Activity
From 01 Jul 2014 to 30 Jul 2014
30 Jul 2014
- 18:28 Revision a5f47392: Updates in CHANGELOG and the release script.
- 17:24 Feature #698 (Backlog): Fix incompatibility between integration over bin and infinite particles
- 17:23 Feature #702 (Backlog): Implement position particle info equivalent in GUI
- 17:23 Feature #703 (Backlog): Implement rotation of particles in GUI
- 17:23 Feature #704 (Backlog): Implement Lattice2D interference function in GUI
- 17:23 Documentation #263 (Backlog): Review morphology file implementation in IsGISASXS and BornAgain
- 17:22 Feature #682 (Backlog): Provide PlotWidget with x,y-axis in both, radians and degrees.
- 17:22 Documentation #692 (Backlog): Put manual into form that is appropriate to be cited
- 17:22 Documentation #691 (Resolved): Provide simple cartoons for concepts that are otherwise difficult ...
- 17:22 Feature #679 (Resolved): Revise location of auto generated BAVersion and similar files
- 17:22 Refactoring #409 (Backlog): Core: remove duplicate functions "evaluate" and "evaluatePol"
- 17:21 Feature #393 (Backlog): Create Mac installer
- 17:21 Documentation #330 (Backlog): Core: beam propagation: take care of kz=0 case appropriately
- 17:20 Revision c6e8130e: Bug in cmake and BAConfigure
- 17:18 Revision 689b6d7d: Merge branch 'BAVersion'
- 17:17 Revision 5f95a89b: Updates in release script.
- 16:31 Revision fee728af: BAConfigure now generated in build directory and not present in source tree at...
- 14:27 Revision 1e6d80d4: On the way to new release script.
- 11:08 Revision 3a8b7d73: Cleaning, commenting, formatting. Erase the function
- IInterferenceFunctionStrategy::evaluateWithoutMC.
29 Jul 2014
- 12:54 Revision 2cf94a93: Few comments.
- 12:33 Revision 03d2b9cd: Different implementation of the complex square root. By default, which one
- which works with this example.
with - 10:58 Revision 2dad2267: Find the problem !! The problem was bound to complex_cu::sqrt, and
- particularly with squqre root of negative real. In fact, because of numeric
unstability during transfert between devi...
28 Jul 2014
- 18:18 Feature #729 (Resolved): Integrate Mahadi's QuickSimulation widget into run simulation evironment.
- 18:15 Revision 548e4772: Merge branch 'QuickSimulationFix'
- 18:15 Revision c1e38e74: Code cleanup.
- 17:05 Revision 1d70adb8: Merge branch 'QuickSimulation'
- 16:59 Revision 9040cb51: Bugfix in DetectorEditorWidget with wrong detector type constants.
- 16:55 Revision 40e1f1a4: Bugfix in PlotWidget with QCPColorMap.
- 15:39 Revision f6e37c5d: New QuickSimulationRunner class and adjustments in JobQueueData for quick simu...
- 14:07 Revision 05afa526: The problem comes from execute in CalculateAsFormFactor.
- 13:21 Revision de364b82: Do not find the error in CUDA: the functions are the same, the inputs are the
- same, but the results are differents, and I do not understand why.
Add a new command line to choose between CUDA wit... - 10:52 Revision 2fcd63fa: Merge branch 'OpenMPI'
- 10:26 Revision b6f779d2: Update in OpenMPI scripts.
- 10:06 Feature #738 (Archived): Perform new round of profiling and memory leackage investigation
- 09:55 Feature #737 (Archived): Provide example demonstrating Monte-Carlo integration for big particles
- And corresponding functional test.
- 09:52 Feature #736 (Archived): Compile GUI under Windows with Qt5.3 and provide installer.
- 09:34 Revision bd58f0ec: Refactoring in QuickSimulation
25 Jul 2014
- 18:44 Revision f0fa1634: SampleTuningDelegate is connected.
- 17:42 Revision 52a80d81: Merge branch 'PlotBugFixes' into develop
- 17:25 Revision bba9e246: Update in OpenMPI scripts.
- 17:21 Revision 3ed996fb: Update in OpenMPI scripts.
- 17:03 Revision 21275302: Updated qcustomplot library
- 16:54 Revision 3fb16929: Reorganize the files. Creating new files in FormFactors folder, named
- CalculateAsFormFactor.* where the calculation functions are saved.
Problem with CUDA!!! For some unknown reasons, th... - 13:10 Revision 6df41855: Update in OpenMPI script.
- 13:04 Revision c11f3439: Resolved lines issue over plot
- 13:01 Revision 3d12001c: Update in OpenMPI scripts.
