On Sat, Feb 14, 2009 at 9:26 PM, Jason Grout <[email protected]> wrote: > > Bob Hanson (the author of Jmol) and I had the following exchange > discussing why the built-in Jmol axes seem say the origin is a different > place than Jmol axes. The end result is that it appears that Sage does > not consider the Jmol coordinate (0,0,0) to be the origin. Does anyone > know what is going on here? > > The email exchange was about why setting "axes Molecular" in jmol does > not always seem to put the built-in axes centered at the origin; > sometimes, it seems like the origin in our graph is different than what > Jmol thinks is the origin. > > The (edited) email exchange follows. > > > Thanks, > > Jason > > > > > > > Jason: > > If you go to http://sagenb.org/home/pub/243/ , you'll see the graph of > x^2+y^2. Right-clicking and displaying the jmol axes doesn't show the > axes centered at what is labeled as (0,0,0). Going to the console and > typing "axes Molecular" shifts the axes, but not to "(0,0,0)". Now, I'm > not sure if we draw the bounding box with coordinates, or if Jmol does. > It might be that we draw things shifted from where they should be for > jmol coordinates. If that is the case, we should probably change things. > > > Bob: > > Jmol is working correctly. Open a console and enter: > > > draw pt1 {0 0 0} > > The point will appear at the axes origin. For some reason your plotting > is off by 2 units in the Z axis. > > I think it will be important for you to plot in actual coordinates so > that people can do interesting things with the plot -- like measure > between points and such. I'm guessing that you have some sort of > transform there that needs revision.
It's been over a year since Robert Bradshaw and I implemented the 3d plotting code, but I distinctly remember that we first tried using jmol coordinates and many bad things happened. It was fine when the graphs were big and chunk and nicely fit into everyday values. But when we tried, e.g., drawing graphs on a small scale -- imagine plotting say from 0 to 0.0000000001 in each direction (say), things went wrong quite wrong in various ways. But for a math software system, plotting should work fine even if the total range of the axes is very small or big. So instead we decided to normalize whatever the math coordinates were to some specific box in jmol coordinates. I encourage you to try experimenting with not rescaling before using jmol to renderer, in order to see the problems I suggest above. -- William --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
