On Wednesday, July 18, 2012 2:00:58 AM UTC-4, jason wrote: > > On 7/18/12 12:50 AM, Dan Drake wrote: > > On Tue, 17 Jul 2012 at 10:27PM -0700, R. Andrew Ohana wrote: > >> This is kind of fixed by #11475 (in that prime_pi becomes part of the > >> symbolic ring). > > > > Cool. > > > >> The bigger issue is that the symbolic ring does sampling for plotting, > >> which is horribly inefficient for prime_pi -- something that is solved > >> with the custom plotting method for prime_pi. > > > > Sampling is definitely a poor idea here. How does one access the custom > > plotting method? It seems like plot(prime_pi, 5, 100) does so -- but why > > then does plot(prime_pi, (5,100)) *not* use it? > > > > Is there a way to get the plot command to be smart enough to > > automatically use the custom method? > > IIRC, if A.plot exists, then plot(A, ...) will call A.plot(...). I > think plot(prime_pi, ...) *is* using the custom plotting method always. > The problem is that the prime_pi.plot method is very limited, and > doesn't recognize prime_pi.plot((5,100)). >
Correct. Dan, you should have asked me yesterday while I was doing the related thing in the workshop - that the only way to plot prime_pi(x)-x/ln(x) is plot(lambda x: prime_pi(x)-x/ln(x), (x,3,100)) which as Andrew points out isn't really good, though I suspect the recursive algorithm would do decently over small things. Any ideas on that one? This also causes problems in helping a different participant. E = EllipticCurve(...) plot(E,-3,3) works but plot(E,(x,-3,3)) doesn't. -- -- 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
