Another late post this week. I partially blame my lack of concentration on tutoring and other extra curricular activities, but they're still excuses...
Anyway, monthly meeting with supervisors yesterday came to a happy conclusion: I no longer have to worry about implementing the Eades best viewpoint finder. Instead, I will continue with my implementation of finding the 'polhodes': http://en.wikipedia.org/wiki/Polhode. [May be incorrect but the description matches].
Essentially, it is what I have implemented previously; finding the two or three most prominent axis in the graph (greatest direction of edges) and then move the viewer perpendicular to them. Bad that I didn't find the paper but reassuring that I was on the right lines without knowing it existed.
Before this, I am preparing my results for the 2D versions of the algorithms. The testing 'suite' is almost finished and last night was the first night of leaving my application working away (as a test run to ensure no errors are found). I have a few simple changes to make, but I should have full tests running over the weekend, with my first set of real results on Monday.
After I have got these running (later today), I will be continuing with the 'polhodes' work so I can finally test the 3D graphs. Again, this shouldnt take too long what with previous coding already existing.
This all finally leading to me returning to the domain of dynamic graph drawing! In retrospect, I feel slightly let down that I have not yet started on it, now being in my second year. Could be worse though, so back to work I go! Will update next week, hopefully to discuss my results, if not, the failure of the testing suite, and the rest of my work.
No comments:
Post a Comment