- 12:57 Revision 9784ab9f: Only currently selected MultiLayer is propagated into QuickSimulation.
- 12:56 Revision aea60d64: Resolved file name without extension issue during saving plot
- 11:30 Revision 9c439bb9: Initialization order changed in QuickSimulation widgets.
- 10:40 Revision a2caef65: Ownership of basic views and models (SampleView, SampleModel etc) is changed t...
- 08:42 Revision 625e3452: Update in OpenMPI script to run bulky mesocrystal simulation.
24 Jul 2014
- 18:03 Revision 14a11326: Renaming campaign in QuickSimulationWidget.
- 16:52 Revision dbe29cd6: Add a new interference function (Interference Function 1D Lattice), as well as
- the FTDistributions. 1D Distributions were adapted to be used in CUDA.
- 16:00 Revision 68f0afe5: Updates in OpenMPI script.
- 15:48 Revision 61124d57: Merge branch 'ParameterViewBugFix' into develop
- 15:45 Revision 6138c706: fixed slider issue in parameter view
- 14:47 Revision ca9f60e8: User Manual - additional figures to illustrate theoretical section
- 14:40 Revision 7693b5b6: Add a new approximation strategy. All works. However, because of a mysterious
- command line, I do not implemented the whole strategy (only the half).
- 14:02 Feature #696 (Resolved): Implement check for unsaved project and corresponding save/discard widget
- 13:50 Revision 10946159: Merge branch 'SaveProjectWidget' into develop
- 13:48 Revision 77550b17: Implemented window close event
- 13:47 Revision 97da429f: Modified init order for item
- 13:15 Revision a125cb65: Adding a new formfactor (formfactorPrism3). Nothing seems wrong.
- Time to update from the old file, about 15 min.
- 12:29 Revision a3feaba9: Saving.
- 10:43 Revision c3013d1b: Upgrading the function IFormfactor::calculate, so save 0.2s.
- The function IFormFactor::calculate is the function in CUDA which spend the
whole time. In this function, there are (... - 10:23 Revision 8b8a4c30: Merge branch 'ParameterModel'
- Conflicts:
GUI/coregui/mainwindow/mainwindow.cpp
GUI/coregui/mainwindow/mainwindow.h - 09:55 Feature #663 (Resolved): Improve MaterialEditor logic
- 09:50 Revision e8c12703: Merge branch 'ParameterModel'
- Conflicts:
GUI/coregui/Views/TestView.cpp
GUI/coregui/mainwindow/mainwindow.cpp - 09:39 Revision b6adfd05: Merge branch 'MaterialEditor'
- 09:38 Revision 452b7dc3: Cleanup in ParameterizedItem.
- 09:20 Revision df559eb4: Cleanup from old GroupProperty.
- 09:07 Revision 926cbc31: New InstrumentModel class inheriting SampleModel.
- 08:49 Feature #729 (Archived): Integrate Mahadi's QuickSimulation widget into run simulation evironment.
23 Jul 2014
- 18:03 Revision 3beea6f5: MaterialModel is connected with SampleModel.
- 16:59 Revision 49463c7b: Change comments.
- 16:58 Revision d752d7d9: SampleModel is introduced which is an alias for SessionModel.
- 16:18 Revision 56660cd2: Simplify the code : now and henceforth, the function IFormFactor::stock is not
- any longer a virtual function, but directely call the function IFormFactor::
stockWithoutSaving, which is still virtu... - 15:59 Revision 54319b44: Bugfix with wrong LayerRoughnessItem construction.
- 15:23 Revision bdf4da1b: Bugfix with mixed int/double variants.
- 15:16 Revision a032d503: Bugfix in RefractiveIndexItem with screwed variant type.
- 13:47 Revision 09808dfd: Change error messages.
- 13:45 Revision a995a547: Forget to delete an unused variable.
- 13:43 Revision eebcbb2c: New commentary. Changing FastestFourierTransform.h. Cleaning files.
- 11:00 Revision 1ba421fd: Solving the problem bound with fftw3.h and nvcc.
- Create a new file FastestFourierTransform, where __float128 is defined or
redefined. Seems working with CUDA and C++. - 10:50 Revision c46d4318: Transform of GUI to domain is restored.
- 09:24 Feature #699 (Resolved): Implement ParameterModel and corresponding view delegates for real time ...
22 Jul 2014
- 18:34 Revision b0f0d28b: Completed:
- 1. Simulation tab view
2. LinkItem implementation - 18:14 Revision 85507d92: Code cleanup.
- 17:26 Revision b640bcd5: Restored serialization of MaterialModel.
