New Features:
Line Plotting: The new Line Plotting utility allows users to generate graphs of their data to help steer their visual analysis. Multiple variables can be plotted between two points in space, or through time at a single point. More info here.
Variable Statistics: A statistics utility has been written to make key metrics in users’ data conveniently accessible. Currently supported values include variable minimum, maximum, mean, median, and standard deviation. These values can be calculated across time at a single point, or within a volume. More info here.
SeedMe Video Encoder: A video encoding tool based on the online video encoder SeedMe has been incorporated into Vapor to make movie production easier. Once users have registered an account with SeedMe.org, they will be able to easily generate video content with the screen captures of their data. More info here.
Enhanced Colorbar Manipulation: Support for mouse-based colorbar manipulation has been added. Users can now control the size and position of their colorbars by clicking and dragging the mouse.
Default Colorbar Titles: An option for a default colorbar title has been added to Vapor’s visualizer-features. This will give the name of the corresponding variable as the title of each colorbar.
Bugs Fixed:
1426 GRIB importers not working on Windows
1422 stats: minimum timestep resets to 1
1421 statistics: crash after setting center at 0 in center/size mode
1420 statistics: core dump
1417 Plot: add variable list changes orders after adding and removal variables
1415 SeedMe: should warn file overwriting when user clicks "go"
1412 Seedme: consistent button reaction.
1411 Plot: Spatial coordinate for time plots should default to the domain midpoint
1410 Statistics: Center textEdits do not properly correct size sliders
1409 Plot sliders do not move on 8spot
1406 Statistics: Max shows inf sometimes
1400 Plot: Timestep slider and text box inconsistent
1399 line plotting: can't change time step
1398 line plotting: state information not retained
1397 line plotting: const check box broken
1396 core dump after using plot utility
1393 GUI false alarm "invalid metadata file" when cancelling an "load data set" operation
1392 Plot: save to file doesn't report error on failure
1391 Vapor fails an assertion with multiple visualizers
1390 Time ordering of WRF data incorrect
1388 ROMS converter not applying missing values
1383 Vapor crashes on certain GRIB files
1380 Windows 10 installer improperly applies environment variables
1377 statistics: unable to use derived variables
1374 SeedMe: download link gone
1373 Statistics: 2nd time open forgets what stats to calculate
1371 Statistics extents change when timestep changes
1370 Cannot load 8spot3d with Windows 10.
1366 SeedMe: First time user screen has overlapping text in OSX 10.8
1365 Stats: need to limit digits of numbers to display when short of space
1363 Stats: Export Text doesn't check if write successful
1359 Statistics: Single point statistics do not appear correct
1356 Stats: auto-update not working
1355 Plot: default coordinate system should be space not time
1350 SeedMe file selection window pop up twice
1346 SeedMe: Qt warning messages from FirstTimeDialog
1340 SeedMe: Advanced tab doesn't function
1336 SeedMe: displays "processing..." even no input image selected
1332 Colorbar: Can't display both variable name and user supplied annotation
1293 SeedMe Select Images: No way to delete what is selected
1291 SeedMe save video overwrites existing file without warning
1287 line plotting: Z coordinate should be locked for 2D variables
1266 Colorbars rendering as white boxes on Debian
1250 ncdfvdfcreate: -zcoordvar variable is not added to -vars for data conversion
New Features:
Line Plotting: The new Line Plotting utility allows users to generate graphs of their data to help steer their visual analysis. Multiple variables can be plotted between two points in space, or through time at a single point. More info here.
Variable Statistics: A statistics utility has been written to make key metrics in users’ data conveniently accessible. Currently supported values include variable minimum, maximum, mean, median, and standard deviation. These values can be calculated across time at a single point, or within a volume. More info here.
SeedMe Video Encoder: A video encoding tool based on the online video encoder SeedMe has been incorporated into Vapor to make movie production easier. Once users have registered an account with SeedMe.org, they will be able to easily generate video content with the screen captures of their data. More info here.
