The kview and kubes programmes can be used to look at two cubes at the same time. They do not volume render the data, but display the data cube along the three principle planes (typically and ). The programmes know about two data sets, one is called Set 1, the other Set 2. The two datasets may be combined in various ways, and allow you to step through the cubes and compare the two data sets. One can also loop over a certain range of channels or position-velocity diagrams.
The original idea behind kubes was to compare a data cube with a model cube and it is used for this in the modeling software, but of course kview and kubes can be used for any one or two data sets. The data sets do not have to be cubes even, they can also be single images, or one can be a cube and the other a single frame. One application of this kind is to overlay the contours of the channels of an HI data cube over an optical image from the digital sky survey. If you are using kubes, you will need to use skyve in aips to regrid to a proper coordinate system) or use skyview to download the image. If you use kview you don't have to worry about regridding.
Another application of kubes is that one uses xray to make a movie of a data cube and one of a model cube, and use kubes to look at the rendered cubes simultaneously (side by side), provided these are not made with the Hot Gas Substances or the Hot Gas Continuous shader. This can be very instructive.
kview should be started with:
See the Karma User Manual section on kview.
You don't have to worry about available colours on the screen: if there are insufficient colours (perhaps because you have a WWW browser or some other visualisation tools running), kview (and the rest of the standard visualisation tools) will automatically create a new colourmap. On most hardware this will mean that you will get colour flashing when you move from one window the another.
kubes should be started with:
kubes [-ncolours n] [-set1 <name>] [-set2 <name>] [-test]
The optional switch -ncolours n sets the number of colours kubes allocates. The default is -ncolours 200, but if you are also running another program that uses many colours, you will have to use a smaller value.
Specifying -set1 and/or -set2 will load the file name into set1 resp. set2 at startup. name should be the full name of the file with extention (e.g. .kf or .fits)
The -test switch enables features that are not yet completely implemented, but work to some extent. At the moment -test enables a display mode very similar to kslice_3d (mode XYZ).
Two Data Channels
As stated, kview and kubes know about two data sets. For loading data in either one of them, click (left) on Set 1 (or Data) or on Set 2 (or Model (kubes only)), depending where you want to load the data. kview and kubes have to go through some bureaucracy with the data, like checking units, determining the min and max etc., so loading a cube is not instantaneous.
There will be some error messages printed on the screen if you use kubes. Ignore these. They come from the karma library and are usually harmless.
kview Display Modes
See the Karma User Manual section on kview.
kubes Display modes
There are a number of ways to display the data sets. The first set of modes controls how the data is shown, the second set of modes determines what you see.
Of course the two sets of modes work independently. To select a mode, click (right) on Display modes and choose one.
For the contourplotting, kubes works in physical coordinates, ie. it looks in the header of the data and uses the reference position and the gridspacings. For the moment only NCP (i.e. ATCA), SIN and TAN projection is handled properly. This means that you can display one image and overlay contours of another and the gridspacings do not have to be the same (e.g. HI contours on top of an optical image), as long as the coordinate systems of both data sets are defined properly. Note that the optical images read from the Sky Survey CD-ROM can be converted to NCP coordinates with aips using the task skyve and than be used with kubes to overlay the radio data on these images.
If the data in data is a data cube and the data in model a 2D data set, the same contours will be drawn on each channel of the data data cube.
One can only use this mode if the sizes in X, Y and Z of the two data set are the same. I intend to make this more flexible.
This mode can only be used if the two data sets have the same size in X, Y and Z, or if one is a data cube and the other is a 2D image. In the last case, the 2D image will be combined with every channel of the cube. For example the data set can be an HI-cube and the model set an optical image.
You can also blink between a data cube and a 2D image. In this case the image will be blinked with every channel.
The modes discussed above control the way the data sets are represented, the modes below determine what you see:
At the moment, if the data set you want to see as an image is a in fact a 2D data set, this is the only mode available (eg. you select image 1 no contour and the data in the data set is a 2D image). This restriction will be lifted in the future so images can be displayed with cubes in a more flexible way.
With the contours menu one can set the contour levels. This is done by typing them in the levels field (be careful not to hit return after filling in the levels!!!). There syntax for the contour levels in as follows:
To apply the new contour levels, click on apply levels.
These options can be mixed, i.e. -0.03:-0.01+0.01 0.003:0.1+0.02 0.2 0.3 is interpreted correctly. If kubes cannot understand what you typed, it will use the default levels (i.e. 10%, 20%,....). Clicking on reset levels will give the same contour levels.
Contours are only drawn in the area defined by clicking the crosshair (i.e. left mouse). Negative contours are dashed, except for velocity fields (option VEL in the contour window).
The button Axis Labels toggles the drawing of axis labels around the image.
See the Karma User Manual section on movies.
kubes enables you to make movies similar to eg. tvmovie in aips, i.e. a sequence of images played in time. Under the button Loop sits a window that gives this control. (see the section on sequences).
An alternative to step through manually through a sequence is to have the mouse in the display window and to click with the middle button to go to the previous frame and with the right button to go to the next frame. This also works when you have set the display mode to blink.
Sequences are also possible if you have overlayed eg. the contours of of a data cube over an 2D image. Stepping through the sequence will change the image of which the contours are shown.
kubes has the same colourtable control as all the other visualisation software. The only difference is that there are a few extra colourtables compared to kview or xray. There are two tables for the Images combined mode: Two images 1 and Two images 2. Both are red and green, the difference being that the colours are swapped with respect to each other.
There is also a colourtable colled Ron&Renzo. This colourtable has 9 distinct colours. The boundary between te blue and the green colour stays fixed at the zero level in the map.
Difference between kview and kubes
Development work on kubes has long since halted, while kview is being actively developed. However, since kubes had the ability to display two datasets long before kview, many people are familiar with kubes. This section gives a quick comparison between the two tools.
kview has the following features over kubes:
kubes has the following features over kview: