Geant4 9.0 Release Notes

June 29th, 2007

The code and binary libraries for the supported systems are available through our Source Code Web page.

Please refer to the Geant4 User Documentation for further information about using Geant4.

Contents

1. Supported and Tested Platforms
2. CLHEP and AIDA
3. Items for migration of the user code
4. New Developments and Capabilities
5. Expected effects on physics and performance
6. Known Run-Time Problems and Limitations
7. Compilation Warnings
8. Known Run-Time Warnings
9. Geant4 Software License
10. Detailed list of changes and fixes

1. Supported and Tested Platforms

• SUN Solaris 5.8, C++ CC-5.5.
• Linux, gcc 3.4.6.
  This configuration was tested on 32 bits architectures (Intel) and 64 bits architectures (AMD) with the Scientific Linux CERN 3 (SLC3) distribution (based on RedHat Linux Enterprise 3) and Scientific Linux CERN 4 (SLC4) distribution (based on RedHat Linux Enterprise 4). Versions of Geant4 have also been compiled successfully on other Linux distributions, like Debian, Suse or other RedHat systems.
• Windows/XP and CygWin Tools with: Visual C++ 8.0 (Visual Studio 2005)

• SUN Solaris 5.8, C++ CC-5.4 Patch 111715-02
• Linux, gcc-3.2.3, gcc-4.2.0
• Linux, Intel-icc 9.1
• MacOS 10.4, gcc-4.0.1
• Windows/XP and CygWin Tools with: Visual C++ 7.1 .NET

• AIX 4.3.2, xlC 6.0
• DEC V4.0, cxx C++ V6.1-027
• HP 10.20, aCC C++ B3910B A.01.23
• SGI V6.5.5, CC 7.2.1

2. CLHEP and AIDA

3. Items for migration of the user code

• When defining a parallel geometry for importance biasing, the parallel geometry now must inherit from G4VUserParallelWorld. This has a GetWorld() method which creates a (named) copy of the mass geometry world volume and later retrieves it.
• The different parallel and mass sampler classes have been combined into one G4GeometrySampler class which requires the following instantiation, e.g.: mgs(detector->GetWorldVolume(),"neutron");
  An additional set method defines whether or not the sampler is in a parallel geometry: mgs.SetParallel(false);
• When a parallel world is created, then parallel navigation must be activated. An application with physics list that uses the AddTransportation method will automatically use G4CoupledTransportation which is picked up by the creation of the parallel world copy.
• All scoring can be implemented only through the Primitive Scorer classes. Users utilising the older scoring classes will need to migrate to using Primitive Scorers. The same functionality is provided apart from G4ScoreTable, but scorers are now attached to logical volumes (in place of geometry cells).. This requires explicit copy numbers to be utilised for the same logical volume and the GeometryCell methods should be accessed through the physical volume and replica number method.
• A demonstration of the new biasing and scoring is available in examples/extended/biasing/B01 and B02.

A few methods of the interface classes G4VEmModel, G4VEnergyLossProcess, G4VEmProcess, G4VMultipleScattering were revised, in order to reduce memory allocations. All classes inheriting these interfaces were modified accordingly (see detailed list of changes). Advanced users with derived classes must make similar changes.

Old implementations of physics processes, of configurations frozen in release 5.2, have been removed (see below detailed list of changes).

• Minimal - equivalent to the original algorithm, as implemented in Geant4 7.1 and earlier releases.
• UseSafety - the current default, which makes use of the geometrical safety
• UseDistanceToBoundary - the most advanced, recommended for accuracy in the cases where no magnetic field is set.

• Isotope production counting flag is now set to OFF by default. If desired, user code must switch it on using the method G4HadronicProcess::EnableIsotopeCounting(). Isotope production models allow users to specify their own isotope production mechanism and data set, or use G4NDL to get the production cross sections. If no model is specified, the largest final state nuclear fragment becomes the produced isotope. If enabled, isotope production will occur for any hadronic model. Only users who have specifically assigned an isotope production model in their physics lists will be affected by tghe above change.
• The old variable NeutronHPCrossSections defining the path where to load the G4NDL data set has been renamed to G4NEUTRONHPDATA, so that this variable shares the G4 prefix used for other environment variables; therefore the user environment must be changed accordingly.