Enhanced Colorbar Manipulation: Support for mouse-based colorbar manipulation has been added. Users can now control the size and position of their colorbars by clicking and dragging the mouse.
Default Colorbar Titles: An option for a default colorbar title has been added to Vapor’s visualizer-features. This will give the name of the corresponding variable as the title of each colorbar.
Bugs Fixed:
#1267: Uninitialized value in idwt
#1261: Can’t import or convert WRF data with no 3D variables
#1259: wrfvdfcreate hangs when converting WRF-FIRE data
#1258: WRF converters report “No Data” when the -vars flag is used
#1257: vaporgui can’t display images with non-georeferenced data
#1256: Vaporgui and vdcwizard hang with idealized wrf data
#1252: Vapor crashes after creating dataset with long y-axis
#1251: Python returns some arrays in reverse ordering
Bugs fixed:
The following significant bugs were fixed in the current release.
#1248 Scipi missing fortran library on OSX 10.10
#1240 wrf2vdf can't find libvdf.dylib
#1239 Stereo contexts causing flicker on unsupported displays
#1238 Undo keyframing disables timestep textbox
#1237 VDCWizard halts after failing to copy variable
#1236 Variable name XML error
#1235 gribvdfcreate timesteps not detected in grib2 file
#1233 Cannot save .tiff on Linux
#1230 NaturalEarth imagery not installed on windows
#1229 Can't build from source - python error
#1228 Missing or incorrect tooltips and context-sensitive help
#1224 asciitf2vtf dumps core
Complete descriptions of these and other bugs are available here.
New Features:
High-resolution cartographic maps: Vaporgui now supports import and display of arbitrary, multi-resolution cartographic maps that follow the OSGeo Tiled Mapping Service conventions. Two 2km resolution maps are included in the VAPOR distribution (NASA Big Blue Marble and Natural Earth). 500m Big Blue Marble imagery may be download separately from the VAPOR web site.
Improved cartographic maps for ocean data: To improve the use of Cartographic maps with ocean data all ocean maps include a transparent alpha channel for pixels over ocean regions. The alpha channel can be optionally honored or ignored.
3D Stereo Rendering: Scenes can now be shown in 3D using quad-buffered stereo rendering. This can be viewed in real-time as users navigate their data in VAPOR, and it can be used to create post-hoc 3D movies.
Color Barbs by Variable: Barbs can now be colored based on the values of a variable at each barb position, using a color map in the Barbs panel.
Contour lines following terrain: Contour lines for 2D variables can be offset vertically by a terrain-height variable, to display the contours on the terrain surface.
Contour line labeling options:
Three methods of labeling contour lines in the scene are supported: Billboard (label displayed in front of the scene), In-Plane (displayed on the plane of the contours), and 3D (positioned in the 3D scene, facing the camera).
Parallel data conversion scripts:
Python scrips have been written that can run VAPOR’s data conversion routines in parallel on compute clusters. This can provide a linear speed-up with the number of cores that a user wishes to use for their conversion.
Quadratic interpolation of vertical coordinates:
Users now have the option to do a quadratic interpolation of the vertical coordinate in layered and stretched grid types.
Python documentation:
Additional Python documentation has been provided in the Python variable editor.
MPAS and ICON support: Prototype data importers have been added to vaporgui for MPAS and ICON model data. Outputs from these models can now be visualized with VAPOR.
Supplementary Changes:
Key-framing simplifications: When the time step changes between two key frames, by default the key frames will be synchronized with the time steps so that the time step will advance by one each frame. Also, automatically generated key frames will no longer be included in the list of key frames displayed in the user interface.
Sampling rate support for DVR and Isosurface renderers: Users can specify a spatial sampling rate for these renderers to improve the image quality when data values are not spatially uniform.
