Re: Dynamic freeze
- To: mathgroup at smc.vnet.net
- Subject: [mg101816] Re: Dynamic freeze
- From: David Bailey <dave at removedbailey.co.uk>
- Date: Sat, 18 Jul 2009 04:49:56 -0400 (EDT)
- References: <h3hjh0$1pd$1@smc.vnet.net>
Andrzej Kozlowski wrote: > From time to time when opening a notebook containing dynamic content > (e.g. a Mathematica demonstration) Mathematica appears to freeze, and > then the following message appears: > > One or more dynamic objects are taking excessively long to finish > evaluating. You may either disable further dynamic evaluations or > continue waiting. Dynamic updating can be re-enabled using the Dynamic > Updating Enabled command in the Evaluation menu. > > All very well, except that my experience has taught me that this > situation is (almost?) always equivalent (from the user point of view > anyway) to a Front End Crash. Clicking the "Continue" button seems to > guarantee that the same dialog will return after a further period of > apparent freeze. Clicking on "Disable Dynamic Updating" is usually > impossible and in fact I am not sure if I have ever succeeded in > getting out of this sort of situation without having to Force Quit > Mathematica. > > What's more worrying is that this sort of thing happens with perfectly > good notebooks containing no errors of any kind, including > demonstrations downloaded from the Demonstrations site. It only > happens from time to time, fortunately not very often, but I have not > been able to discover any reproducible way to arrive at this > situation. It looks like it depends on the internal state of > Mathematica rather than on the contents of the notebook. > > I am experiencing this now with version 7.01 under Mac OS X, 10.5.7, > but I think the same thing used to happen in every version since > version 6 and Mac OS X. > > Andrzej Kozlowski > Yes - this happens with Windows too. Unfortunately, I think the FE is now so complicated, because of Dynamic, that it is no longer very stable :( One way of rescuing the FE (at least most times), that I have discovered, is to use a system command (e.g. task Manager under Windows) to kill the kernel process - don't try to do this from the FE. The system usually seems to hiccup a bit, but recovers, which can be important if you have unsaved work! David Bailey http://www.dbaileyconsultancy.co.uk