The components for electro-magnetic physics have been renamed and three variants are now available:

• G4EmStandardPhysics - the default
• G4EmStandardPhysics_option1 - providing similar physics to that implemented in release 7.1
• G4EmStandardPhysics_option2 - using sub-cutoff options, which enable lower production thresholds for delta-electrons near geometry boundary.

• New data for neutron cross-sections, G4NDL.3.11.
• New low-energy data set, G4EMLOW.4.3.
• Fix in revised data set for radioactive-decay, G4RadioactiveDecay.3.2.

4. New Developments and Capabilities

• The first implementation of error progration for tracks is provided. This capability is new to Geant4, and is a new implementation of GEANE capabilities. Users are invited to provide feedback.

• The structure of UI commands was changed: a new directory /process/msc contains all commands for multiple scattering. New UI commands are added to control the parameters of the Multiple Scattering process:
  • /process/msc/Skin
  • /process/msc/RangeFactor
  • /process/msc/GeomFactor
  • /process/msc/StepLimit
  • /process/msc/LateralDisplacement.
• The interface to G4EmProcessOptions is modified to be consistent with this change.
• As a result, the possibility to change the values of parameters between runs is now extended to all parameters which can be defined via the UI.

• When using the high precision neutron module, a user may now choose to force the use of the Photon Evaporation model, instead of the neutron capture final state data. The environment variable G4NEUTRONHP_USE_ONLY_PHOTONEVAPORATION must be set to do this. For cases in which the final state data does not exist in the G4NDL data library, this model was already used and thus there is no change.
• A user can force high precison neutron code to use only exact isotope data files, instead of allowing nearby isotope files to be used. In this case if the exact file is not available, the cross section will be set to zero and a warning message will be printed. To enable this, set the environment variable G4NEUTRONHP_SKIP_MISSING_ISOTOPES.
• A user can switch off the Doppler broadening due to the thermal motion of the target nucleus. The broadening is used to correct the G4NDL data, which are at zero temperature. This must be done only for appropriate applications. To switch it broadening off, the environment variable G4NEUTRONHP_NEGLECT_DOPPLER must be set. Then the Doppler broadening in the cross section calculation is neglected. However, the thermal motion of the nucleus is still taken into account in the Final State Generator. This option provides a significant CPU performance advantage. It is not suitable, though, for simulation in which low energy (thermal) neutrons are important, as related observables are affected.

5. Expected effects on physics and performance

• Results for calorimetry are expected to be unchanged within 0.5%. The width of energy deposition in thin layers (vertex detectors) should be more close to data - current accuracy is few percent.
• In simple setups a speedup of simulation of EM showers of about 10% is observed for the 'default' EM physics configuration (used in most lists, including QGSP) and about 5% for the faster 'optional' physics configuration (used in EMV lists, such as QGSP_EMV).

• The FTF model is now used in some physics lists down to 5 GeV.
• Bug fixes in LEP models remove an un-physical peak at 180 degrees and reduce the number of very low energy nucleons produced. The net effect is to shift the angular distributions forward by about 5-10 degrees and increase the mean secondary kinetic energy by roughly 20 - 30 MeV. Effects of this size occur at incident particle energies of 2 - 10 GeV. At higher energies the effect is small. Benchmark tests show little change in shower shape.

6. Known Run-Time Problems and Limitations

7. Compilation Warnings

8. Known Run-Time Warnings

    G4Transportation: Statistics for looping particles 
       Sum of energy of loopers killed: 
       Max energy of loopers killed: 

9. Geant4 Software License

10. Detailed list of changes and fixes

• Added modules biasing, error_propagation and lll_fission to configuration.
• Removed G4LIB_BUILD_LISTS flag. Physics lists module is built by default.
• Updated Configure script to detect new data sets and to remove search for old physics_list module in installation of libraries.

