Visualization of Tokamak Simulation

This page presents a set of images and videos made to show different frequency bandpass data obtained from a tokamak simulation. The entire dataset (999 time steps) was used.

Multi-band enhanced visualization

Here we hope to gain a better understanding of the results of a tokamak simulation by using direct volume rendering to render multiple bands from the simulation. Scalars are mapped to colors and opacities with a one dimensional transfer function. An example is shown in the figure 1-(a), the original dataset is rendered with blue representing the negative values while red represents the positive values. Without frequency analysis, it is difficult to discern movement between different frequency bandpasses.

In figures 1-(b) and 1-(e), we show the lowpass (blue/red represent negative/positive) and the highpass (green/violet represent negative/positive) parts.

In figure 1-(f), we have rendered the lowpass and the highpass together. Compared to figure 1-(a), it is much easier to identify the intertwined bandpass behavior in the dataset.

Volume Rendering

Timestep	1-(a) Original	1-(b) Lowpass (0~5kHz)	1-(c) Region2 (5~10kHz)	1-(d) Region3 (10~30kHz)	1-(e) Highpass (>30kHz)	1-(f) Combined
380
600
1-999	1080p DIVX(199.4MB) 720p H264(23.6MB)	1080p(24.6MB) 720p(19.9MB)	1080p(16.8MB) 720p(15.3MB)	1080p(53.4MB) 720p(20.1MB)	1080p(171.5MB) 720p(23.2MB)

Slice and Volume Rendering

Timestep	2-(a) Original	2-(b) Lowpass (0~5kHz)	2-(c) 5-10kHz	2-(d) 10-30kHz	2-(e) Highpass (>30kHz)
349
1-999(Log)		1080p DIVX(74.8MB) 720p H264(34.5MB)	1080p(200.1MB) 720p(53.9MB)	1080p(208.7MB) 720p(62.9MB)	1080p(214.9MB) 720p(62.7MB)

Discussion

1. Currently we are using different colors to display positive and negative values. Is this useful? What should we expect to see from the simulation (e.g., a difference in the rotation of positive vs. negative rings)?

2. As shown in figures 1-(b) and 1-(e), we extract the lowpass and the highpass frequency parts from the original dataset and render them separately. Are there any significant phenomena that can be observed in the lowpass or the highpass parts?

3. In figure 1-(f), both the lowpass and the highpass parts are rendered together with different transfer functions to represent the lowpass/highpass positive/negative values. Is this helpful?

4. We would like to know what you can see now previously impossible.

5. What would you like to see but the current visualization still does not offer?

6. Would it help if you can isolate spatially or temporally a region to better follow selected features? Can you envision how this is done?

7. What other raw data do you want to look together with the potential data?

8. What following probing and analysis would you like to do?

9. Please imagine you have unlimited power to manipulate your data temporally and spatially. What kind of operations would you like to perform with visual means?

Please contact Cheng-Kai Chen (ckchen@ucdavis.edu) for any question.
August 16, 2009