More Marching Squares Qs

[ThatGuyCraig]

So, I was wondering exactly how the example figure 5 was generated given that

Quote:you should isocontour on the original data and then scale up the position of the isocontours appropriately.

When I do this, I get just square shaped outlines of the original data. Which in my mind somewhat makes sense, since the original data is at such a low resolution. However, when I scale the original data using bi-linear interpolation and perform marching squares on it I get something closer to figure 5.

Are we supposed to perform bi-linear interpolation on the original contours returned by the marching squares algorithim to get the proper results? Or are we supposed to have jagged results returned?

Illustration of the problem: http://sta.sh/01zj4874qn9z

[accidental_PhD]

You should run marching squares on the original data, and scale those coordinates up (see this thread). Yes, we want to see jagged results.

[u0788158]

The isocontours should be jagged (like in the example), but it looks like you are shading the values on one side of the isovalue instead of drawing the lines. The lines should end up almost exactly like the lines in the example image.