From: Giuseppe Battistoni <Giuseppe.Battistoni_at_mi.infn.it>

Date: Mon, 10 May 2010 13:32:16 +0200

Date: Mon, 10 May 2010 13:32:16 +0200

Dear Hantao

in FLUKA there exist a possible routine for your case, provided that you

have equally spaced point

in X and Y. If this is the case you can call a subroutine named D2INTP:

CALL D2INTP( XX, ILAST, TINTP, D1TINT, D2TINT, XH, X0,

& F, NI, NJ )

This subroutine allows two-dimensional weighted parabolic interpolation.

F(NI,NJ) is the the matrix to be interpolated,

XX(2): X and Y coordinate of the point where the interpolations is needed.

X0(2) are the starting points in the two coordinates (X and Y),

XH(2) are the grid spacings.

TINTP is the interpolated in XX

Depending on the value of ILAST (<2, or >=2) the interpolation

is carried out in different order to get easily the first (D1INTP) and

second (D2INTP)

derivatives

IMPORTANT!!!: Rememember: in FLUKA it is mandatory to use always Double

Precsion

Best regards

Giuseppe Battistoni

Hantao Jing wrote:

*> Dear fluka experts,
*

*> Firstly, thank Mr. Battistoni very much for help!
*

*> I have more problems when sample a generic distribution. Now I
*

*> have numerically obtained the relation between
*

*> the uniform sample of (1,0) and self variable x, y, namely, F(x,y) =z
*

*> and 0<=z <1. I can obtain z by FLRNDM(DUMMY) function. Then
*

*> I want to obtain the x,y by inverse interpolation method to finish the
*

*> sample of f(x,y) . I want know if there is an inverse-interpolation
*

*> routine in the fluka .
*

*> Or can anyone give me some code and references? On the other hand,
*

*> please tell some details about interpolation function of yinter, finter,
*

*> d..intp in fluka. Is it possible to give me the math functions in
*

*> mathmvax folder?
*

*> Thank you very much!
*

*>
*

*> Hantao Jing
*

*> 2010-5-3
*

*>
*

*>
*

*>
*

*>
*

*>
*

*>
*

*>
*

*>
*

*>
*

*>
*

*>
*

*>
*

*>
*

*>
*

*> Giuseppe Battistoni wrote:
*

*>
*

*>> Dear Hantao
*

*>>
*

*>> You are already familiar with the RANDOMIZ command.
*

*>> Therefore you have to prepare different input files for the different
*

*>> machines, just changing the WHAT(2) value of that command,
*

*>> i.e. giving different initial seeds to the pseudo-random sampling
*

*>> algorithm.
*

*>>
*

*>> Please, see also
*

*>> http://www.fluka.org/web_archive/earchive/new-fluka-discuss/2787.html
*

*>>
*

*>> Giuseppe Battistoni
*

*>>
*

*>> Hantao Jing wrote:
*

*>>
*

*>>> Dear fluka experts,
*

*>>> I want to sample a generic distribution, so I use the
*

*>>> FLRNDM(DUMMY) function to produce a random
*

*>>> sequence. I need run my input files on different computer, but I find
*

*>>> the FLRNDM always uses the same random
*

*>>> sequence when input file start to run. How to generate a different
*

*>>> random sequence by FLRNDM(DUMMY) function
*

*>>> every time?
*

*>>> Thank you very much!
*

*>>>
*

*>>> Hantao Jing
*

*>>> 2009-5-3
*

*>>>
*

*>>>
*

*>>
*

Received on Mon May 10 2010 - 14:06:49 CEST

*
This archive was generated by hypermail 2.2.0
: Mon May 10 2010 - 14:06:50 CEST
*