|Message: Re: a question on replica's usage||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)|
Dear Victor, If you are trying to create just 200-300 volumes, there is no need to save memory. The memory saving is typically important if you are creating at least 10,000 or 100,000 volumes, depending on your computer. If M and N are each order 100, it could be interesting, but potentially not mandatory. To use replicas, you should understand that all copies of a replica volume will be almost exactly the same for all characteristics: material, contents (number and type of daughter volumes), sensitivity, type of volume and size. The only possible change of any of these is for a slicing along a radial axis - in that case the increment of the radius is constant and the shape varies (eg when slicing a cylinder of radius R into 3 slices along the radius, the first slice is from 0 to R/3 in radius, the second from R/3 to 2R/3 and the third from 2R/3 to R. This is a different shape.) So you cannot have different contents inside a replica. A simple way to proceed would be to place individual placement volumes in your patern in M*N locations except for the empty space sub-space. If you do not care about this much, it can be a single volume of dimension P*Q. In case you do have a very large number of volumes, you could consider to use a parameterisation. But I do not see the need for this added complication 200-300 volumes. Even if you have a large number of volumes, eg 200,000 - but not massive (eg 10,000,000) - creating the geometry in the simpler way and seeing whether the application takes too much memory would likely be the way I'd suggest. If you are convinced that you want to use replicas, you will need to separate volumes by material. Eg if a line had materials AAAAABBBBBCCCC then creating three volumes AAAA, BBBB, CCCC and slicing them further into individual volumes A1, A2, A3 and A4 will be ok. But you cannot mix A and B, or B and C inside a single replica. Best regards, John P.S. As you are submitting your question from an email that is not tied into an institution, we request that you let us know your institution and your project or its application area, in order to help enable us to continue support. Victor_Makarov wrote: > *** Discussion title: Geometry > Email replies to PublicHyperNews@slac.stanford.edu must include: > In-Reply-To: <"/geometry/820"@geant4-hn.slac.stanford.edu> > Subject: ...change this to be about your reply. > > I would like to build such a model. A boxes pattern of dimension M * N > which stand tightly to each other. The M*N number is quite big - about > 200-300. So I consider usage of replicas in order to save memory. So at > first I create a line G4PVReplica of M boxes into mother logical volume > A. Then I create a line G4PVReplica of N logical volumes A in direction > perpendicular to previous one into mother logical volume B. I hope at > this moment I'm doing right. But finally I need to put an empty space > boxes pattern of dimension P * Q in the center of already created > pattern M*N. Suppose I create a new logical volume containing P*Q vacuum > boxes. But is there any possibility to place it in pattern M*N? > According to this from Application tutorial "Normal placement volumes > may be placed inside replicas, provided that they do not intersect the > mother's or any previous replica's boundaries" it seems to me that it's > impossible. What is the solution then? Tnank you very much. >
|Inline Depth:||Outline Depth:||Add message:|