• Introducing new scorers for event biasing options.
• Removed G4CellStoreScorer class associated with old biasing/scoring scheme in detector module.

• In all processes inheriting from G4VEnergyLossProcess or G4VEmProcess, removed the method SecondaryPostStep().
• Modified interface of SampleSecondary() method in all models inheriting from G4VEmModel base class.
• High Energy
  • G4mplIonisation: fixed destruction of monopole ionisation model.
• Low Energy
  • Removed obsolete class G4LowEnergyBraggModel.
• Muons
  • G4EnergyLossForExtrapolator: added Set methods for first and last bins in tables. Default maximum value is set to 10 TeV on CMS request.
  • G4MuPairProductionModel: fixed usage of cut value inside SelectRandomAtom(), providing correct sampling of atom of mixture for production of e+e- pair.
  • New class G4ErrorEnergyLoss for error-propagation.
  • Removed obsolete classes G4MuIonisation52, G4MuPairProduction52, G4MuBremsstrahlung52 and G4VMuEnergyLoss.
• Polarisation
  • G4PolarizedCompton, G4ePolarizedIonisation, G4eplusPolarizedAnnihilation: added method PostStepGetPhysicalInteractionLength() in order to provide simulation of polarized target. Addition required following the design change in utils module.
  • G4PolarizedCompton: fixed compilation warning and incorrect usage of G4PhysicsVector.
• Standard
  • G4ionIonisation:
    • Added InitialiseMassCharge() method; removed GetMeanFreePath() implementation.
    • Added NuclearStopping flag.
    • Added stopping data only for particle type GenericIon.
  • G4MultipleScattering, G4hMultipleScattering:
    • Reorganized methods for setting parameters of G4UrbanMscModel providing a possibility to change parameters between runs.
  • G4UrbanMscModel:
    • Reorganized initialisation and ComputeTruePathLengthLimit() method using G4MscStepLimitType enumerator to choose step limit algorithm.
    • Optimisation of step limitation for skin>0 (about 5% CPU save).
    • Use G4SafetyHelper for computation of safety and distance to the boundary; removed pointer to G4Navigator.
  • G4UniversalFluctuation: correction of the algorithm in order to have better agreement between sampled width and data - precision is few percents.
  • G4eBremsstrahlung: removed definition of gamma threshold from constructor, define this parameter at initialisation of the process from G4LossTableManager - allow to use UI command.
  • Use fParticleChange in G4eeToTwoGammaModel::SampleSecondaries().
  • Removed obsolete classes G4VeEnergyLoss, G4VhEnergyLoss, G4GammaConversion52, G4PhotoElectricEffect52, G4eBremsstrahlung52, G4eIonisation52, G4eplusAnnihilation52, G4hIonisation52 and G4MultipleScattering52.
• Utils
  • G4VEmModel: changed interface of SampleSecondary() method, return type becomes void, added extra parameter std::vector<G4DynamicParticle*>*. All classes using or inheriting this interface are modified. About 5% speedup of EM shower simulation observed. G4VEmProcess and G4VMultipleScattering modified to follow new interface.
  • G4VEnergyLossProcess, G4VEmProcess:
    • Removed method SecondaryPostStep(); use a direct call to selected model instead.
    • Speedup code by adding implementation method PostStepGetPhysicalInteractionLength().
    • Cleanup comments and reorganized code for easier reading; moved short methods to inline.
  • G4VEnergyLossProcess:
    • Added methods InitialiseStep() and InitialiseMassCharge() to define mass and charge of ions at each step.
    • Added implementation method AlongStepGetPhysicalInteractionLength() in order to speedup computations.
    • Remove pointer to G4Navigator and use G4SafetyHelper.
    • Added proper initialisation of preStepKinEnergy.
  • G4VMultipleScattering:
    • Speedup code by adding implementation of methods AlongStepGetPhysicalInteractionLength() and PostStepGetPhysicalInteractionLength().
    • Cleanup comments and reorganized code for easier reading; moved short methods to inline; added method to set multiple-scattering parameters.
  • G4EnergyLossMessenger:
    • Defined new UI sub-directory /process/msc and reorganized UI commands for multiple-scattering processes. Using G4EmProcessOptions as an interface to all EM processes.
  • G4LossTableManager: removed setting of multiple-scattering parameters and fixed deletion of G4EmCorrections class in destructor.
  • G4EmProcessOptions: updated interface to multiple-scattering processes, added extra methods to set multiple-scattering parameters.
  • Added class G4MscStepLimitType.