- 17:07 Revision 48796a98: Just some comments, and cleaning the files.
- Remark : in CUDA, with this parameters, the function IFormFactor::calculate
delays 1.63s on 1.88s, id est 86.7% of th... - 16:46 Revision 1f0006f8: MaterialProperty now relies on unique identifier.
- 15:59 Revision 1a6d9545: Reducing the number of calling of Bessel functions => Improving the algorithms.
- 15:37 Revision da702155: A precision about arrays in CUDA.
- 15:05 Revision 215243e3: Updates the pdf files.
- Improving the pow function in complex_cu.
- 14:04 Revision 7e52b86c: Finnally, the CUDA algorithm is quicker than the C++ algorithm.
- Just know that in CUDA, using arrays on device is VERY slow. So try to avoid
local arrays, and reduce the number of g... - 13:32 Revision 9bc5f613: MaterialItem switched to ParameterizedItem.
- 11:06 Revision cebbab08: Increasing the algorithms. Now CUDA code is nearly as quick as C++ code.
- Changing mainly the IFormfactor.cpp files, where the outputs coefficients
are calculated only once (instead of 4). Mo...
21 Jul 2014
- 18:18 Revision 14eb9bd7: Refactoring in MaterialEditor.
- 15:32 Revision c0d9cfdc: Bugfix with not deleted FancyGroupEditor
- 15:32 Revision 5ab05f1b: Cycling refactoring in FancyGroupProperty.
- 15:32 Revision 4a4a90f4: Switching to FancyGroupProperty part 1.
- 15:32 Revision c9e147f0: Further refactoring in ItemFactory
- 15:32 Revision c831a447: Further refactoring in ItemFactory.
- 15:32 Revision 002ff269: Functionality of GroupProperty is restored.
- 15:32 Revision 7fd741f0: Materials VariantManager and VariantFactory are removed, they functionality is...
- 15:32 Revision bc62547c: New ScientificDoubleProperty.
- 15:32 Revision 9e509139: Serialization for ScientificDoubleProperty.
- 15:32 Revision ae8d6a1b: On the way to new ItemFactory.
- 15:32 Revision 51b57dd6: Bunch of QString constants for item factories.
- 15:32 Revision ddfb0df2: Refactoring in ItemFactory and more QString constants.
- 15:32 Revision ff48f6f0: More const QString in FormFactor items.
- 15:32 Revision c45ccc0c: Refactoring in PropertyAttribute.
- 15:32 Revision e5c6dd99: On the way to GroupProperty refactoring.
- 15:32 Revision a11f253a: Refactoring in ParameterizedItem to be aware of property change in sub items.
- 15:32 Revision e54ac9b2: FancyGroupProperty test
- 15:32 Revision 504023dc: On the way to FixedGroupProperty
- 15:32 Revision fe00c31c: First working version of FancyGroupProperty.
- 15:28 Revision 94b088c8: It does not work, and I do not know why...
- The calculation on kernel takes too much time, and CUDA returns an error,
which stops the calculation.
Remark : lauch... - 13:27 Revision 802953c8: Added stylesheet for tree view
- 13:27 Revision 99e0b82e: Added additional images for tree view
- 13:09 Revision 05194565: Some tests to try to understand why CUDA was so lang, compare to C++.
- 11:35 Revision 1d383b26: Finish to adapt the code. Now, the calculation are not made in a loop anymore,
- in CUDA code. Yet, the parallelize code is not faster as C++ code !?!
Perhaps, the problem comes from there are 4 pro... - 10:19 Revision 38f56c68: Second part of amelioration. Changing evaluate_as_DWBA and stock in
- FormFactorDWBA. Changing IformFactor::stock and stockWithOutSaving.
Changing CalculateIntensitySimulation.cpp. Now, g... - 09:22 Revision 94e7fc01: First part of amelioration : changing evaluate_as_DWBA_Constz and stock in
- FormFactorDWBAConstZ.
Did Not test in CUDA yet !
18 Jul 2014
17 Jul 2014
- 14:49 Revision cd5f3ffe: Updates the pdf files.
- Remark : if you still have problem in CUDA, like "segmentation fault" when you
increase the number of points and/or f... - 14:04 Revision 624f1133: Solving the problem bound with max length array of indicators array.
- Now, you can create until 100 particules as wish.
- 12:56 Revision 4420eca0: Find the new problem : length of arrays are too huge. Solving should come
- later.
- 11:11 Revision bce44a3e: Solving the problem bound to array lengths.
- Adding a line which allows you to choose if you want to have details on
execution times or not (commenting /decomment...