Default 16-bit DVR and Isosurface rendering: By default, these renderers operate in 16 (instead of 8) bits, improving the default rendering quality.
Keyboard controls: VAPOR’s orientation can now be manipulated with keyboard arrow keys for rotation (add ‘shift’ for translation) and the +/- keys for zooming in/out.
Bugs fixed:
The following significant bugs were fixed in the current release.
1201: Lower accuracy data not used.
1177: Support user setting of ISO and DVR sampling rate.
1153: Usability problems with key framing.
1145: Contour line annotation incorrect.
1144: Spinning cursor delayed.
For a list of all fixed bugs please visit the vapor sourceforge bug list.
1154 vaporgui error message: libexpat.so.0: failed to map segment from shared object: Cannot allocate memory
1147 Libraries on Mac installed with @executable_path
1146 vaporgui problems with libexpat.so on Ubuntu 12.04
This version includes the following new features:
Support for GRIB data: Data in GRIB (GRIdded Binary) format can now be visualized in VAPOR. Currently this capability has only been tested with GRIMs (Global and Regional Integrated Modeling System) model data, as described in the GRIMs User’s Manual. This model data can be converted to a VDC using VDCWizard, or converted using two new command-line apps “gribvdfcreate” and “grib2vdf. GRIB data can also be imported directly into vaporgui, using the Data menu.
Support for CAM (Community Atmospheric Model): VAPOR can now visualize CAM model data. NetCDF output Data files can be converted to a VDC using VDCWizard, or converted using two new command-line apps “camvdfcreate” and “cam2vdf”. CAM data can also be imported directly into vaporgui using the Data menu.
Contour lines: Vapor now supports displaying contour lines for 2D and 3D variables, using the “Contour” tab in vaporgui. Contour lines of 2D variables are displayed in a horizontal planar plot. Contour lines of 3D variables are displayed in a planar slice that can be arbitrarily oriented and positioned in the 3D scene. Users may specify any number of isovalues and the contour lines can be annotated or colored to indicate the associated isovalue.
Latitude/Longitude axis annotation: The horizontal axes of geo-referenced data can now be annotated with latitude and longitude, for most supported map projections. This option is set in the axis annotation properties in the visualizer features panel.
Multiple color bars: Vaporgui now supports display of up to 6 color bars in the 3D visualizer; one color bar for each renderer that has a transfer function. A title can be displayed below each color bar.
In addition, users will note the following changes:
Windows PATH and PYTHONHOME usage: On Windows, VAPOR will no longer set the system PATH or the PYTHONHOME. Users of the VAPOR command-line apps will need to perform a vapor-setup operation in the current terminal. This will not affect users of vaporgui or VDCWizard.
Copying box extents. The Region panel includes a "Copy Box Extents" option that allows users to copy the box extents from one renderer instance to another.
Color map by field magnitude: The flow line color-mapping by field magnitude option is no longer supported. Instead, to color lines by flow magnitude, users should define a 3D derived variable equal to field magnitude, and specify this derived variable as the color/opacity mapped entity.
The following list includes the more significant bugs that have been fixed in this release. For more details and a list of all fixed bugs please visit the vapor sourceforge bug list.
1142 Can't process MOM/POP/ROMS files with varying time steps
1140 momvdfcreate generates wrong extents with some POP data
1131 Core dump during keyframe animation
1121 VDF translators always output all wavelet coefficients even if not wanted.
1117 Seed bias = 16 gets strange results
1106 Zero 2D variable in restored session
1102 GRIB conversion fails on Mac OSX 10.10 (Yosemite)
1101 Isosurface crash when reading 2.3 session file
1098 Many repetitions in image capture sequence
1024 Silent setting environment variables on Windows
A number of high severity known bugs is listed below. For more details and a list of all known bugs please visit the vapor sourceforge bug list .
