4PhotonEvaporation-5.3</TT>, <TT>G4INCL-1.0</TT>,
<TT>G4RadioactiveDecay-5.3</TT>, <TT>G4SAIDDATA-2.0</TT> and
<TT>G4PARTICLEXS-1.1</TT> should be used.</LI>
<LI>Requiring new <TT>G4INCL-1.0</TT> data set and corresponnding G4INCLDATA path
environment variable when using INCLXX physics.</LI>
<LI>The old G4NEUTRONXS data set is now replaced by <TT>G4PARTICLEXS-1.1</TT>,
and the corresponding G4PARTICLEXSDATA environment variable should be
now used.</LI>
</UL>
<P> </P>
<B>Analysis</B>
<P> </P>
<UL>
<LI>The default mode for Root n-tuple merging (in multi-threaded runs) is
changed to row-wise in order to preserve an "event point of view" (the
n-tuple rows) after merging. The output files written with this mode use a
different data schema (organisations of TBranches and TLeaves) than
before and so require adapting the macros for reading back the data.</LI>
</UL>
<P> </P>
<B>Data Sets</B>
<P> </P>
This release introduces new data set versions.
Please see the corresponding details in
<A href="#data-notes">Section 9</A> of this document.
<UL>
<LI>New data set versions: <TT>G4EMLOW-7.7, G4RadioactiveDecay-5.3,
G4PhotonEvaporation-5.3, G4INCL-1.0, G4SAIDDATA-2.0,
G4PARTICLEXS-1.1</TT>.</LI>
<LI>In order to use <TT>ParticleHP</TT> for charged particles
(protons, deuterons, tritons, He3 and alphas), an optional data
set <TT>G4TENDL-1.3.2</TT> is required, and should be downloaded in
addition from the
<A HREF="http://cern.ch/geant4/support/download">Geant4 web site</A>.</LI>
</UL>
<A NAME="4."></A>
<HR>
<!-- ============================================== -->
<H2>4. New Developments and Capabilities </H2>
<B>General</B>
<UL>
<LI>Replaced POSIX threading with C++11 threading, enabling multi-threading
on Windows for use with either static or dynamic libraries.</LI>
<UL>
<LI><TT>G4THREADCREATE</TT> is now a variadic template function
(unlimited arguments) instead of a macro</LI>
<LI><TT>G4THREADCREATE</TT> can be used outside of MT. In sequential
mode, <TT>G4THREADCREATE</TT> will execute the function on the
calling thread.
</UL>
<LI><TT>G4Mutex</TT> is now a typedef to <TT>std::mutex</TT>,
even in sequential mode.
<LI>Added <TT>G4RecursiveMutex</TT> typedef to
<TT>std::recursive_mutex</TT>.</LI>
<LI>Provided generic template functions <TT>G4TypeMutex<T>()</TT> and
<TT>G4TypeRecursiveMutex<T>()</TT> to create and access static
instances of mutexes without the need for declarations.</LI>
<UL>
<LI>When combined with <TT>G4AutoLock</TT>, this enables easy creation of
class-specific locks, e.g.
<TT>G4AutoLock lock(G4TypeMutex<decltype(this)>());</TT></LI>
</UL>
<LI><TT>G4AutoLock</TT> now inherits from <TT>std::unique_lock</TT>, even in
sequential mode.</LI>
<UL>
<LI>In sequential mode, <TT>G4AutoLock</TT> does not lock upon construction,
i.e. <TT>std::defer_lock</TT></LI>
<LI>Added overloads to accept reference to mutex, in addition to previous
form of a pointer to a mutex.</LI>
</UL>
</UL>
<B>Analysis</B>
<UL>
<LI>Switched the default Root ntuple merge mode to row-wise in analysis
tools and added new <TT>G4RootAnalysisManager::SetNtupleRowWise(G4bool)</TT>
function which can be used to override the defaults. </LI>
<LI>Added "Boolean reset" argument in <TT>G4AnalysisManager::CloseFile()</TT>
function, allowing to close a file without resetting data.</LI>
</UL>
<B>Digitization & Hits</B>
<UL>
<LI>Added new class G4StatAnalysis, a lightweight statistics class that
calculates the mean, FOM, relative error, standard deviation, variance
and other quantities. Compatible with usage in G4THitsMap and (new)
G4THitsVector -- replacing <TT>G4THitsMap<G4double></TT> with
<TT>G4THitsMap<G4StatAnalysis></TT> and can provide a quick
implementation of accumulating statistics for scoring.</LI>
<LI>Added new class G4THitsVector, a generic hits container compatible with
G4THitsMap, that enables the underlying storage container to utilize a
sequential container instead of a map (saving memory); additionally, it
allows the underlying container to store data as non-pointer types
(saving more memory).</LI>
<UL>
<LI>These containers are intended to be used at the G4Run level to
accumulate data over the course of the run and it is not recommended
for use at the event level -- a G4THitsMap is ideal at the event level
due to the general sparsity of hits per index but at the run level,
G4THitsVector becomes ideal with respect to memory usage as
approximately 40% of the total available indices accumulate non-zero
hits.</LI>
<LI>See <I>DICOM2</I> and <I>ThreadsafeScorers</I> for usage examples,
including demonstrations of G4THitsVector + non-pointer data storage
and the new style of generic iteration over the hits container classes
that is valid for both G4THitsMap and G4THitsVector (the previous form
of iteration over G4THitsMap is still valid, and will remain valid,
for G4THitsMap types).</LI>
</UL>
<LI>Introduced a new interface class, G4VScoreNtupleWriter, which allows
automated storing of scorers hits with Geant4 analysis tools. Concrete
implementations (G4ScoreNtupleWriter with associated messenger) are
demonstrated in new provided extended/analysis examples.</LI>
</UL>
<B>Geometry</B>
<UL>
<LI>A configuration option allows an installation to use the
<A HREF="https://gitlab.cern.ch/VecGeom/VecGeom/">VecGeom Unified Solids
Library</A> primitives, to replace the original Geant4 solids, as in the
previous release. The path to an external installation of the VecGeom
library (VecGeom version <TT>v01.01.00</TT>) should be specified.
