Re: [Visualpython-users] Errors running Vpython on OSX

[Chris B. <bu...@hv...>]

Hi all.

Thanks for the quick reply!  To clarify, the script I ran was the standard
doublependulum.py that ships with the software.  It works fine on my Linux
computer (as well as other scripts I've written).  I've tried installing
VPython on two seperate Macs and get the same behaviour on both.  Details:

  - Mac OSX 10.1.5
  - python 2.2.1 installed using fink version 0.4.1
  - Installed VPython (Visual-2002-07-22) using the install script

Regarding the theta1 suggestion, the script sometimes bombs even before
the pendulum completes an oscillation, so I don't think it's a problem with
theta becoming too large.  It does seem to have something to do with the 
number of computations, though.  I wrote a double-pendulum of my own with 
Runga-Kutta and it consistently dies more quickly.

Unfortunately, I don't maintain the Mac's, I just installed fink to get
python working, so I don't know if there might be some esoteric software
that might interfere.  But fink seems to make its own little world below
/sw, so I don't see how there could be a conflict.

I also tried installing python from source (ie., not using fink) and I get
the same problem.  

Chris

> doublependulum.py is part of the standard VPython demo suite. All the demos
> work on other OSX machines.
> 
> Bruce Sherwood
> 
> ----- Original Message -----
> From: "Andy Dougherty" <dou...@la...>
> To: "Bruce Sherwood" <bas...@un...>
> Cc: "Andy Dougherty" <dou...@ma...>;
> <cb...@sw...>; <vis...@li...>
> Sent: Friday, October 04, 2002 3:20 PM
> Subject: Re: [Visualpython-users] Errors running Vpython on OSX
> 
> 
> > On Fri, 4 Oct 2002, Bruce Sherwood wrote:
> >
> > > This vague possibility would seem to be ruled out by the fact that these
> > > programs run on other people's Mac OSX machines, with presumably the
> same
> > > math libraries, no?
> >
> > Not quite, unless I've misunderstood the original poster's question.
> > Yes, Vpython runs fine on other people's Mac OS X machines, but I'm
> > unclear if this particular script runs fine on other people's Mac OS X
> > machines.
> >
> > > > > Traceback (most recent call last):
> > > > >   File "doublependulum.py", line 71, in ?
> > > > >     atheta2 = -(A*atheta1+C*sin(theta1))/B
> > > > > ValueError: unexpected math error
> >
> > What I'm imagining is that the user is simulating a (possibly damped,
> > driven) double pendulum that might have swung over many many times, so
> > that theta1 and/or theta2 might be very large compared to 2 pi.  That's
> > the sort of thing I've seen eventually lead to odd problems elsewhere.
> >
> > Perhaps if the original script were posted, others could test it.
> >
> > --
> >     Andy Dougherty dou...@la...
> >     Dept. of Physics
> >     Lafayette College, Easton PA 18042
> >
> >
> 
>