1141 VDCWizard crashes on Yosemite after data conversion completed
1128 User extents do not update on 2D, Image, and Probe tabs.
1127 TF Domain Bounds and Isocontrol bounds should be obtained on-demand
1041 NCL geotiff creation fails
1009 Nudging does not work on Mac
918 Isosurface missing in ROMS scene
Support for GRIB data: Data in GRIB (GRIdded Binary) format can now be visualized in VAPOR. Currently this capability has only been tested with GRIMs (Global and Regional Integrated Modeling System) model data, as described in the GRIMs User’s Manual. This model data can be converted to a VDC using VDCWizard, or converted using two new command-line apps “gribvdfcreate” and “grib2vdf. GRIB data can also be imported directly into vaporgui, using the Data menu.
Support for CAM (Community Atmospheric Model): VAPOR can now visualize CAM model data. NetCDF output Data files can be converted to a VDC using VDCWizard, or converted using two new command-line apps “camvdfcreate” and “cam2vdf”. CAM data can also be imported directly into vaporgui using the Data menu.
Contour lines: Vapor now supports displaying contour lines for 2D and 3D variables, using the “Contour” tab in vaporgui. Contour lines of 2D variables are displayed in a horizontal planar plot. Contour lines of 3D variables are displayed in a planar slice that can be arbitrarily oriented and positioned in the 3D scene. Users may specify any number of isovalues and the contour lines can be annotated or colored to indicate the associated isovalue.
Latitude/Longitude axis annotation: The horizontal axes of geo-referenced data can now be annotated with latitude and longitude, for most supported map projections. This option is set in the axis annotation properties in the visualizer features panel.
Multiple color bars: Vaporgui now supports display of up to 6 color bars in the 3D visualizer; one color bar for each renderer that has a transfer function. A title can be displayed below each color bar.
The following bugs have been fixed in this release (Full descriptions can be found on the Sourceforge Web site):
1069 Memory leak in pathline integration using color mapping
1058 romsvdfcreate seg faults when no 3D variables are present
1043 .user_prefs not identified in file save on OSX 10.9
1040 Implicit time steps do not work
1036 Session does not save barbs variables
1035 Inconsistent height values with models at higher elevation
1032 LD_LIBRARY_PATH usage on Linux
1030 ncdf2vdf outputs wrong data range for 2D Data
1028 Memory leaks occurring in ROMS, WRF, and CAM converters.
1027 vdfcreate truncates domain extents precision
1026 libpng12 dependency
1019 crash when copying renderer instances
1012 assimp library version 3.0 not supported
1010 Color edit has no effect until another control point clicked
1005 Cannot read installed images or colormaps on OSX 10.9
999 wrf2geotiff.ncl is broken
A number of high severity known bugs is listed below. For more details and a list of all known bugs please visit the vapor sourceforge bug list .
1079 Import GRIB file crashes (when fed a bad file).
1024 Silent setting environment variables on Windows
1102 GRIB conversion fails on Mac OSX 10.10 (Yosemite)
1101 Isosurface crash when reading 2.3 session file
918 Isosurface missing in ROMS scene
vdcwizard is a brand new tool in VAPOR, providing GUI for simplified data conversion from foreign data to VAPOR’s VDC data format. Currently, vdcwizard is capable of converting data from ROMS, MOM4, POP, and WRF models to the VDC. For more information see https://www.vapor.ucar.edu/docs/vapor-data-preparation/data-conversion-vdcwizard
Direct model import: vaporgui can now read ROMS, MOM4, and POP model outputs directly, without prior conversion to a VAPOR Data Collection.
Parallel data translation: Support for parallel (threaded) data conversion on Windows has been added (Linux and Mac OS X were previously support), speeding the process of reading and writing VDC data
CDO post-processed data support: Supported model outputs (e.g ROMS, MOM, POP) that have been post-processed with the Climate Data Operators (CDO) can now read by VAPOR.
On-line documentation: VAPOR’s on-line help has been extended and now references the VAPOR documentation web site.