See the associated instructions file for configuration and installation.</LI>
<LI>Enabled VecGeom wrappers for G4ExtrudedSolid, G4TessellatedSolid, G4Hype
and G4Tet.</LI>
<LI>Reimplemented <TT>EstimateSurfaceArea()</TT> for approximate calculation
of the surface area of a solid with new algorithm providing more
accurate estimation without performance penalty.</LI>
<LI>Introduced Bulirsch-Stoer method of integration, an alternative to
Runge-Kutta based on the mid-point method.</LI>
<LI>New G4InterpolationDriver field driver class using Runge-Kutta method
with interpolation property to integrate EoM with error control.</LI>
<LI>Updated field steppers to make use of cached field vales.</LI>
<LI>Improved reporting of looping particle, adding info on density,
fraction (%) of proposed step that was completed and momentum.</LI>
</UL>
<B>Electromagnetic physics</B>
<UL>
<LI>Use new parameter <TT>MscEnergyLimit()</TT> (default is <TT>100 MeV</TT>)
in EM physics constructors, allowing to customise transition from
low-energy multiple scattering to high energy combination of multiple
and single scattering.</LI>
<LI>Alg96 lateral displacement algorithm used in G4UrbanMscModel in all
EM physics.</LI>
<LI>Default angular generator for bremsstrahlung is G4ModifiedTsai.</LI>
<LI>Revisions of G4BetheHeitlerModel and G4PairProductionRelModel with
improved screening function approximation, improved LPM function
approximation, angular generation and efficiency of sampling
of the final state.</LI>
<LI>New gamma conversion model G4BetheHeitler5DModel.</LI>
<LI>Added optional possibility to use ICRU90 data.</LI>
<LI>New optional ion ionisation models G4LindhardSorensenIonModel,
G4AtimaEnergyLossModel, G4AtimaFluctuations.</LI>
<LI>Added new classes for the simulation of elastic scattering of gammas.</LI>
</UL>
<B>Hadronic physics</B>
<UL>
<LI>For both Fritiof (FTF) and Quark-Gluon-String (QGS) string models,
we have released the latest versions, which have been developed since
the release 10.2 but have not been included in the two previous releases,
10.3 and 10.4, to keep stable the simulation of hadronic showers
(note that snapshots of these developments have been made available
in previous beta releases: 10.3.beta, 10.4.beta and 10.5.beta).</LI>
<LI>New tuning of the FTF model parameters, for both string formation
and string fragmentation. Introduced smearing of delta-isobar mass and
improved di-quark fragmentation into baryons in Lund string fragmentation.
First implementation of rotating strings in FTF, at the level of string
fragmentation, with introduction of Mt distribution of hadrons.
Improved process probability parameterizations for Pi-Nucleon interactions;
corrected calculation of nuclear residual excitation energy.</LI>
<LI>Major revision of the QGS final-state model: implemented Reggeon cascading
and "Fermi motion"; new algorithm for the determination of the kinematical
properties of partons; improved formation of the residual nucleus.
Improved cross-sections of K-meson - nucleon interactions in QGS: Pomeron
and two non-vacuum exchanges are taken into account.
Gamma-nucleon cross sections are also improved by tuning of parameters.
Improved string fragmentation (G4QGSMFragmentation): new algorithm for
last string decay a-la Lund is implemented; refined the algorithm to stop
the fragmentation.