• New module for backwards error propagation of tracks (Geant4e).

• Fixed normalisation of source intensities with multiple sources in G4GeneralParticleSource::IntensityNormalization().
• Removed classes using old biasing/scoring scheme: G4ScoreTable, G4Scorer.

• Biasing
  • New module for process-based biasing.
  • Migrated classes from processes/scoring and transportation modules.
• Cuts
  • Fix in G4VRangeToEnergyConverter::convert(), range-energy conversion, correcting artifact observed in the visible energy plot over range-cut, where a jump in the distribution was evident. The correction can influence the result only for the tests, where the e+/e- energy cuts are below 30 keV and even in this case the change in results should be small.
  • Fix in G4ProductionCutsTable::CheckMaterialCutsCoupleInfo().
• Management
  • Added PreparePhysicsTable() method in G4WrapperProcess.
• Optical
  • Use call to G4GeometryTolerance instead of kCarTolerance constant in G4OpBoundaryProcess.
• Parameterisation
  • Removed classes implementing old interface to fast-simulation manager process and setting of parallel world: G4FastSimulationManagerProcess81, G4FlavoredParallelWorld and G4GFSManager81.
• Scoring
  • Moved biasing classes and tests into processes/biasing module.
  • Fixed bug in G4ScoreProcess which didn't get applied to non-parallel geometries.
• Transportation
  • G4CoupledTransportation: make treatment of mass and full safety coherent. Added storing of full safety in state; retrieve it after compute step from G4PathFinder; calculate them same way at start of step.
  • Use call to G4GeometryTolerance instead of kCarTolerance constant in G4VTrackTerminator.
  • Archived classes for old biasing scheme, now replaced in module processes/biasing. Removed classes: G4MImportanceConfigurator, G4MScoreConfigurator, G4MScoreProcess, G4MWeightWindowConfigurator, G4MassGCellFinder, G4MassGeometrySampler, G4MassImportanceProcess, G4MassWeightWindowProcess, G4PImportanceConfigurator, G4PScoreConfigurator, G4PScoreProcess, G4PWeightWindowConfigurator, G4ParallelGCellFinder, G4ParallelGeometrySampler, G4ParallelImportanceProcess, G4ParallelTransport, G4ParallelTransportConfigurator, G4ParallelWWnTransportProcess, G4ParallelWeightWindowProcess, G4PlaceOfAction, G4ProcessPlacer, G4SamplingPostStepAction, G4VProcessPlacer, G4VSampler, G4VSamplerConfigurator, G4WeightCutOffConfigurator.