Simplified LOD and refinement control: The vaporgui user interface now offers a simplified, single-parameter control of the “refinement” level and level-of-detail (LOD).
Discrete color maps: Discrete color maps are available for all Transfer Function editors.
Flow panel TF: The Flow visualizer tab Transfer Function editor has been made consistent with TF editors used in all other visualization tabs.
Automated release notification:vaporgui will now alert users when new releases of the VAPOR package become available.
A summary of fixed bugs is below. For details see https://sourceforge.net/p/vapor/bugs
1002 asset when resizing probe
1001 roms2vdf dumps core when -var option is used
1000 'Barb Length Scale Factor' not being initialzed properly
996 vdcwizard does not allow storage block > 99
992 Text in the vaporgui tabs gets clipped in certain areas
991 Wrong region centering
990 animation stops in the middle
987 vaporgui model renderer ignores color
983 PYTHONVERSION incorrect on Windows
981 lat-lon extents not displayed by wrfvdfcreate or vdcwizard
975 vdcwizard: can't change start time step
970 Python imports not behaving as expected
952 Probe histogram and smoothing incorrect
946 Probe crop too small
923 Animation hangs after session 2 days old
920 Environment Variables in Windows Source Installation
907 lat-lon projections not support in WRF data
743 core dump when rendering over X11
vdcwizard is a brand new tool in VAPOR, providing GUI for simplified data conversion from foreign data to VAPOR’s VDC data format. Currently, vdcwizard is capable of converting data from ROMS, MOM4, POP, and WRF models to the VDC. For more information see https://www.vapor.ucar.edu/docs/vapor-data-preparation/data-conversion-vdcwizard
Direct model import: vaporgui can now read ROMS, MOM4, and POP model outputs directly, without prior conversion to a VAPOR Data Collection.
Parallel data translation: Support for parallel (threaded) data conversion on Windows has been added (Linux and Mac OS X were previously support), speeding the process of reading and writing VDC data
CDO post-processed data support: Supported model outputs (e.g ROMS, MOM, POP) that have been post-processed with the Climate Data Operators (CDO) can now read by VAPOR.
On-line documentation: VAPOR’s on-line help has been extended and now references the VAPOR documentation web site.
Simplified LOD and refinement control: The vaporgui user interface now offers a simplified, single-parameter control of the “refinement” level and level-of-detail (LOD).
Discrete color maps: Discrete color maps are available for all Transfer Function editors.
Flow panel TF: The Flow visualizer tab Transfer Function editor has been made consistent with TF editors used in all other visualization tabs.
Automated release notification:vaporgui will now alert users when new releases of the VAPOR package become available.
A summary of fixed bugs is below. For details see https://sourceforge.net/p/vapor/bugs
969 can't delete python variables
967 Can't save or load user prefs on Mac
963 flow display lists broken
962 Error message after saving seeds
961 Core dump saving seeds
956 vaporgui model tab: instance confusion
947 OpenGL errors on Mac
934 conversion tools abort with invalide compression ratios
928 Vapor crashes on direct data importation
927 Mouse scroll whell disrupting visualizer settings
922 IDL VDC_VAROPENWRITE broken under 2.2.4
921 keyframing doesn't work with "zoom"
919 ncdfvdfcreate ignores extents
917 incorrect default in romsvdfcreate,momvdfcreate help
916 Unsteady timestep display interval change does not force rerender
914 keyframing gets confused about time step
910 Pre-integrated volume rendering is broken
890 ncdfvdfcreate doesn't handle implicit time steps
882 Unable to complete image capture sequence rendering
739 Users need other python modules to work with Vapor
712 Unsteady flow opacity mapping incorrect
448 flow lines penetrate terrain
Parallel data translation on Windows: VDC data translation (reading and writing) has been parallelized (threaded) on Windows platforms. Windows platforms with multiple cores can now take advantage of parallel data translation by VAPOR's command line data translation utilities (e.g. raw2vdf, wrf2vdf, etc.) and vaporgui. Note, parallel data transformation already exists on Mac and Linux platforms.