Tuning the parameters related to both string formation and string
fragmentation to improve the description of thin-target data.</LI>
<LI>For both string models (FTF and QGS): first implementation of alpha
cluster structure of carbon nuclei (affecting only hadron-Carbon
interactions), according to P.Bozek et al., Phys.Rev.C90, 064902 (2014).</LI>
<LI>For Bertini Cascade, extended strange pair production channels, and
improved pion-quasideuteron reactions.</LI>
<LI>Extended upper limit of applicability (from 1 to <TT>100 TeV</TT>) for the
diffuse-elastic processes.</LI>
<LI>New class for diffuse-elastic process, where the underlying physics model
code has been revised and significantly improved in performance.</LI>
<LI>Introduced switch in cross-sections to allow transuranic elements.</LI>
<LI>Coherent use of the same parameterisation of level density and pairing
correction between all models in hadronic de-excitation.</LI>
</UL>
<B>Physics Lists</B>
<UL>
<LI>In all hadronic builders fixed use of data for neutron capture in HP
physics-lists.</LI>
<LI>Major revision of all ion builders.</LI>
<LI>New tuning of parameters for several EM constructors.</LI>
<LI>The physics lists QGSP_BIC_HP and QGSP_BIC_AllHP use EM Opt4
(instead of EM Opt0 as before).</LI>
<LI>For FTF/QGS/INCLXX physics-lists builders, use Barashenkov-Glauber-Gribov
inelastic cross-section for pions, avoiding to use Gheisha cross-sections
for pion inelastic on Hydrogen target.</LI>
</UL>
<B>Run</B>
<UL>
<LI>Added track/step/volume information in G4ExceptionHandler when an
exception happens while event is being processed.</LI>
</UL>
<B>Visualization and Interfaces</B>
<UL>
<LI>Implemented simple algorithm in visualization to allow colouring of
detector elements based on materials density.</LI>
<LI>RayTracer now visualizes layered-mass-geometry volumes defined in
parallel worlds.</LI>
<LI>G4UIcommand: additional implementation for adding
a mechanism of carrying back error message from the messenger
if a command is rejected due to the target class or the messenger.</LI>
</UL>
<B>Examples</B>
<UL>
<LI><I>B3aScoreWrgiter</I> and <I>B4dScoreWriter</I>: new examples (in
extended/analysis) demonstrating saving of scorers hits in form of
n-tuples in a Root file using Geant4 analysis tools. </LI>
<LI><I>DICOM2</I>: new example (in extended/medical) demonstrating how
to specialise and extend functionalities from DICOM example.</LI>
<LI><I>chem5</I>: new example (in extended/medical/dna), variation of
chem4, showing how to activate chemistry code and score the
radiochemical yield G, using preliminary <TT>G4EmDNAPhysics_option8</TT>
and <TT>G4EmDNAChemistry_option1</TT> constructors.</LI>
<LI><I>splitting</I>: new example (in extended/medical/dna) demonstrating
acceleration of Geant4-DNA physics simulations by particle splitting.</LI>
<LI><I>OpNovice2</I>: new example (in extended/optical) for investigation
of optical properties and parameters. Details of optical photon boundary
interactions on a surface are recorded, as well as the details of
optical photon generation and transport.</LI>
<LI><I>exMPI04</I>: new example in extended/parallel/MPI demonstrating
merging n-tuples via MPI. The <I>G4mpi </I> classes were adapted to
allow to define (an) extra worker(s) for collecting data
from processing workers (needed for merging n-tuples).</LI>
<LI><I>field01</I>: Added code to demonstrate use of G4PhysicsListHelper's
<TT>Use[Low|High]LooperThresholds()</TT> methods and code to demonstrate
fine grained control of G4[Coupled]Transportation parameters for (killing)
looping tracks (tracks which take too many iterations to finish integration;
typically due to electrons in vacuum, but also can affect other charged
particles).</LI>
</UL>
<A NAME="5."></A>
<HR>
<!-- ============================================== -->
<H2>5. Expected effects on physics and computing performance</H2>
<B>Electromagnetic physics</B>
<UL>
<LI>Extended G4SeltzerBergerModel model with optional sampling tables for
faster run time sampling of the emitted photon energy. Measuring 40-60%
model speedup.</LI>
</UL>
<B>Hadronic physics</B>
<UL>
<LI>Both hadronic string models (FTF and QGS) describe generally better the
thin-target data with respect to the previous versions.
Moreover, the QGS model becomes competitive with the FTF model, for
most thin-target observables for projectile kinetic energies above
about 20 GeV.</LI>
<LI>For the simulation of hadronic showers, both hadronic string models
give higher energy response with respect to the previous versions,
and consistently higher when compared with test-beam and collider data.
We believe at least part of this is due to an underestimate with the Birks
quenching factor which must be applied to the simulation in order to
compare to measured data. The Birks coefficient is a phenomenological
function of energy loss and not of energy deposit. It therefore neglects
the effect of discrete delta ray production performed in Geant4.
We therefore suggest re-fitting from data the Birks coefficient to the
h/e ratio: energy response of a hadron (charged pion or proton) at a
given energy with respect to the energy response of an electron of the
same energy.</LI>
</UL>
<P> </P>
<A NAME="6."></A>
<HR>
<!-- ============================================== -->
<H2>6. Known Run-Time Problems and Limitations</H2>
For a complete list of outstanding run-time problems and to submit any
problem you may find while running this version of Geant4, please refer
to the
<A TARGET="ext" HREF="http://cern.ch/geant4/problemreport">Geant4 Problem
Reporting System</A>.
<P> </P>
<A NAME="7."></A>
<HR>
<!-- ============================================== -->
<H2>7. Compilation Warnings</H2>
There may be a few compilation warnings on some platforms.
We do not believe that any of these lead to incorrect run-time behaviour.
<P> </P>
<A NAME="8."></A>
<HR>
<!-- ============================================== -->
<H2>8. Geant4 Software License</H2>
A Software License applies to the Geant4 code.
