|Message: Re: Stepper for a Rapidly Changing Magnetic Field||Not Logged In (login)|
Click on the Forum title, e.g. on the "Forums by Category" page, to read a sequence of postings to the Forum and its threads all in one page. If you are only interested in one thread or the thread following a specific posting, click the thread or the posting, which takes you to a smaller page, which contains only the part you are interested in and may be easier to navigate.
Messages are "chained" if there are only replies at the first level, i.e. 1/1.html, 1/1/1.html etc. In case of "chained" messages the message number is replaced by the icon and there is no indentation.
Inline: Display the subject line only or also the text of the posting(s); for the choice "All" the "Outline" choices are switched off.
|1||0||1||no text / full text of posting|
|2||1||All||text for level 1 only / text for All postings|
Outline: Choose the depth of the posting thread, successive toggle controls provide increasing detail.
|1||2||1||2 levels / 1 level (original posting)|
|2||3||2||3 levels / 2 levels|
|3||3||All||3 levels / all levels (all postings)|
Thanks for the explanation. So the classical RK4 is a very robust one.
If you do not mind, could I ask you one more question? From the Geant4 document, I found an interesting magnetic field calculation option involving "cashing" of the old calculated value. I can keep the old value under certain conditions instead of calculation new one, which could reduce the CPU time consumption.
I guess this "cashing" method may not be ideal for the iron/air interface region, but may work for the other detector area. Does the community have enough experience on this option?
Best Regards, Doris
On Fri, 15 Nov 2013 17:26:27 GMT, John Apostolakis wrote:
> Dear Doris, > > ClassicalRK4 is the most robust field integration stepper, and for this reason it is the default. It should cope well with the case that you mention - where the field changes at the interface between iron and air. > > You will have a challenge to describe this field accurately, of course, in this region because of its complexity - but it will not cause a significant problem to ClassicalRK4. > > It is possible to choose a different stepper for other reasons - for example if tracking in field is a significant CPU host spot for your application, or if you want to check whether your results or CPU speed are influenced by it. > > Two suggestions for steppers which could be interesting: > - G4NystromRK4 was developed in ATLAS (Gavrilenko and colleagues - there is a paper written about it) and has fewer field evaluations per step - a plus if your field is costly to calculate. Note: this can be used only for pure magnetic fields - it does not work in any other case; > - G4SimpleHeum is general, and *may* cope a bit better with a very-discontinuous magnetic field - if its value is not smooth when you approximate it and if it has large steps (discontinuities). > > Since replacing the default stepper requires a few operations, I refer you to the Users Guide for Applications Developers to see how to do it. > > Best regards, > John > =================================================== > > On Nov 15, 2013, at 6:19 AM, <firstname.lastname@example.org> > <email@example.com> wrote: > > > > > *** Discussion title: Fields: Magnetic and Otherwise > > > > Hello, > > > > In my understanding, RK4 is the default integration stepper for the magnetic field. I also heard that this method would not work for a rapidly changing magnetic field. > > > > For example, a track would go through a detector consisting of big chunks of iron and air gaps in between. I guess the magnetic field would change rapidly at every boundary between iron and air. In such cases, which stepper would you recommend to use? And what kind of commands should I use to implement the change? > > > > Regards, > > Doris > > > > ------------------------------------------------------------- > > Visit this GEANT4 at hypernews.slac.stanford.edu message (to reply or unsubscribe) at: > > http://hypernews.slac.stanford.edu/HyperNews/geant4/get/emfields/209.html >