• Use call to G4GeometryTolerance instead of kCarTolerance, kAngTolerance and kRadTolerance constants where necessary.
• Biasing
  • Removed classes from old biasing scheme. These are now replaced by either G4CoupledTransportation (parallel navigation) or in processes/biasing module. In addition the scoring is implemented with the new scoring framework (in digits_hits and processes/scoring modules). The following classes are removed: G4ParallelNavigator, G4ParallelStepper, G4ParallelWorld, G4VParallelStepper, G4VPGeoDriver, G4VScorer, G4ImportanceSplitExaminer, G4VImportanceSplitExaminer, G4VWeightWindowExaminer, G4WeightWindowExaminer.
• Magnetic field
  • New class G4HelixMixedStepper, mixing helical stepper and Runge-Kutta G4MagHelicalStepper.
  • New calculation of DistChord() for SteppingAngle>pi in G4RKG3Stepper.
  • Relaxed fatal exception to warning in G4PropagatorInField::AccurateAdvance() for case of step zero and return previous value. This fixes problem observed by ATLAS in release 8.3 when using very conservative values for field parameters.
  • Introduced new class G4ErrorMag_UsualEqRhs used to reverse the magnetic field when propagation is backwards for error propagation.
• Management
  • Introduced new classes for handling error propagation: G4ErrorTarget, G4ErrorSurfaceTarget, G4ErrorTanPlaneTarget, G4ErrorPlaneSurfaceTarget and G4ErrorCylSurfaceTarget.
  • G4Region may have assigned a regional user stepping action. This local action is invoked after the global user stepping action which is set to G4RunManager.
• Navigation
  • G4SafetyHelper can now be accessed via G4TransportationManager. Depending on type of navigation (single world or multiple navigation) it uses mass navigator or G4PathFinder. All calls from EM processes to geometry are provided by G4SafetyHelper. Initialisation of the class is done in G4TransportationManager and G4PathFinder.
  • G4Navigator: added new method CheckNextStep(), which makes a trial step, without changing the state of the navigator. It's a variant of ComputeStep(), whose signature it shares.
  • Fixed bug in G4GeomTestSegment in case of point on the surface, reducing cases of reported spurious overlaps.
  • G4PropagatorInField: modified ReEstimateNewPoint() in order to handle cases with zero step.
  • Introduced new class G4ErrorPropagationNavigator, a specialised navigator for performing double navigation in the detector geometry and on the target surface for error propagation.
  • Added method CheckPointExiting() in G4AuxiliaryNavServices utility class.
• Solids (CSG)
  • G4Tubs, G4Cons: fixed bug for DistanceToOut(p,v) for phi section. The bug was causing segmentation fault in recursive_test (problem report #784).
• Solids (Specific)
  • Fixed default response of DistToIn/DistToOut for twisted solids.

• Added G4GeometryTolerance singleton class for the definition of tunable geometrical tolerances. The Cartesian tolerance can either be set to a fixed value (kCarTolerance=1E-9 mm) or to a value computed on the basis of the maximum extent of the world volume assigned through the G4GeometryManager at the beginning of the application and -before- any geometrical object is created.
• Old static constants kCarTolerance, kRadTolerance and kAngTolerance have been removed.
• Added new class G4SimplexDownhill in HEPNumerics module: a Simplex method for function minimization.
• Introduced new utility class G4ErrorPropagatorData to provide access to mode, state, target and manager verbosity for the error propagation classes.
• Fixed bug in G4LPhysicsVector constructor: initialise vectors of length=nbin+1 (one bin is hidden for users). Addresses problem report #942.
• Changed date for release 9.0.

• G4AttDefStore: now comprises utility functions in namespace.
• G4AttHolder: Made copy constructor and assignment private to enforce proper usage. Replaced forward declarations in header with explicit inclusions.

• Cross sections
  • G4GlauberGribovCrossSection: added new methods GetRatioSD() and GetRatioQE() allowing for retrieving qe/in and sd/in ratios.
  • G4HadronNucleonXsc: added new method GetHadronNucleonXscMK() to get hadron-nucleon cross section.
  • Fix for non-initialised total cross section for pi- in G4PiNuclearCrossSection.
  • Fixed memory leak at initialisation in classes G4ElectroNuclearCrossSection and G4PhotoNuclearCrossSection.
  • G4HadronCrossSections: replaced PSIZE with NPARTS to avoid name conflict with SDK Windows environment.
• Management
  • G4HadronicProcess:
    • In method FillTotalResult() partly remove spurious peak at 90 degrees by using different G4DynamicParticle constructor and prohibiting kinetic energy from being exactly zero.
    • Changed default for isotope production to off to fix a memory leak. Fixed memory leak at initialisation. Removed effective charge calculation.
• Binary Cascade
  • G4BinaryLightIonReaction: added protection against divide by 0 when energy momentum correction stalls.
• Cascade
  • First implementation of interfaces for pre-equilibrium model and equilibrium evaporation model.
• Chiral Invariant Phase Space (CHIPS)
  • More accurate initialization of dZ, dN, dS (dense area) in G4QNucleus.
  • Elastic cross-section for low energy neutrons upgraded and tuned in G4QElasticCrossSection. High energy, heavy nuclei nA correction.
  • Memory leak fix in G4StringChipsParticleLevelInterface, G4QuasiFreeRatios and G4QElasticCrossSection.
  • First implementation of the G4QDiffractionRatio class.
  • First implementation of the charge-exchange: coherent in the new process G4QCoherentChargeExchange and incoherent as an extention of the G4QuasiFreeRatios.
• De-excitation
  • Fixed bug in G4ExcitationHandler caused by attempt to access deleted pointers and iterators. Now a final state iterator list is created and filled only when necessary.
  • Fixed bug in low A,Z false condition in G4DiscreteGammaDeexcitation.
• Elastic scattering
  • G4ElasticHadrNucleusHE: optimised computation of tables. Fixed compilation of nuclear parameters for heavy nucleus (A &ft;100). General cleanup and redesign of private methods.
  • Added model names to constructors in G4LEnp, G4LEpp, G4ChargeExchange, G4HadronElastic and G4ElasticHadrNucleusHE.
  • G4HadronElastic: use QElasticCrossSection for p, n incident for any energy and any target; use HElastic for mesons only.
  • G4UHadronElasticProcess: use true flag calling CHIPS to compute cross section.
  • New class G4DiffuseElastic, new model for elastic scattering.
• High precision Neutron
  • Try to limit sum of secondary photons energy while keeping distributions in G4NeutronHPPhotonDist.
  • Renamed variable NeutronHPCrossSections to G4NEUTRONHPDATA for the path definition to the G4NDL data library.
  • Added possibility to neglet Doppler broadening mode to boost performance in G4NeutronHPElasticData, G4NeutronHPInelasticData and G4NeutronHPCaptureData, by using the environment variable G4NEUTRONHP_NEGLECT_DOPPLER.
  • Added environment variable G4NEUTRONHP_SKIP_MISSING_ISOTOPES which will force the use of the exact isotope file and set the cross section to zero and print a warning if the file is not found. Modified method of searching substitute isotope data in G4NeutronHPNames.
  • Added G4NEUTRONHP_USE_ONLY_PHOTONEVAPORATION flag in G4NeutronHPCaptureFS.
  • Bug fixes for Valgrind run-time errors and NaNs. Fixes for memory leaks.
• LLNL Fission
  • New module for Livermore LLNL neutron-induced and spontaneous fission model is now available. Model class G4FissLib must be instantiated and assigned to G4HadronFissionProcess. Pointer to G4NDL is required as in high precision neutron models.
• Parton String
  • Separate G4DiffractiveExcitation for FTF from QGS. Added original G4DiffractiveExcitation to qgsm as G4QGSDiffractiveExcitation. Moved G4SingleDiffractiveExcitation class from management to qgsm module.
• Radioactive Decay
  • Minor fix to remove compilation warnings on Windows in G4RIsotopeTable and G4RadioactiveDecay.
• Utils
  • Removed unused classes G4Fuzzy, G4EffectiveCharge and G4ParticleVector.
  • Three bug fixes in G4ReactionDynamics::TwoCluster(): removed a sharp peak at 180 degrees by fixing a units error and fixed an error in sampling the momentum transfer. Fixed array bound violation which caused in some rare occasions an infinite loop.

• Implemented signal handler for soft-abort in G4UIterminal.

• Added implementation file for G4SurfaceProperty.
• Added static G4SurfacePropertyTable to G4MaterialTable.
• Added exception to SetGROUPVEL() when E_photon<<=0.
• G4NistManager and G4NistElementBuilder: added GetNistFirstIsotopeN() and GetNumberOfNistIsotopes() methods.
• G4NistElementBuilder: added protection (Z>=limitNumElements) inside BuildElement() on Mokka request.
• G4SandiaTable: fixed deletion of elements of the table in destructor and non-initilised pointer in constructor. Use G4 native types.