Users must accept this license in order to use it. The details and the list of
copyright holders is available at
<A href="http://cern.ch/geant4/license">http://cern.ch/geant4/license</A>
and also in the text file <TT>LICENSE</TT> distributed with the source code.
<P> </P>
<A NAME="9."></A>
<HR>
<!-- ============================================== -->
<H2>9. Detailed list of changes and fixes</H2>
These are the main new features/fixes included in this release since the
last patched public release (for more detailed lists of fixes/additions,
please refer to the corresponding History files provided in most packages):
<P> </P>
<A NAME="conf-notes">
<H3><I>Configuration</I></H3></A>
<UL>
<LI><U>CMake</U>
<UL>
<LI>Added c++17 to C++ standards against which Geant4 can be compiled.
Requires a minimum CMake version of 3.8 to enable.</LI>
<LI>Added "auto" dummy Thread Local Storage model; when "auto" is used,
no explicit compiler flag is being set to select the TLS model;
selection is left to compiler default.</LI>
<LI>Added configuration for optional profiling package TiMemory.</LI>
<LI>Skip validation checking for external zlib sources.</LI>
<LI>Added TESSELLATEDSOLID flag to enable optional wrapping of
G4TessellatedSolid with VecGeom implementation.</LI>
<LI>Added configuration support for MT on Windows; removed restrictions
and warnings on enabling MT on Windows.</LI>
<LI>Support use of WINDOWS_EXPORT_ALL_SYMBOLS for DLL builds with
cmake >=3.4 retaining old capability for CMake 3.3.</LI>
<LI>Updated FindInventor.cmake to properly setup SoQt if existing.</LI>
<LI>Geant4InterfaceOptions.cmake: added check for existance of SoQt package.</LI>
<LI>Geant4MakeRules_cxx.cmake: added -DGL_SILENCE_DEPRECATION to
CMAKE_CXX_FLAGS_INIT for clang to silence GL deprecation compilation
warnings on MacOS 10.14 Mojave.</LI>
<LI>Corrected URLs for data installation to use https; added -L flag to
curl to follow redirects when accessing a dataset URL.</LI>
<LI>Removed entry for G4NEUTRONXS data set, no longer necessary.</LI>
<LI>Updated data-sets versions.</LI>
</UL></LI>
<LI><U>GNUMake</U>
<UL>
<LI>In geomconf.gmk, added G4GEOM_USE_UTESSELLATEDSOLID flag to enable
VecGeom wrapper for G4TessellatedSolid.</LI>
<LI>Fixed setting of G4RUNPATHOPTION in Darwin/Linux-clang/g++ and
binmake.cmk. Addressing problem report
<A HREF="http://bugzilla-geant4.kek.jp/show_bug.cgi?id=2086">#2086</A>.</LI>
<LI>Cleanup settings for GL in Darwin-clang.gmk and Darwin-g++.gmk.
Added -DGL_SILENCE_DEPRECATION to CPPFLAGS to silence GL deprecation
compilation warnings on MacOS 10.14 Mojave.</LI>
</UL></LI>
</UL>
<A NAME="ana-notes">
<H3><I>Analysis</I></H3></A>
<UL><LI><U>Ntuples</U>
<UL>
<LI>Switched the default Root n-tuple merge mode to row-wise.</LI>
<LI>Added function <TT>SetNtupleRowWise(G4bool)</TT> to G4RootAnalysisManager, allowing
to choose row-wise/column-wise mode also in sequential mode.</LI>
<LI>Updated <TT>G4TNtupleManager::FinishTNtuple()</TT> to support creating n-tuples
both before and after <TT>OpenFile()</TT>.</LI>
<LI>Added friend class declarations to allow an implementation of MPI
n-tuple merging.</LI>
<LI>Allow overriding of selected functions in G4RootAnalysisManager,
G4RootNtupleManager and G4TNtupleManager, needed MPI ntuple merging.</LI>
</UL></LI>
<LI><U>Output files</U>
<UL>
<LI>Added "Boolean reset" argument in <TT>G4AnalysisManager::CloseFile()</TT>
function, allowing to close a file without resetting data.</LI>
</UL></LI>
<LI><U>Fixes</U>
<UL>
<LI>Fixed compilation warnings on gcc-8.1.0.</LI>
<LI>Fixed a deadlock issue with G4RootPNtupleManager where the mutex
was locked before being passed to Root.</LI>
<LI>Modified management of plotting: G4PlotManager is now instantiated
with G4AnalysisManager what makes possible to remove static
G4PlotParameters. Addressing problem report
<A HREF="http://bugzilla-geant4.kek.jp/show_bug.cgi?id=2083">#2083</A>.</LI>
<LI>Use the same mutex for add_row and end_fill. This should fix
occasional failures of tests in MT mode.</LI>
</UL></LI>
<LI><U>g4tools</U>
<UL>
<LI>Updated the g4tools version to g4tools-4.2.3.