16 Jul 2014
- 13:05 Revision 2d2da98a: Inserted timing and possibility to create more particles
- 11:08 Revision 6255dd7a: Merge branch 'ModelViewTest'
- 11:01 Revision df6f9d5b: Added slider in TestView's QStandardItem
- 11:01 Revision 1fbd9382: Implemented slider and editor box
15 Jul 2014
- 13:55 Revision 88be21e7: Merge branch 'SuperMUC'
- 13:54 Feature #713 (Resolved): Provide simplified OpenMPI support in runSimulation
Mesocrystal simulation is working for OpenMPI, need to run now something really time consuming.
See explanatio...- 13:48 Revision 54217363: Update in OpenMPI scripts.
- 12:44 Revision 0b66dc82: Update in OpenMPI scripts.
- 12:38 Revision 41e72d4b: Update in OpenMPI scripts.
- 12:35 Revision e3742484: Update in OpenMPI scripts.
- 11:40 Revision 69186861: Update in OpenMPI scripts.
- 11:33 Revision 2e8c5b11: Update in OpenMPI scripts.
- 11:18 Revision 3e5c6c44: Update in OpenMPI scripts.
- 11:05 Revision fc06359c: Update in OpenMPI scripts.
- 10:50 Revision 23c6b22c: Updates in OpenMPI scripts.
- 10:37 Revision f0c9467f: Change in debug printing, updates in openmpi scripts.
14 Jul 2014
- 17:49 Revision 552abdd2: Fix in OpenMPI example
- 17:44 Revision e92a43a2: Fix in OpenMPI example.
- 17:18 Revision 68ba47a4: Updates in OpenMPI simulation scripts
- 17:08 Revision 83fc4e34: Updates in OpenMPI examples
- 16:03 Feature #600 (Resolved): Provide compilation of BornAgain at lrz cluster
- 15:54 Revision b08b46c7: Bugfix in FindEigen
- 15:39 Revision 2a35ce2e: Bugfix in FindGSL module
- 12:54 Revision b3ce8baf: Boost version is decreased
- 12:54 Revision cf8a0d4f: Minimal Boost version decreased to 1.47, setup script for SuperMUC
11 Jul 2014
- 10:37 Bug #709 (Resolved): Take care of thick layers
- 10:25 Revision 40853742: Regenerated Python API
- 10:18 Revision ed089eb6: Merge branch 'thicklayers3' into develop
- 10:10 Revision f6ba4041: Fixed unit test
10 Jul 2014
- 17:44 Feature #695 (Resolved): Provide panning functionality for SampleView
- 17:44 Feature #694 (Resolved): Provide select all functionality for SampleView
- 17:44 Feature #694 (Sprint): Provide select all functionality for SampleView
- 17:40 Revision 8e5cea23: Merge branch 'DesignerViewPan'
- 17:32 Revision 34cec15f: Bugfix with unconnected resetToDefaultLayout in SampleView
- 17:04 Revision 679608bb: Bugfix with double delete in the case of the complex sample which was fully se...
- 15:56 Revision 50b09f96: Adjusted behaviour of rubber band in the case of connection being established.
- 15:28 Revision 7bbcbec3: New zooming logic for DesignerView
- 14:05 Revision 673da47d: SampleToolBar logic for select/pan communications with SampleView
- 13:49 Revision de57e4b0: New staffs to have an estimation of execution times (in C++ and CUDA code).
- Changing the pdf.
09 Jul 2014
- 18:10 Revision cf8d86e4: New icon set of SampleView
- 17:42 Revision c889b910: Regenerated reference files after adapting normalization in Detector class
- 14:18 Revision f8e94c31: Adapted GetSolidAngle() for normalization of detector cell intensities
- 13:34 Revision ec940117: Magnetic R and T coefficients are also calculated when needed now (no
- longer precalculated)
- 10:27 Feature #726 (Resolved): Polarized: Provide numerically stable calculation of R,T coefficients fo...
- As part of the refactoring for curing the instabilities caused by thick layers, the scalar calculation for R, T coeff...
08 Jul 2014
- 17:23 Revision 76d45ca6: Multiple delete of selected items, new scale view button.
- 16:10 Revision 1d789b91: Allowed multiple selection on graphics scene, pan functionality for graphics v...