911 Vertical coordinates incorrect with some MOM and POP data sets
906 Derived variables don't work when resume saved session
905 Can't display multiple fields with 2D only data sets
904 2D visualizer can't load transfer functions if no 3D data are present
903 Mixed up variable names from session files
897 Program crash due to steady field parameters
893 Flow Image capture crash
891 TF window does not refresh
884 Can't overlay data in
876 IDL broken on Windows
863 Apply to terrain does not correctly handle missing values
861 transparency slider not updated
command line translators: New versions of the command line data translators romsvdfcreate, roms2vdf, momvdfcreate, mom2vdf, and ncdf2vdf exist. The new commands provide a much simplified interface for data conversion. An entirely new command, ncdfvdfcreate, was written to facilitate conversion of generic netCDF data sets.
64bit Windows support: Binary installers supporting 64bit addressing are now available for Windows platforms.
vdfcp.pl command: vdfcp.pl is a command line utility that facilitates copying all, or a selected subset, of a VDC data set from one location to another
3613859 Clipping planes too tight
3613851 viz features not changeable
3613230 Infinite recursion in VariableExists for derived variables
3613073 Many errors when time step missing
3613069 no elev grid refresh when change height variable
3611854 Unwritable default directories
3610027 vapor.BOUNDS is wrong
3609065 linux installer omits libexpat
3608833 tab changes very slow with big data
3608754 vector_rotate broken
3607921 path handling under windows
3607919 spherical volume rendering broken
3607074 python conflict
3605413 In layered datasets mouse drags show like not layered
3605410 square box visibile in isosurface
3603624 Data range incorrectly calculated
3602835 probe orientation is wrong
3593358 incorrect derivative operators for WRF, MOM, ROMS
Animation key framing support: Complex animaitons can now easily be created through key framing: after specificying a small number of viewpoints a smooth animation can be constructed by automatically filling in the frames in between.
Missing data values: The VAPOR Data Collection, translation tools, and vaporgui now provide first class support for missing data values: grid points for which no data value is defined are permitted and correctly handled by all VAPOR data operators.
Stretched grids: In addition to regular and terrain following grids, VAPOR now supports stretched (rectilinear) grids.
Ocean data support: Translators for converting from MOM, POP, and ROMS ocean models to the VAPOR VDC are now available.
Improved layered grid support: Layered (terrain following) grids, such as those generated by WRF, are no longer resampled to a uniform grid. vaporgui now operates directly on layered grids, which can significantly improved rendering quality.
Improved probe rendering quality: The probe tool now supports optional linear interpolation (previously only nearest neighbor interpolation was supported).
Display window size user preference: The visualizer display window width and height may now be specified as a user preference.
3602862 random seeds change often
3602449 Data Cache fills up
3601808 vdfedit does not insert time-varying extents
3601282 assert in romsvdfcreate
3601269 errors generating unsteady seeds
3601195 Barb nudge broken
3601158 degenerate flow lines from nonrandom rake
3601156 Missing values not in 2D derived variable
3601111 cvs library version produce not valid vdc
3601034 Missing values in probe are not transparent
3600297 Error when depth peeling isosurface on NVS5200
3600262 keyframe speed wrong in stretched domains
3599886 keyframing prevents traditional animation
3599505 random seeds not uniform
3599023 speckles intermittently in ISO
3598663 problems mapping variable onto isosurface
3598087 problems with frame capture
3597326 Can't enable diagnostic messages
3597310 animation max wait is useless
3595657 Cruft in unsteady flow
3595025 problems after disabling depth peeling
3594997 image capture fails with depth peeling
3594996 another depth peeling blending problem
3594994 depth peeling blending problem
3594833 missing values are visualized with renderer, iso, probe ...
