ParaView Reading in Data
This tutorial will show you the basics of Loading Data in ParaView, along with the basic usage of the Display, Parameter, and Information tabs.
First launch ParaView and then select File->Open Data from the main menu. This will open the "Open ParaView File" panel.
This gives the user the ability to load data in many different formats. The user will need to select the appropriate format for the data to load it, or allow ParaView to guess the format based on the extension. Allowing ParaView to guess the format can cause problems. For example, both Plot3D and XMol have the same *.xyz extension. ParaView chooses Plot3D, which may or may not be appropriate format. If the file selected does not match any known format extension, then a pop-up will be displayed to allow you to select the specific reader to use.
Once the reader is known, the Parameter panel will be displayed. The specifics of this panel will vary depending on the reader. Also, some readers will load the file automatically while some will highlight the "Accept" button, waiting for you to make final selections, such as specific fields or formatting. In this case, a legacy VTK file is being loaded. Click on the file, and then click on the Open button. This will automatically load the file into ParaView. Once the file is loaded, the user can interact with the data by performing transformations such as rotating, translating, and zooming in the 3-D View Pane.
- Transform Data
- Rotate with left mouse button.
- Translate with middle mouse button.
- Zoom with right mouse button.
Below the Selection Pane, there are three tabs: Parameters, Display, and Information. First select the data in the Selection Pane to operate on it. It will be highlighted yellow when this occurs.
The Parameters Tab is used for browsing and loading data, especially time series data sets. (Be sure to refer to the notes relating to Transient (Time Varying) Data if you plan to use time-varying data).
From the Parameter tab you have a few basic settings (Some file readers have significantly more.)
- Accept - This button will turn green to indicate something has changed and now needs to be accepted.
- Reset - This will reset all the UI elements back to the last "Accepted" state.
- Delete - This will remove the reader from the interface, freeing the memory. This option is only available if no other visualization elements are dependent on it. (i.e., you cannot delete a file reader if an Isosurface has been extracted from it and still exists).
You can also change the filename here, so long as the new file is of the same format. By using the "Time steps" button you can specify multiple files to load as a time-series.
The Display Tab is used to modify the display of the data. It is divided into four sections: View, Color, Display Style, and Actor Control.
In the view section, there are five options.
- Data - When checked, this object is visible. This has the same effect as clicking on the "Eye" in the pipeline Selection Window.
- Scalar Bar - This controls whether a scalar bar (legend, color bar, etc.) will be visible in the window. More than one item can have a scalar bar visible at a time, but each item can only have one scalar bar.
- Cube Axes - When this is checked a rectangular bounding box will be constructed around the object, and the axis names and dimensions will be overlaid in the 3-D Window.
- Label Point Ids - When checked, an Id number will be shown at every vertex. This is only useful for very small data sets.
- Set View to Data - When clicked, the 3-D View will be reset to show the selected object. This is helpful when the viewport is accidentally rotated or panned in a direction that causes the data to slide off the edge of the screen.
The Color section of the Display tab allows the user to change the color of the data being displayed. The first option allows the user to color the data by a property (solid color), or data field. If the user chooses to color by property, then they can also choose the color by clicking on the Actor Color button. This will open the Color pop-up window. The Color window allows the user to choose either one of several basic colors, or to create a custom color using hue, saturation, and luminence values, or red, green, blue values.
More interesting is to color the data by a scalar or vector. Clicking on the dropdown next to Color by will give a list of all fields in the data acceptable for coloring. The list will detail whether it is a Point (Vertex) attribute or Cell (face) attribute, and how many components are in the field. Simply selecting one will load a default color map across the full range of the field. The color map can be changed by clicking on the Edit Color Map button (Refer to Coloring Data in ParaView for more Information). The Range of the color map can be reset to the bounds of the selected data field by clicking the Reset Range button.
Typically, Map Scalars will be disabled. In the rare event that a three-component Unsigned-Char (8-bit) vector field is chosen, this checkbox will be available. When "Map Scalars" is checked (the default), then the selected field will be mapped through your color map. When it is unchecked, however, the three-component unsigned-char field will be used directly for the color. This is helpful for data sets with pre-generated colors (Images like JPG or BMP, or PLY models with colors). Typically this option is not used.
The Interpolate Colors checkbox is typically unchecked and controls how colors are mapped between points in the data. By default (unchecked) colors will be mapped through OpenGL textures to ensure that the color map is preserved between data points. If checked, then the colors will be RGB interpolated between vertices. This is mostly important when dealing with non-continuous (banded, contoured, striped) color maps or rapidly changing data. The default of unchecked is typically the proper setting.
In the third section of the Display Tab, the user can change the appearance of the data by choosing to be displayed in Outline mode, in Wireframe mode, in Shaded mode, as Points, or in some cases, it allows Volume Rendering. Volume Rendering is allowed only for unstructured data sets, and will not be visible if the data set does not meet this criteria.
By setting the Interpolation, you can control if the colors are smoothly interpolated across the faces (Gouraud) or solid (Flag). Solid offers a slight performance improvement over extremely large data sets.
Also, the width of lines in Wireframe and the size of Points in Point-cloud can be set, along with the typeface size of Point ID's.
In the fourth and final section of the Display Tab, the user can translate, scale, change the orientation, or change the opacity of the data.
It is important to note that these settings will not alter the coordinates of the data, but simply move it about in space. This is important when probing the data or working with multiple-stage pipelines. When comparing the visualizations of two different data sets or two different algorithms, if they have been moved with these Actor controls, then the results will not align, even if they occupy the same XYZ coordinate space.
The Information Tab provides the user with information about the data set. An example is shown below:
The Statistics section includes the type of data read in, the number of cells, the number of points, and how much memory the data set consumes in memory.
The Data Arrays section includes information about the data fields in the data set, including name, whether it's Point or Cell centered data, the data type and number of components, and the lower and upper bounds of each component.
The Bounds section includes the lower and upper bounds of the X, Y, and Z components.
If the data set is Rectilinear, then another section similar to Bounds will appear called Extents, showing the upper and lower bounds of the I, J, and K extents.