- 14:04 Revision 26ded373: Implemented MC integration for thick layers; switched to MISER algorithm
- for Monte Carlo integration
- 14:01 Revision 35f7cf48: Merge branch 'OpenMPI'
- 11:47 Feature #523 (Resolved): Expose to python MSG::SetLevel
- 11:47 Feature #523 (Sprint): Expose to python MSG::SetLevel
Occasionally implemented it.- 11:46 Revision b3bb6037: SetMessageLevel is propagated into python API
- 11:00 Feature #720 (Rejected): Provide OpenMPI support in FitSuite
- 10:22 Revision 909995bb: OpenMPI switched off
- 09:45 Revision 2d91944a: OpenMPI example in dev-tool and experimental class OMPISimulation
07 Jul 2014
- 14:17 Revision e1581652: Fixed a unit and functional test
- 14:12 Revision f7449fe1: OpenMPI hello world.
- 13:07 Revision 807badb2: Scalar R and T coefficients are no longer precalculated (calculated when
- needed now)
- 11:10 Revision cda0ed26: New functional test to check batch simulation.
04 Jul 2014
- 16:28 Revision f5c743df: Force usage of RTcoefficients of midpoint for alpha
- 15:39 Feature #713 (Archived): Provide simplified OpenMPI support in runSimulation
- 15:09 Feature #710 (Resolved): Provide toy OpenMPI application to run at LRZ
- 11:27 Feature #710 (Archived): Provide toy OpenMPI application to run at LRZ
- 14:00 Feature #701 (Resolved): Implement layer roughness in GUI
- 13:55 Revision 1a670a93: Merge branch 'LayerRoughnessItem'
- 13:33 Revision 9ab21e37: Unit test for LayerRoughnessItem.
- 13:03 Revision a8729ea4: The same file as before, but as pdf format.
- 11:24 Feature #600 (Sprint): Provide compilation of BornAgain at lrz cluster
- 11:16 Revision 9d7c1fb5: A file which sums up the most important changes between C++ version and
- C++/CUDA version. (In fact a C++ version is different form a C++/CUDA version,
which is also different from a CUDA ve... - 11:15 Revision 9aa525c6: Merge branch 'thicklayers2' into develop
- 11:12 Revision 3395ec0d: Fixed bug in qz calculation (different methods in two parts of the code)
- 10:06 Revision 4b95f27b: Other comments.
03 Jul 2014
- 12:48 Revision b13e8da5: LayerRoughnessItem GUI functional test.
- 12:36 Revision 2ea1fbd3: New small feature in complex multiplication.
- 12:31 Revision 5ab297f1: LayerRoughness transform from GUI to domain is implemented.
- 11:17 Revision c678d8db: LayerRoughness from domain to GUI.
- 09:11 Revision e91d56c1: Correct few little mistakes in commentary. Confirmation : rccursion is not
- allowed in CUDA code (generate a warning with nvlink).
02 Jul 2014
- 17:44 Revision a514f89f: MultiLayerImte updates visibility of Layer's properties
- 16:25 Revision c802a128: Add some commentary, to try to explain the code (if it can help ...).
- 16:12 Revision 9108dce0: New LayerRoughnessItems
- 13:29 Feature #688 (Resolved): Include GUI in nightly build
GUI+tests are included in nightly build. Due to our misantropic application server intended for usage by hard-boi...- 13:09 Revision dc29f070: Merge branch 'BuildIssues3'
- 13:06 Revision 18de39b9: Attempt to run QTests in headless mode.
- 11:42 Revision b4f044f1: Merge branch 'BuildIssues2'
- 11:27 Revision 12a41cdc: CMake message about Qt version found.
01 Jul 2014
- 17:32 Revision e4e7fe9d: Bug with the order of static members initialization
- 17:19 Revision 1a9ffa49: Implemented scalar R,T coefficient calculation that is more resistent to
- thick layers
- 17:07 Revision b318ec6c: Merge branch 'CoreShellItem'
- 15:40 Feature #700 (Resolved): Implement core shell particles in GUI
- 15:31 Revision b7fff79f: Last verification, and all seems working. Deleting the useless files (because
- thre are created by CMake).
Remark : compile with nvcc still generates a warning (in LLData.h), but it does not matte... - 15:30 Revision 5562efc7: Bugfix with propagation of core position from domain.
- 15:22 Revision cee78102: Modification of Core/CMakeFil;es.txt. Now, anyone should be able to launch
- this program.
- 15:16 Revision 1a23ecc1: New functional test for core shell particle.
- 13:44 Revision 803c5924: CoreShellParticle example from GUI to domain is working.
- 11:56 Revision 0dda88b9: Automatic disabling of depth,abundance property while connecting to CoreShellItem
- 11:53 Revision 46de2a7f: Finnally, it works (on my computer)!!!! I save the datae, and I begin to write
- the Core/CMakeFiles.txt.
- 11:41 Revision d0b0b0a7: GUIObjectBuilder now works for ParticleCoreShell samples.
Also available in: Atom