• Added magnetic moment in G4ParticleDefinition, G4DynamicParticle and G4IsotopeProperty. Updated magnetic moment for leptons. Added G4IsotopeMagneticMomentTable and data file.
• Allow for registration of several isotope property table in G4IonTable.
• Allow for any Z,A combination (A>=Z) for G4IonTable::GetIon().
• Added deletion of IsotopeTableList in G4IonTable destructor.
• Fix in G4PDGCodeChecker::GetQuarkFlavor().

• In QBBC list, use consistent cross sections for elastic and inelastic.
• Activated quasi-elastic for FTF lists. Lower threshold for FTF: FTF starts at 4.0 GeV, overlaps with LEP up to 5.0 GeV.
• Added G4QEmExtraPhysics for gamma, e+-, mu+- nuclear CHIPS.
• Removed obsolete physics lists LHEP_HP, QGSP_HP, LHEP_BIC, LHEP_BIC_HP and LHEP_PRECO. Removed physics lists using Mars leading particle classes (*LEAD*).
• Removed obsolete G4EmStandardPhysics72 and G4EmStandardPhysics71; using new commands for G4EmStandardPhysics_option1,2.
• QGSP_BIC_HP is no longer experimental.
• Removed definition of name of EM constructor, now name of EM constructor defined inside the constructor itself.
• Incremented version number.

• G4VUserDetectorConstruction class now has access methods to the parallel worlds.
• Fix against accessing to the world logical volume that had already been deleted.
• Fixed memory leak in G4VUserPhysicsList and G4VModularPhysicsList.
• Fixed the name of random number status file.
• Fixed case of segmentation fault with stacking events; addressing problem report #951.
• In G4RunManagerKernel, UpdateRegion() method now internally invokes CheckRegions() method to make sure all region information are updated.

• Fixed weight treatment in G4ParticleChange.
• Added non-ionizing energy deposit in G4Step and G4VParticleChange.
• Added magnetic moment in G4DynamicParticle.

• Management
  • Added command /vis/viewer/set/picking.
  Picking is false by default, except for OpenInventor.
  • Removed G4GAGTree class.
  • Removed /vis/scene/add/eventID command from /vis/drawVolume.
  • Improved error trapping and bug fix.
  • Bug fix for zero pointer in G4VscenHamdler::LoadAtts().
• Modeling
  • Added further information on volume G4Atts.
  • Added solid dump and volume transformation to G4Atts.
  • Simplified G4Atts for Physical Volume path.
• HepRep
  • Fixed memory leak in G4HepRepFile.
• OpenGL
  • Implemented picking for OGLIX.
  • Some reorganisation of display list construction to fix rare bug in Display-by-Time feature.
  • Fixed some compilation warnings.
  • implemented picking for OGLSX (as well as OGLIX).
  • /vis/ogl/printEPS should work for all viewers.
  • Improved error trapping and bug fixes.
• OpenInventor
  • Picking is true by default.
  • Added scale factor to list of changes requiring kernel visit, and some rationalisation.
• RayTracer
  • Fixed memory leak at initialisation.
• ASCIITree
  • Removed GAGTree.

• g4py: implemented software abort by <Ctrl-C>.

• New data for neutron cross-sections, G4NDL.3.11:
  • Added Germanium data and Silver data.
  • Added Special file 0_0_Zero which give 0 cross section over all energy range.
• New low-energy data set, G4EMLOW.4.3:
  • Added auger file au-tr-pr-100.dat.
• New data set for radioactive-decay, G4RadioactiveDecay.3.2:
  • Fixed z81.a188 element.