See History_tools for the complete list of modifications.</LI>
</UL></LI>
</UL>
<A NAME="digits-notes">
<H3><I>Digitization & Hits</I></H3></A>
<UL>
<LI>Introduced a new interface class, G4VScoreNtupleWriter which allows
implementing in the user code automated storing of hits collections of
G4THitsMap (used by primitive scorers) with Geant4 analysis tools;
concrete implementations (G4ScoreNtupleWriter with associated messenger)
are demonstrated in new provided extended/analysis examples.</LI>
<LI>Updated G4THitsMap: generic <TT>GetContainer(), size(), GetIndex(...),
GetObject(...), begin(), end(), cbegin()</TT> and <TT>cend()</TT>.</LI>
<LI>Added G4THitsVector class, a generic container compatible with G4THitsMap
such that vectors can be used to score over the course of a run (saving
memory); it allows the underlying container to store data as non-pointer
types.</LI>
<LI>Updated use of thread-local static variables to function returning
a thread-local static reference. New model for allowing support of
multi-threading on Windows DLLs.</LI>
</UL>
<A NAME="emstd-notes">
<H3><I>Electromagnetic Processes</I></H3></A>
<UL>
<LI>Requires data-set G4EMLOW-7.7.</LI>
<LI><B>DNA</B>:
<UL>
<LI>Updated G4DNAOneStepThermalizationModel and other classes for
chemistry.</LI>
<LI>Updated use of thread-local static variables to function returning
a thread-local static reference. New model for allowing support of
multi-threading on Windows DLLs.</LI>
<LI>Included upper energy limit of models.</LI>
<LI>Updated Si mass in G4DNAGenericIonsManager.</LI>
<LI>Added protection in Auger emission.</LI>
<LI>Fixed header guards in G4ITFinder and G4AllITFinder.
Addressing problem report
<A HREF="http://bugzilla-geant4.kek.jp/show_bug.cgi?id=2084">#2084</A>.</LI>
</UL></LI>
<LI><B>High Energy</B>
<UL>
<LI>G4AnnihiToMuPair: added Coulomb correction to cross-section at
threshold.</LI>
<LI>Minor code cleanup in G4eeToHadronsMultiModel, G4eeToHadronsModel and
G4mplIonisationWithDeltaModel.</LI>
<LI>G4eeToHadrons, G4eeToHadronsMultiModel, G4hBremsstrahlung,
G4hPairProduction, G4hhIonisation, G4mplIonisation: automatic
documentation format now rST.</LI>
</UL></LI>
<LI><B>Low Energy</B>
<UL>
<LI>Added new classes for elastic scattering of gammas by M.Omer
and R.Hajima.</LI>
<LI>Updated JAEA elastic model classes.</LI>
<LI>G4LivermoreBremsstrahlungModel: no functional changes. Some base class
member names have been changed and these modifications propagated down
to the derived class.</LI>
<LI>Fixed problem G4PenelopeIonisationModel responsible for rare energy
non-conservation and run-time warning.</LI>
<LI>Fix in G4PenelopeComptonModel and G4PenelopePhotoElectricModel for rare
energy non-conservation.</LI>
<LI>Fix for differential cross-section of photo-electrons in
G4LivermorePolarizedPhotoElectricModel. Addressing problem report
<A HREF="http://bugzilla-geant4.kek.jp/show_bug.cgi?id=2089">#2089</A>.</LI>
<LI>G4RayleighScattering: changed automatic documentation format to rST.</LI>
</UL></LI>
<LI><B>Muons</B>
<UL>
<LI>G4MuMultipleScattering, G4MuPairProduction, G4MuIonisation,
G4ePairProduction, G4MuBremsstrahlungModel, G4MuBremsstrahlung,
G4MuBetheBlochModel: use default destructor.</LI>
<LI>Added minor protection in G4MuPairProductionModel.</LI>
<LI>G4ePairProduction: fixed model name.</LI>
<LI>G4EnergyLossForExtrapolator: fixed MT initialisation and destruction.</LI>
<LI>G4MuMultipleScattering, G4MuPairProduction, G4eePairProduction,
G4MuBremsstrahlung, G4MuIonisation: automatic documentation format
now rST.</LI>
</UL></LI>
<LI><B>Polarisation</B>
<UL>
<LI>G4PolarizedAnnihilationModel: implemented sampling of final state
at-rest.</LI>
<LI>Code cleanup; removed unused headers and methods.</LI>
</UL></LI>
<LI><B>Standard</B>
<UL>
<LI>G4GoudsmitSaundersonMscModel: implemented interface method for
macroscopic first transport cross-section computation per volume
(used only for testing).</LI>
<LI>G4GSMottCorrection, G4GSPWACorrections: fixed initialization of first
moment correction factor.</LI>
<LI>G4GoudsmitSaundersonTable: fixed gcc-8.1.0 compilation warnings.</LI>
<LI>G4WentzelOKandVIxSection: fixed form-factor for proton.
Added Mott corrections to e+-.</LI>
<LI>G4WentzelOKandVIxSection: fixed destructor.</LI>
<LI>G4ScreeningMottCrossSection: added extra method for more CPU effective
computation of the Mott correction.</LI>
<LI>G4UrbanMscModel: correction in order to get better backscattering for
low energy.</LI>
<LI>G4eSingleCoulombScatteringModel: fixed calculation of scattering angle.
