Release Notes

Version 2.4.23 (November 2022)

Added a helper class ViewPolyline to display a user-defined polyline. See help(ViewPolyline).

The ViewLines (view user-defined lines) is deprecated. Please make use of ViewPolyline instead .

Version 2.4.22 (June 2022)

Added the ViewLines class to view user-defined lines in a model.

Version 2.4.21

Added --mesa option to baspl++ command line. Was there previously…

Added mesa target to make.

Version 2.4.20

The baspl++ desktop icon is located in <prefix>/share/applications/baspl++.desktop and can be copied to your desktop. Ubuntu example:

cp <prefix>/share/applications/baspl++.desktop $HOME/Desktop/baspl++.desktop

The utility class ViewPoints for viewing user-defined points has been added. Example of viewing a point a position (-1.0, 10., 0.0) for model model:

pts = ViewPoints(model)
pts.add((-1.0, 10., 0.0), color=(1.0, 0.0, 1.0), size=7., text='my point')

CentOS8 could not find MemCom libraries from Python2. Fixed.

CentOS8 could not find gtk python path. Fixed.

The build process has been updated (configure files for Python 2.7 and for batch mode).

Version 2.4.19

Updated user manual. Release notes in separate document.

Version 2.4.18

baspl++ still works with Python2 only.

The batch mode (--batch command line option) has been reactivated (it was broken since version 1.x).

The subcase argument to the Field object has been added. It is equivalent to the mode argument.

Version 2.4.17

flc videos have been deactivated (make use of mp4 instead).

Version 2.4.16 (April 2018)

Video creation: Use png’s instead of ppm’s.

Improved display of ViewLINC to distinguish between master and slave nodes.

Version 2.4.15 (December 2017)

Support of Q36, Q5X5, Q2X2 element.

Version 2.4.14 (November 2017)

Support CSDA coordinates and fields.

Version 2.4.13 (october 2016)

Use new implementation for detecting coinciding nodes, because of memory run-away problems with b2aabb_intersection. The new implementation is faster (but not very fast), has a lower memory footprint, and seems more robust.

Versions older than 2.4.13

See repository.

Known Problems and Bugs

baspl++ still works with Python2 only. Python3 scripts submitted to baspl++ may not work if the scripts are not compatible with Python2.

Vectors at Integration Points (Sampling Points)

Vectors at integration points (Sampling Points), such as heat flow vectors at the element integration points, cannot be plotted.

Grid

Plotting a grid can fail if the model is a pure 2-dimensional model defined in the x-y-plane.

baspl++ still works with Python2 only. Python3 scripts submitted to baspl++ may not work if the scripts are not compatible with Python2.

Contour Plots

The Mesa OpenGL libraries (driver) can lead to unwanted effects with the “Gouraud” and “step texture” shading. Remedies:

  • Select smooth color map texture mode in menu or specify in the script for the NPart concerned (here: p):

    p.contour.colmap.texture.mode='smooth'
    
  • Make use of Intel or NVIDIA drivers (see below).

baspl++ and the NVIDIA Driver

If no other driver is installed baspl++ selects the default OpenGL libraries (Mesa). This can lead to unwanted effects with the “Gouraud” and “step texture” shading. To make use of Intel or NVIDIA drivers, please start baspl++ and check:


_images/nvidia.baspl.png


The “openglinfo” tab will tell you about the OpenGL version used in baspl++:


_images/nvidia.openglinfo.png


If “Vendor” or “Renderer” or “Version” says “Mesa”, then the Gouraud and “step texture” shading will not work property.

If you got a NVIDIA graphics hardware on your device you should activate it (Ubuntu: see https://www.linuxbabe.com/desktop-linux/switch-intel-nvidia-graphics-card-ubuntu). The shell comand lspci

lspci -k | grep -A 2 -i "VGA"

will then show (NVIDIA example):

03:00.0 VGA compatible controller: NVIDIA Corporation GP108 [GeForce GT 1030] (rev a1)
        Subsystem: Gigabyte Technology Co., Ltd GP108 [GeForce GT 1030]
        Kernel driver in use: nvidia

The NVIDIA settings can be checked with

nvidia-settings


_images/nvidia.nvidia.png