On Sunday, July 1, 2012 2:47:16 AM UTC-7, Andrea Lazzarotto wrote: > > I don't know if it can be considered to be related, but I noticed in some > cases that when plotting a 3d implicit plot if I put the "mesh=True" in the > plot it doesn't show up, but when inside the show method it does show up > (in Jmol, not in Tachyon). > I know nothing about the Jmol code, but I don't see how it could be related. To display a real-time 3d object, Jmol has already solved this, and numerous other intermediate problems (viewport, data representation, coordinate transforms). Somewhere in Jmol, a manager/interpreter stands between the solver's output and the displayport methods.
For a 2D plot, there is clearly no interpreter at all. It's not a bug, it's just logic that doesn't match the structure of what it's trying to do. For example, the axes might always display correctly, but the parameter "axes" is in the wrong place. This is confusing because it seems natural to define various properties of > the plot inside the plot itself, as many other parameters work in both > methods. > > -- > *Andrea Lazzarotto* - http://andrealazzarotto.com Can we hang on to these words? I think they're important. Defining the properties in that display isn't just natural, it is an accurate logical structure. Win-win. Right now (in 2D) there are no *graph * data to display. An equation is rendered to an image container, losing all information about the structure of its contents. It *is* possible to continue this way. We can hijack the image format and use, say, Value to define the height of the curve at a given point, and pack the necessary transform information along with it. This works for static objects, but I think it would be a mistake to use it to build graphs, because equations change shape as you vary the parameters. Vertex list; connection information. Points, lines, surfaces. We need to manipulate *objects* composed of these elements, then render the result to the monitor. This is what we all want, right? In the long run, there is no reason to differentiate 2D and 3D plots. A 2D plot has an orthographic camera aligned to one axis. Say it's aligned to *z *. The magnitude of all vectors <0,0,k>, from your point of view is 0. Or however people normally solve this problem. I know you all know that. I'm just reminding you. There are an *awful lot * of different plot methods to memorize, and no interpreter-with-display structure to manage them. -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