• Updated reference outputs.
• advanced
  • brachytherapy
    • Few improvements resulting from code review. No functionality changes.
  • cosmicray_charging
    • Fixed compilation problem in physics-list.
  • hadrontherapy
    • Geometry upgrade (hadrontherapyBeamLine class) according to the experimental CATANA proton therapy beam line.
    • Improved the modulator geometry and fixed a bug on its rotation.
    • Physics component redesigned and reimplemented as a result of code review.
  • human_phantom
    • Corrected bug in MIRDLeftBreast and MIRDRightBreast.
    • Introduced voxel Left/Right breast, VoxelBreastSensitive detector and PhantomHeadBuilder.
    • Introduced Left and Right Adrenals and Scapulae.
    • Minor correction to documentation.
  • radioprotection
    • Correction to documentation.
  • underground_physics
    • Migrated to new interface for SetMscStepLimitation(fMinimal) in DMXPhysicsList.
    • Changed to hIonisation from hLowEnergyIonisation.
  • Tiara
    • Currently not supported and not migrated to new python, scoring or biasing.
• extended
  • analysis/N03Con
    • New example showing how to apply convergence test in simulation.
  • biasing
    • Migrated example B01 and B02 to new biasing/scoring scheme.
    • Removed B03 (redundant/obsolete use of Python).
  • electromagnetic/TestEm1
    • Use G4EmProcessOptions to define bremsstrahlung threshold energy.
    • Completed PhysListEmStandard options and added new input macro for e-.
    • Make visualization only in interactive mode.
  • electromagnetic/TestEm2
    • Completed PhysListEmStandard options.
    • Updated main input macro file; use visualisation only for interactive session.
  • electromagnetic/TestEm3
    • Use renamed EM builders and update macro files.
    • More options in PhysListEmStandard.
  • electromagnetic/TestEm5
  • electromagnetic/TestEm7
    • Fixed bug in standardSS Physics Lists.
    • Added EM builders from physics_lists library.
    • Added output of mean number of primary steps and kaon+ run in main input macro.
  • electromagnetic/TestEm9
    • Use standard names for histograms and updated macro files.
    • Substituted 'autoptr' by normal pointers in Histo to avoid crash.
    • Use renamed EM builders, and emstandard_opt1 EM physics in TestEm9.in.
    • Extended reference output: showing both mean/RMS values and truncated mean/RMS.
  • errorpropagation
    • New example illustrating the use of the error propagation utility. A G4ErrorFreeTrajState is created to simulate a muon track of 20 GeV along the X axis. Then the track is propagated until the target is reached.
  • field
    • Removed un-necessary initialisation of G4FastSimulationManagerProcess from setup.
  • hadronic/Hadr01
    • Physics List updated according to recent changes in hadronics and physics_lists packages.
    • Substituted usage of autoptr by normal pointers in Histo to avoid crash.
    • Added histogram on total energy deposit in the target.
    • Use renamed EM builders G4EmStandardPhysics_option1 and G4EmStandardPhysics_option2.
  • medical/fanoCavity
    • Updated to follow change of interface of G4VEmModel::SampleSecondaries().
    • Updated EM options.
  • medical/Gammatherapy
    • Use renamed EM physics-list components from physics_lists module.
    • Substituted usage of autoptr by normal pointers in Histo to avoid crash.
    • Fixed singleton GetPointer() method.
  • parameterisation/gflash
    • Updated physics-list: register the fast simulation process as a discrete process.
  • runAndEvent
    • Fixed indexing scheme of GetIndex() method in the scorers.
• novice
  • N02
    • Show usage of G4GeometryTolerance tuned for the geometry topology.
  • N03
    • Changed visTutor/exN03Vis12 and visTutor/exN03Vis13 macros, following change to /vis/ogl/printEPS command.
    • Increased verbosity in rndmSeed input macro.
  • N04
    • Delete materials and elements when quitting application.
  • N05
    • Migrated to new parallel geometry scheme.
    • Added vis.mac macro.
  • N07
    • Introduced ExN07ParallelWorld and related modifications.
    • Modified vis.mac to allow for visualization of the parallel world.