Addressing problem report
<A HREF="http://bugzilla-geant4.kek.jp/show_bug.cgi?id=2090">#2090</A>.</LI>
<LI>G4ScreeningMottCrossSection: use G4MottData and improved time consuming
computations. Removed obsolete G4MottCoefficients class, replaced by
static data class G4MottData.</LI>
<LI>G4ScreeningMottCrossSection, G4eSingleCoulombScatteringModel: added
extra protections on argument of std::acos avoiding crash due to
precision lost; code cleanup.</LI>
<LI>G4BetheBlochModel: use form-factor for projectile ion as in formulation
of NIM A 488 (2002) 282.</LI>
<LI>G4ionIonisation: apply correction to energy loss only to BetheBloch
ionisation model.</LI>
<LI>G4BraggModel, G4BraggIonModel, G4BetheBlochModel: added optional
possibility to use ICRU90 data.</LI>
<LI>G4PSTARStopping: minor code cleanup.</LI>
<LI>G4LindhardSorensenIonModel, G4LindhardSorensenData: new ion ionisation
model for moderate and high energies.</LI>
<LI>G4AtimaEnergyLossModel, G4AtimaFluctuations: new ion ionisation model.</LI>
<LI>G4BetheHeitler5DModel: new 5D model.</LI>
<LI>G4BetheHeitlerModel, G4PairProductionRelModel: new versions with
improved screening function approximation, improved LPM function
approximation, efficiency, documentation and cleanup.
Corrected call to selecting target atom in final state sampling.</LI>
<LI>G4PairProductionRelModel, G4BetheHeitlerModel: use G4ModifiedTsai
angular generator for the final state sampling.</LI>
<LI>G4PairProductionRelModel: fixed computation of cross-section in the
new version of the high energy model.</LI>
<LI>G4DipBustGenerator, G4ModifiedTsai: implemented new interface for
sampling of directions of e+e- pair.</LI>
<LI>G4eplusAnnihilation: moved sampling of final state from AtRestDoIt()
method to <TT>SamplingSecondaries()</TT>.</LI>
<LI>Code cleanup in G4eeToTwoGammaModel, G4BetheHeitlerModel and
G4NISTStoppingData.</LI>
<LI>G4eplusAnnihilation: set weights for directional splitting.</LI>
<LI>G4eplusTo2GammaOKVIModel, G4eplusTo3GammaOKVIModel: make consistent
definition of cross-section for 2-gamma and 3-gamma models; use cuts
for gamma in sampling of 3-gamma final state.</LI>
<LI>Added new models, G4eplusTo2GammaOKVIModel and G4eplusTo3GammaOKVIModel.</LI>
<LI>G4eBremsstrahlungRelModel, G4PairProductionRelModel, G4PAIySection,
G4PAIxSection, G4ICRU73QOModel: moved inline method which use static
constants to source. Improved code formatting.</LI>
<LI>G4SeltzerBergerModel: extended model with optional sampling tables for
fast run time sampling of the emitted photon energy from the
Seltzer-Berger scaled DCS, instead of using the rejection.
The sampling tables are hadled by the new G4SBBremTable class.
Measuring 40-60% model speedup with some additional memory consumption.</LI>
<LI>G4eBremsstrahlungRelModel: improved LPM function approximation; code cleanup
and optimization.</LI>
<LI>G4SeltzerBergerModel, G4eBremsstrahlungRelModel: use by default
G4ModifiedTsai generator instead of G4DipBustGenerator.</LI>
<LI>G4DipBustGenerator: optimisation, speedup; using G4Pow::A13.</LI>
<LI>G4GammaConversion, G4eBremsstrahlung, G4hIonisation: use upper energy
limit for low-energy model from physics-list instead of hard-coded
values.</LI>
<LI>G4CoulombScattering, G4eBremsstrahlung, G4[e,h]MultipleScattering,
G4eplusAnnihilation, G4[e,h,ion,alpha]Ionisation, G4ComptonScattering,
G4NuclearStopping, G4PhotoElectricEffect: changed auto-generated
documentation format to rST.</LI>
</UL></LI>
<LI><B>Utils</B>
<UL>
<LI>G4EmCorrections, G4VEmModel, G4ionEffectiveCharge, G4VAtomDeexcitation:
avoid use of static class members in inline methods; moved long methods
to source; added thread safe definition of model index for atomic
de-excitation.</LI>
<LI>G4EmCalculator: fixed <TT>GetCrossSectionPerVolume()</TT> method in case of
multiple scattering process; the proper value of the first transport
cross-section is retrieved now.</LI>
<LI>G4EmCalculator: fixed computation dEdx for ions.</LI>
<LI>G4EmParameters, G4EmParametersMessenger: added 5D gamma conversion
model parameters; added <TT>MscEnergyLimit()</TT> and corresponding UI command;
addressing problem report
<A HREF="http://bugzilla-geant4.kek.jp/show_bug.cgi?id=2045">#2045</A>;
added extra DNA UI directory and command; improved dump of EM
parameters.</LI>
<LI>G4EmParameters: added extra protection to avoid simulteneous definition
of options which should not be enabled together.</LI>
<LI>G4EmParameters, G4EmParameterMessenger, G4LossTableManager:
minor update of interfaces; added extra parameters.</LI>
<LI>G4EmParameters, G4EmParametersMessenger: added extra parameter and
