Hi Libby,
I've attached a quick shot at what you wanted. Please read through it and let me know if it does what you want and if I did something wrong (likely!), or if you don't understand why I did something.
There is one subtlety that I should point out. It is statistically improper to consider a pair of points farther apart that 1/2 the shortest dimension of the box. Consider your starting point at the center of the box. If you look epsilon less than 1/2 of the shorted dimension of the box away towards the closest face, you're looking at a point near the periodic edge of the box. If you then look epsilon more than 1/2 the box edge past that previous point, you've reflected around the periodic boundary, and your *actual* distance is epsilon less than 1/2 because of the period. So, any distances greater than 1/2 the minimum box edge are statistically improper.
I've also attached a plot which looks encouraging, I think! It would look smoother with more random points (bigger 'loopsize')
I hope this helps! If it does, please let me know and I'll make it prettier and such and stick it into the yt codebase with documentation.
_______________________________________________________ sskory@physics.ucsd.edu o__ Stephen Skory http://physics.ucsd.edu/~sskory/ _.>/ _Graduate Student ________________________________(_)_\(_)_______________
that looks amazing, good job. thank you so much,
and yes I only want k>2 (where k = 1 / distance in code units) ;-)
Libby
2010/1/14 Stephen Skory stephenskory@yahoo.com
Hi Libby,
I've attached a quick shot at what you wanted. Please read through it and let me know if it does what you want and if I did something wrong (likely!), or if you don't understand why I did something.
There is one subtlety that I should point out. It is statistically improper to consider a pair of points farther apart that 1/2 the shortest dimension of the box. Consider your starting point at the center of the box. If you look epsilon less than 1/2 of the shorted dimension of the box away towards the closest face, you're looking at a point near the periodic edge of the box. If you then look epsilon more than 1/2 the box edge past that previous point, you've reflected around the periodic boundary, and your *actual* distance is epsilon less than 1/2 because of the period. So, any distances greater than 1/2 the minimum box edge are statistically improper.
I've also attached a plot which looks encouraging, I think! It would look smoother with more random points (bigger 'loopsize')
I hope this helps! If it does, please let me know and I'll make it prettier and such and stick it into the yt codebase with documentation.
sskory@physics.ucsd.edu o__ Stephen Skory http://physics.ucsd.edu/~sskory/ http://physics.ucsd.edu/%7Esskory/ _.>/ _Graduate Student ________________________________(_)_\(_)_______________
yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
Libby,
and yes I only want k>2 (where k = 1 / distance in code units) ;-)
Ah, sorry if I told you something you already knew ;)
_______________________________________________________ sskory@physics.ucsd.edu o__ Stephen Skory http://physics.ucsd.edu/~sskory/ _.>/ _Graduate Student ________________________________(_)_\(_)_______________
participants (2)
-
Elizabeth Harper-Clark -
Stephen Skory