|
From: | Patrick Kreissl |
Subject: | Re: Glitch in simple pore constraint visualization with openGL |
Date: | Sun, 24 Oct 2021 15:41:54 +0200 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.11.0 |
Hi Le,
thank you for your Email.
What you observe is a known problem on some hardware including
Mac.
The OpenGl-visualizer uses clip planes to cap objects at the box
boundaries. Also, some Constraint-visualizations, such as the
SimplePore, require additional clip planes to render the object.
In general it is possible to request the additionally needed clip
planes from OpenGL. This works on most Linux machines. However,
unfortunately, on Mac there is an upper limit for the clip planes
allowed (6). Shapes that would need additional clipping will in
this case not be clipped on one face of the box.
I suppose that choosing the simple pore in x direction, you are
lucky as so far as that the pore is not clipped on one of the
boundaries that are not crossed by the Shape, so visualization
seemingly works -- I imagine that having an additional
ShapeBasedConstraint that crosses one of these box faces would
lead to a similar 'glitch'.
What happened when not adding particles is not clear to me,
however.
On my machine that works quite fine, but I am a Linux user...
Pleate note that the online visualizer is not meant to produce
high quality renderings, but rather to help you debug your system.
Still, I guess you can mitigate the issue by staying with the
working choice for the direction of the SimplePore or, if possible
for you, switch to a Linux machine for online visualized
debugging. For high quality rendering, you could consider writing
out *.vtk files for your simulation and animate the data using
visualization software such as ParaView.
Best wishes,
Patrick Kreissl
Hi,
Recently I encounter a glitch while using espresso 4.1 and dev on my M1 Mac (python 3.9.1), not sure if you have the same issue or just me. This happens when you place the the simple pore constraint in y- or z-direction, the wall is oddly extended in one direction for some reason, as shown below. But it is alright when you change this to x-direction (axis=[1.0, 0.0, 0.0]).
Also, when you add no particle, the visualization of the simplepore constraints is very strange, see the third figure.
I tested this with the script “visualization_constraints.py” in the SAMPLES directory.
1. Simple pore placed in y-and z-direction2. placed in x direction
3. simple pore without particlesCheersLe
[Prev in Thread] | Current Thread | [Next in Thread] |