UI command to enable usage of ICRU90 data; fixed DNA parameters;
added checks of EM physics type.</LI>
<LI>G4EmParameters: fixed typo in printout.</LI>
<LI>G4VEmAngularDistribution: added <TT>SamplePairDirections()</TT> method for
sampling of e+e- pair production.</LI>
<LI>Added new enumerator G4DNAModelSubType.</LI>
<LI>G4LossTableManager: moved inline run time method to source.</LI>
<LI>Minor code cleanup in G4VMscModel and G4VEmModel.</LI>
<LI>G4EmDataHandler: added public interfaces.</LI>
<LI>G4VEmProcess: code cleanup in <TT>StorePhysicsTable(..)</TT> methods;
introduced 'isTheMaster' flag.</LI>
<LI>G4EmElementSelector: code optimisation.</LI>
<LI>G4EmModelManager, G4VEmModel, G4VEmProcess, G4VEnergyLossProcess,
G4VMultipleScattering: auto-generated documentation now in rST format.</LI>
<LI>G4EmBiasingManager, G4EmParameters, G4EmParametersMessenger,
G4VEnergyLossProcess, G4VEmProcess: enable directional splitting
and moved logic to G4EmBiasingManager.</LI>
<LI>Fixed Coverity defects.</LI>
</UL>
<LI><B>Xrays</B>
<UL>
<LI>Updated use of thread-local static variables to function returning
a thread-local static reference. New model for allowing support of
multi-threading on Windows DLLs.</LI>
<LI>G4Cerenkov. added protection against infinite loops due to very small
steps. Addressing problem report
<A HREF="http://bugzilla-geant4.kek.jp/show_bug.cgi?id=1992">#1992</A>.</LI>
<LI>G4Cerenkov, G4Scintillation: correctly reset number of photons to
zero each step. Addressing problem report
<A HREF="http://bugzilla-geant4.kek.jp/show_bug.cgi?id=2056">#2056</A>.</LI>
</UL></LI>
</UL>
<A NAME="event-notes">
<H3><I>Event</I></H3></A>
<UL>
<LI>Code cleanup in G4StackManager to remove branches for muonic atom
settings.</LI>
<LI>Updated use of thread-local static variables to function returning
a thread-local static reference. New model for allowing support of
multi-threading on Windows DLLs.</LI>
</UL>
<A NAME="externals-notes">
<H3><I>Externals</I></H3></A>
<UL>
<LI><U>CLHEP</U>:
<UL>
<LI>Updated to CLHEP version 2.4.1.0.</LI>
<LI>Enabled thread-local storage and atomics on Windows and clang,
as now all supported by c++11.</LI>
<LI>Ranlux64Engine: fixed compilation warning on Windows for shift count
negative or too big.</LI>
<LI>Fixed compilation warnings from gcc-8.1.0 when using
-Woverloaded-virtual compilation option.</LI>
</UL></LI>
<LI><U>Zlib</U>:
<UL>
<LI>Updated to Zlib version 1.2.11</LI>
<LI>Use original files format and adopt C Compiler for installation.
Addressing compilation warnings on gcc-8.1.0.</LI>
</UL></LI>
</UL>
<A NAME="gen-notes">
<H3><I>General Processes</I></H3></A>
<UL>
<LI><B>Biasing</B>
<UL>
<LI>Updated use of thread-local static variables to function returning
a thread-local static reference. New model for allowing support of
multi-threading on Windows DLLs.</LI>
</UL></LI>
<LI><B>Cuts</B>
<UL>
<LI>Code cleanup; use of nullptr and c++11.</LI>
</UL></LI>
<LI><B>Management</B>
<UL>
<LI>Code cleanup; use of nullptr and c++11.</LI>
</UL></LI>
<LI><B>Optical</B>
<UL>
<LI>G4OpMieHG: changed <TT>GetProperty()</TT> to use enum indeces instead of
strings.</LI>
</UL></LI>
<LI><B>Scoring</B>
<UL>
<LI>Updated use of thread-local static variables to function returning
a thread-local static reference. New model for allowing support of
multi-threading on Windows DLLs.</LI>
<LI>G4ParallelWorldProcess: fixed track velocity of optical photons for
the first step in a layered mass geometry.</LI>
</UL></LI>
<LI><B>Transportation</B>
<UL>
<LI>G4Transportation and G4CoupledTransportation: adjusted killing of
looping particles, to apply only to stable particles. Unstable
particles will be propagated indefinitely. At object destruction,
added printing of the maximum energy and sum of energies of particles
"saved". Added method <TT>SetHighLooperThresholds()</TT> to restore
old default parameters for killing looping particles (warning energy
to 100 MeV - tracks below this are killed silently; important energy
to 250 MeV - above this, tracks are given multiple chances). Default
constructors now use these high values.
New auxiliary methods <TT>PushThresholdsToLogger()</TT> and
<TT>ReportMissingLogger()</TT>.
New method <TT>SetLowLooperThresholds()</TT> for low-E applications.
New method <TT>ReportLooperThresholds()</TT> to print out the values.
Added extra simple statistics at destruction: stats on non-electrons
loopers killed (sum E, number, max E & its PDG); stats for non-electron
loopers killed (total energy, maximum energy and PDG of the type of
that track). Statistics are now printed by a new method and are
printed also in non-verbose mode.