3594362 no color by variable with unsteady flow
3593743 deriv operators fail with short dimensions
3593359 Need windows diagnostics
3592259 DVR of MOM4 has wrong color map at top refinement
3592131 DVR of ROMS displays missing values
3590415 2.2.0.RC1 (cvs version) do not compile
3590130 Python problems on windows
3588020 windows version crashes on exit
3579789 ELEVATION variable may contain missing data values
3579524 flow bias doesn't work with ocean data
A partial list of "high severity" known bugs appears below. For more details and a list of all known bugs please visit the sourceforge web site
3602451 Iso render crashes mac
3590735 Some POP datasets cannot be converted with mom2vdf
3603218 Crash with no error message with low-end intel card
3602835 probe orientation is wrong
3602726 DVR does not update colors
3602724 WRF results inconsistent with guide
3601015 Windows netcdf 4.2 problem with reading large files
Two new visualization capabilities:
Several improvements have been made to the VAPOR Python support, including a Python module that calculates several WRF-related derived variables.
Several of the NCL colormaps most useful to WRF have been provided as transfer functions.
VAPOR provides a number of built-in global terrain images, enabling users to avoid generation of custom terrain images.
Support is provided for reading compressed tiff images.
A new loadable shader capability allows advanced users to customize or write their own GPU shaders
Support is provided for rotated lat-lon (Cassini) transform, as used by WRF-ARW version 3.
The command-line apps ncdf2vdf and wrf2vdf now support wavelet compression using the VDC2 data representation.
Users can specify the order and showing or hiding of tabs in the user interface as a user preference.
The user interface style is now customized to each platform.
Probe and Flow panels were rearranged for improved usability, with support for hiding and showing subpanels
The performance of the VDC2 data encoder has been improved on platforms with multiple cores.
Numerous instability problems have been fixed. Other fixed bugs include:
3341880 unreliable temporal animation
3300606 Preferences should be forwards compatible
3392570 VDC2 data not machine portable
3447289 Python variables don’t use lowered resolution
3290522 Wrong data bounds with derived variables
Known Bugs and Workarounds:
2210024: Unsteady flow integration results are incorrect if the domain is moving in time.
1903402: Transparency does not work correctly when displaying multiple partially transparent surfaces in the same scene.
3123643: The VAPOR IDL API does not work with VDC2 data. Users should use Python to calculate derived variables on VDC2 data.
3308059: Existing derived variables do not recalculate when the vertical reinterpolation rate changes.
A full list of known bugs and feature requests can be found on the SourceForge VAPOR project web site
VAPOR’s spin animation is now an option that is set in the visualizer features panel. By default, spin animation is turned off.
A new Python method, vapor.GetFullVDCDims(refinement), enables Python programs to find the dimensions of the current VDC.
The Python finite-difference derivative operator, deriv_findiff(), now has its dimension ordering in agreement with the VAPOR user coordinates.
This release fixes several bugs:
3131758: python memory leak was fixed
3131860: region extents no longer jump when slider is released
3132554, 3136874: various problems with the use of local/global parameters were fixed
3150390: data bounds are now evaluated lazily
3181360: multiple isosurfaces no longer cause a core dump
3151444: multiple animating windows are now synchronized
3124772, 3126751: When changes are made to a Python startup script, the previously cached derived variable data is now re-calculated.
3113074: At the end of an animation sequence, the front tab now updates to display the data for the final time step.
3117797: Only the currently available compression levels are indicated in the gui. When lower compression is not available, a higher compression may be used (if specified in user preferences).
3158384: The encoded significance map used by VDC2 data contained incomplete information. Note: VDC2 data generated by VAPOR 2.0.2 and later will not be readable by pre 2.0.2 releases.
Known Bugs and Workarounds:
2210024: Unsteady flow integration results are incorrect if the domain is moving in time.
