Newer
Older
Markus Frank
committed
nanosecond s=1e9 second ns=1e-9,s=1
Mega electron Volt GeV=1e3 GeV GeV=1
radian rad=1 degree rad=180/pi
Unclear to me was if in Geant3 steradian is also 2*2*pi or 2*2*180 degree.
No-date F. Gaede
-------------------
- changes to build system:
(see ReadMe.txt for details)
- moved examples to ./example directory
- they are not built as part of DD4hep anymore
- install thisdd4hep.sh in ./bin
(modified to have the correct pathes)
- added -DINSTALL_DOC=on/off option
to build doxygen documentation (in ./doc/html/index.html)
- create DD4hepConfig.cmake for easy building against DD4hep
( see examples CMakeLists.txt)
- fixed doxygen API documentation (C.Rosemann)
(greatly improved with many graphs)
--------
| v00-04 | fourth beta release ...
-------- - placement options are now compatible with geant4
- fixed Geant4SensitiveDetector::getVolumeID()
- modified ILDEx model to not use assemblies (for now !?)
- removed obsolete geant4 example ILDEx
- Fix population procedure of volume manager
- Fix uniqueness of volume identifiers.
- Remove call with lookups by placed volume to the volume manager
- improved ILDEx toy example:
- fixed creation of cellIDS for simhits
- creates SimTrackerHits for VXD, SIT and TPC
- creates SimCalorimeterHits for AHcal barrele and endcap
- works now wigth ILDExSimu _and_ SLIC if no assemblies are used
- assemblies work with the DD4hep Geant4Converter and VolumeManager
- added a prototype example for a ROOT independent
plugin mechanism for SensitiveDetectors
- added example implemetation for Calice test beam
calorimeters: CaliceTbeam (Shaojun Lu, DESY)
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
2013/21/06 Markus Frank
--------------------------
1) Since the placement model for combined translations and rotations
caused more confusion than good, it was agreed to move to a more
Geant4 like placement model using Vectors, Rotations and Transformations
from ROOT::Math. To place a physical volume only one of these
is allowed as an argument: either a rotation or a translation
for simple placements. Combinations may be constructed with Transform3D
objects allowing more complicated placements.
The same mechanism was applied for the operations to construct
boolean solids.
2) The CLICSid Examples were updated according to the changes necessary
from 1)
3) The Tesla drivers from the ILDExDet example were removed. They are
not compliant to 1).
4) The G4 package allows now to translate TGeo geometries to Geant4.
Visual inspection has shown an agreement between the two geometries.
5) The TGeo to LCDD/GDML conversion is still buggy. Hence, simulations
using slic as an engine do not yet work. This is being looked at.
IMPORTANT NOTICE:
The changes described in 1) may affect existing detector constructors.
Unless all problems are solved you should be careful to move to the svn
head. In any case save your checkout or use the tag v00-03.
--------
| v00-03 | third beta release ...
-------- - broken geometry (for some rotations/translations)
- 'backup release' before placement options
will be reduced
--------
| v00-02 | second beta release ...
--------
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
2013/02/06 Markus Frank
--------------------------
1) Simplyfy inheritance for common objects from TNamed.
Use direct inheritance of Object from TNamed rather than hidden
using the Value<a,b> construct.
2) Added physical volume manager to simplyfy the detector element
and sensitive detector lookup from a given physical volume.
This implies: If the volume manager is instantiated, the readout specifiers
MUST be correct. This is in most xml files NOT the case.
The "system" field holding the sibdetector-id is MANDATORY!
Otherwise you will receive messages like this:
Exception:SITCollection: This ID descriptor has no field with the name:CellID0
while programming VolumeManager. Are your volIDs correct?
3) To check the volume identification, add argument -volmgr to geoConverter,
geoDisplay, etc.
4) To trace possible memory leaks:
export DD4HEP_TRACE=Yes
then run you converter etc. At the end a table is displayed with a "leakage"
column, showing how many objects were not deleted.
+----------------------------------------------------------------+
| I n s t a n c e c o u n t e r s b y T Y P E I N F O |
+----------+---------+-------------------------------------------+
| Total | Leaking | Type identifier |
+----------+---------+-------------------------------------------+
| 13| 0|DD4hep::Geometry::DetElement::Object
| 3| 0|DD4hep::Geometry::SensitiveDetector::Object
| 3| 0|DD4hep::Geometry::Readout::Object
| 1| 0|DD4hep::Geometry::OverlayedField::Object
| 1| 0|DD4hep::Geometry::CartesianField::Object
....
Ideally the second column only has "0"s. Instances of 1 may be OK (singletons).
Prenotice:
As soon as ROOT v 6.00 is out, we will have to use it!
Andrei kindly agreed to implement a few changes to TGeo, which will make
life much easier and the implementation cleaner and less cumbersome.
--------
| v00-01 |
--------
- first beta release...
2013/20/03 Markus Frank
--------------------------
1) Finished the compact->lcdd converter
Extract lcdd information
$ > geoConverter -compact2lcdd -input file:<compact-input-xml-file> -output <detector>.lcdd
$ > <SimDist>/scripts/slic.sh -o output.slcio -g SiD.lcdd -m <geant4-macro>.mac -r 100
2) Finished the compact->gdml converter
Full GDML cycle can be tested (deplace <detector> with some meaningful identifier):
a) Extract gdml information
$ > geoConverter -compact2gdml -input file:<compact-input-xml-file> -output <detector>.gdml
b) Extract visualisation hints to csv file
$ > geoConverter -compact2vis -input file:<compact-input-xml-file> -output <detector>.vis.csv -ascii
c) Load GDML into ROOT:
$ > root.exe ../doc/gdml_root.C+\(\"<detector>\"\)
The macro expects *at least* the <detector>.gdml file. If no visualisation
information is found, the resulting graphics however is not very nice!
2013/22/02 Markus Frank
--------------------------
1) Today a decision was taken to remove all _X() and _A() macros
accessing static unicode strings. A common macro will be used to
address both XML elements and XML attributes using the macro _U(...).
The original usage only created confusion.
All _X and _A macros were replaced by _U.
2) Improve the executables running the plugin mechanism in UtitlityApps.
They now complain if the plugin was not found.
2013/21/02 Markus Frank
--------------------------
Require strict checks for the existence of xml attributes.
e.g. default=false for Element::attr<bool>() is no longer an option.
if the attribute is not present in the element, an exception
is thrown.
Slightly change the factory method for creating detector elements.
Use a more generic pattern do support also more generic XML
processing.
Install rootmap mechanism for the usage of plugins.
It is no longer necessary to have separate programs
for the various detector models (ILD/SiD).
One generic display program servest them all.
==> Creation of package DDExamples/UtilityApps
example: $> bin/displayGeo -compact <compact-xml-file>
This move has another consequence:
All libraries are build into a common install area:
the <build directory>/lib.
All executable binaries are build into a common install area:
the <build directory>/bin.
The LD_LIBRARY_PATH and PATH variables as defined in
'thisdd4hep.(c)sh' do include these directories
DD4hep release notes. Better start them late than never.
If you perform significant changes to the DD4hep core,
plase leave a small notice here.