Addressing problem report
<A HREF="http://bugzilla-geant4.kek.jp/show_bug.cgi?id=2063">#2063</A>.</LI>
<LI>New class G4TransportationLogger to consolidate reports from the
Transportation processes.</LI>
<LI>Use G4Exception for warnings, instead of streaming to <TT>cout/cerr</TT>.
<LI>Fixed Coverity defects.</LI>
</UL></LI>
</UL>
<A NAME="geo-notes">
<H3><I>Geometry</I></H3></A>
<UL>
<LI><B>Magnetic field</B>
<UL>
<LI>Correction in G4MagneticField to no longer inherit from
G4ElectroMagneticField; this allowed a user to provide a full EM field,
but obtain the force only from the B-field when using the equation
created in the G4ChordFinder constructor.
It is no longer possible to use the equation of motion for a pure
magnetic field for the case of a mixed electromagnetic field. Instead
the full chain of classes for equation of full EM field, steppers and
driver must be used for the case of the 'full' EM field.</LI>
<LI>Introduced Bulirsch-Stoer method of integration, an alternative to
Runge-Kutta based on the mid-point method. A separate driver derived
from G4VIntegrationDriver was needed, using specialisation of
G4IntegrationDriver template class.</LI>
<LI>Change in <TT>G4FieldManager::CreateChordFinder()</TT> to cope better
with case where field argument is null.</LI>
<LI>Fix for missing interface method in G4VIntegrationDriver; added method
<TT>RenewStepperAndAdjust()</TT> which used to exist in G4MagInt_Driver. This
method must be defined in those derived driver classes which accept
a Runge Kutta stepper.</LI>
<LI>Defined default implementation of the method in G4IntegrationDriver
and G4FSALIntegrationDriver, and implementations in the templated
classes G4IntegrationDriver and G4FSALIntegrationDriver.</LI>
<LI>Removed static data from G4FSALIntegrationDriver, G4IntegrationDriver
and G4MagIntegratorDriver and moved to base class as normal data
members. Fix was required for allowing proper symbol exporting on
Windows DLLs.</LI>
<LI>Use G4Exception warnings instead of streaming to cerr.</LI>
<LI>Added override final for virtual methods in G4UniformMagField.</LI>
<LI>Corrections in calculation of chord distance in G4DormandPrince745
(use 4th order interpolation from Shampine 1986 for mid-point).</LI>
<LI>Revised G4FieldUtils: moved <TT>copy()</TT> method from steppers,
added two new setters for simple & variadic methods.</LI>
<LI>Fixes to copy input time in ModifiedMidpoint, CashKarp, G4RK547FEq1/2/3
Bogacki-Shampine 4/5 and 2/3, Dormand Prince 4/5, 5/6 and 7/8 and
Tsitouras steppers.</LI>
<LI>Fix for unitialised pointer in G4FSALIntegrationDriver constructor.
Revision of data member variable names.</LI>
<LI>G4NystromRK4 now uses G4CachedMagneticField for caching values and
creates it if not provided. Any other stepper will work with that cached
field too. </LI>
<LI>Removed <TT>G4MagIntegratorStepper::ComputeRightHandSide()</TT>
virtual method, previously introduced for use with G4NystromRK4
(calling non-virtual <TT>RightHandSide()</TT> instead).</LI>
<LI>New experimental class G4BFieldIntegrationDriver for pure magnetic
field.
In <TT>QuickAdvance()</TT> it uses helix stepper when rotation angle
is larger than <TT>pi/3</TT>.</LI>
<LI>New experimental class G4InterpolationDriver; using Runge-Kutta method
with interpolation property to integrate EoM with error control.</LI>
</UL></LI>
<LI><B>Management</B>
<UL>
<LI>G4AffineTransform: added explicit copy constructor, assignment and
destructor; <TT>InverseProduct()</TT> has been optimised, now it does
36 multiplications instead of 45; optimised also for the case of pure
translation; added methods for inverse transformation:
<TT>InverseTransformPoint(), InverseTransformAxis(),
InverseNetRotation()</TT> and <TT>InverseNetTranslation()</TT>.
Minor optimisation to multiplication of matrix by vector, to make use
of temporaries in accessing vector components.</LI>
<LI>G4GeomSplitter: replaced use or <TT>realloc()/free()</TT> and
<TT>memcpy()</TT> with normal allocation/deallocation through
G4Allocator. Also addressing compilation warnings on gcc-8.1.0.</LI>
<LI>G4VSolid: reimplemented <TT>EstimateSurfaceArea()</TT> for approximate
calculation of the surface area of a solid, by estimating volume of a
thin shell surrounding the surface using Monte-Carlo method; algorithm
providing more accurate estimation without performance penalty.</LI>
</UL></LI>
<LI><B>Navigation</B>
<UL>
<LI><TT>G4Navigator::ComputeLocalAxis()</TT>: removed unnecessary check
for <TT>IsRotated()</TT>.</LI>
<LI>Replaced use of <TT>Inverse()</TT> from G4AffineTransform with new
dedicated methods, to avoid creation of temporaries in
G4RegularNavigation, G4Navigator and G4ReplicaNavigation.</LI>