1903402: Transparency does not work correctly when displaying multiple partially transparent surfaces in the same scene.
3123881: Jpeg image files cannot be applied to the terrain in the Visualizer Feature panel. Users should use the Image panel, which supports applying Tiff images to the terrain.
3123643: The VAPOR IDL API does not work with VDC2 data. Users should use Python to calculate derived variables on VDC2 data.
3199940 When using a nonrandom rake that extends below the terrain, a crash can result. This can be prevented by setting flow field variables to small nonzero values below the terrain (in the visualizer features panel)
3199770 On windows when converting very large WRF output files, wrf2vdf can crash, if the number of timesteps times the number of variables in one file is greater than 10000. This crash can be avoided by converting a subset of the variables at one time.
The WRF Cassini projection is not supported. Results will be displayed in user's coordinate system (meters).
A full list of known bugs and feature requests can be found on the SourceForge VAPOR project web site:
Python support for calculating derived variables: VAPOR provides a built-in Python calculation engine, enabling users to define and visualize new variables interactively during data analysis. Simple Python expressions or larger programs can be used to derive new variables from existing variables. VAPOR data is handled as Numpy (numerical python) arrays. The derived variables can be visualized like other variables in the VDC.
Support for data compression: VAPOR now supports an improved biorthogonal-wavelet based data model with support for data compression as well as refinement level control. This model is called VDC2 for second-generation Vapor Data Collection. The VDC2 model is designed to improve performance on extremely large datasets. Data that is converted to a VDC2 model can be retrieved at a user-specified compression level and refinement level. Compression ratios up to 500:1 can be used with minimal image degradation, enabling rapid viewing of massive datasets by reducing the I/O requirement.
Direct import of WRF-ARW output files: Vapor now supports direct importing of wrfout files, bypassing the VAPOR data conversion step. This is especially useful to get a quick visualization of small WRF- ARW datasets.
New UI implementation: VAPOR has updated the user interface to Qt version 4.6. Users will notice a somewhat changed appearance (especially on Mac), but the functionality of the UI should not be significantly different.
Faster rendering of flow lines: VAPOR’s flow line rendering has been accelerated, so that display of many thousands of flow lines can now be performed interactively.
Native Mac OSX support for both 32 and 64 bit systems: VAPOR version 2.0.0 now runs as a native Mac application. There is a 64-bit as well as a 32-bit version. The user interface is similar to other Mac apps, and does not rely on X11 any more.
The VAPOR IDL API now uses IDL version 7.
VAPOR now provides limited support for display of semi-transparent surfaces. Multiple two-dimensional semi-transparent surfaces (Probe, 2DData, and Image) will correctly display in a scene provided:
This release fixes several bugs:
2877268: Typing in a new region size no longer changes its position.
2912903: Improved rendering of isosurfaces with volume rendering. Opaque isosurfaces are properly rendered with a DVR in the same visualizer.
2961337: The text in the color scale no longer gets cropped on Linux platforms.
2989430: The elevation of a 2D surface can now be typed into the gui (with both Image and 2D renderers) and the specified elevation is correctly used.
2991491: DVR pre-integration can now be used in the same window as a color scale.
A complete list of bugs fixed by this release may be found on the SourceForge VAPOR project web site
Unsteady flow integration results are incorrect if the domain is moving in time.
Transparency does not work correctly when displaying multiple partially transparent surfaces in the same scene.
Jpeg image files cannot be applied to the terrain in the Visualizer Feature panel. Users should use the Image panel, which supports applying Tiff images to the terrain.
VDC merge crashes if there are Python-derived variables in the current VDC.
The VAPOR IDL API does not work with VDC2 data. Users should use Python to calculate derived variables on VDC2 data.
At the end of an animation sequence, the front tab does not update to display the data for the final time step. Users should click the “stop” button to see the correct settings and images in the front tab.
A full list of known bugs can be found on the SourceForge VAPOR project web site.
