From farhi at ill.fr  Sun Apr  1 11:54:24 2007
From: farhi at ill.fr (Emmanuel FARHI)
Date: Sun, 01 Apr 2007 11:54:24 +0200 (CEST)
Subject: [neutron-mc] Monte Carlo and Vista
Message-ID: <51569.193.49.43.124.1175421264.squirrel@195.83.126.3>

Vista and the perfect order system.

With the release of the latest Microsoft system, a new standard in
sotfware programming has emerged. It puts emphasis on security,
clean-looking and order. Together with the Vista patent, Microsoft has
patented the use of randomness in software, with some restrictions.
As a consequence, the use of random numbers now depends on Microsoft. This
also applies to open source projects, in order to gradualy improve the
open source quality, by reducing the non-sense chaos that comes with it.

However, the US court has recently requested that an alternate method
should be provided to the open source world as a replacement of the newly
restricted random number use. As a consequence, you should now replace in
the McStas code all occurencies of the functions:
    'rand01' 'randpm1' and 'randnorm'
with the following Java call:
    world.ms.NET.getvalue.stockmarket.random(share=Microsoft, value=Rate)

The Mc$tas team will implement this method as soon as possible. Other
Monte Carlo codes will probably be affected too.





From percival at physics.queensu.ca  Tue May  8 16:06:01 2007
From: percival at physics.queensu.ca (Aaron M. Percival)
Date: Tue, 08 May 2007 10:06:01 -0400
Subject: [neutron-mc] MPI support for SMP in Windows
Message-ID: <000f01c7917a$05d0dc00$11729400$@queensu.ca>

Hello all,

 

I'm having trouble using MPI support on my Window's based PC.  I have MPICH
installed and am able to successfully get the example programs provided to
use both processors in my PC, however, I am unsuccessful at getting McStas
to do the same.

 

The problem is I cannot compile the file to generate an executable with MPI
support.  Mpicc is a unix based program, and I cannot use it.  Also when
using mcrun with the --mpi=2 argument, I get the following message in the
command prompt:

 

mcrun: No MPI/grid machine list.

     Grid disabled, MPI will run locally...

  Define /.mcstas-hosts

  or C:\McStas\lib/tools/perl/mcstas-host

  or use option --machines=<file>!

mcrun: You have no mpicc/mpirun available, --mpi disabled...

 

Any suggestions?

 

Thanks,

 

Aaron

 

*************************************************

Aaron M. Percival

M.Sc. Candidate

Dept. of Physics, Engineering Physics & Astronomy

Queen's University

Kingston, Ontario, Canada, K7L 3N6

Office: 613-533-6000 ext. 74789

Fax: 613-533-6463

*************************************************

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman2.mcstas.org/pipermail/mcstas-users/attachments/20070508/5a674622/attachment.html>

From percival at physics.queensu.ca  Tue May  8 22:18:20 2007
From: percival at physics.queensu.ca (Aaron M. Percival)
Date: Tue, 08 May 2007 16:18:20 -0400
Subject: [neutron-mc] Addentum: MPI support
Message-ID: <000001c791ae$08d78eb0$1a86ac10$@queensu.ca>

Hello again,

 

I've been playing around with the MPI support for McStas for most of the
day.  I believe that I have solved a few of my problems:

 

I have defined a  'machines' file named 'mcstas-hosts' in the perl directory
of the mcstas directory.  The file works as I've used it in the test program
provided with mpich.

 

I have modified the mcstas_config.perl file as follows:

 

    HOSTFILE => 'C:\mcstas\lib\tools\perl\mcstas-hosts',

    MPIRUN => 'mpiexec.exe',

    THREADS => 'no',

    SSH => 'no',

 

My problem now is that I'm not sure how to define the MPICC variable, as
from my understanding mpicc is unix only.  Therefore, the  MPICC => 'no', is
not allowing me to use MPI support.  Is there a windows based compiler that
I could used here, or another way around this?

 

Thanks,

 

Aaron

 

*************************************************

Aaron M. Percival

M.Sc. Candidate

Dept. of Physics, Engineering Physics & Astronomy

Queen's University

Kingston, Ontario, Canada, K7L 3N6

Office: 613-533-6000 ext. 74789

Fax: 613-533-6463

*************************************************

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman2.mcstas.org/pipermail/mcstas-users/attachments/20070508/a3da3f52/attachment.html>

From chris2 at lns.mit.edu  Tue May  8 22:27:03 2007
From: chris2 at lns.mit.edu (Christopher Crawford)
Date: Tue, 08 May 2007 14:27:03 -0600
Subject: [neutron-mc] Addentum: MPI support
In-Reply-To: <000001c791ae$08d78eb0$1a86ac10$@queensu.ca>
References: <000001c791ae$08d78eb0$1a86ac10$@queensu.ca>
Message-ID: <CFC93184-13CA-4E17-934E-54BF811D5E44@lns.mit.edu>

Hi Aaron,
   Check out 'cygwin.com'.  They provide almost full Unix  
functionality for Windows, including the standard 'gcc' compiler and  
and x-server, under the GPL.
--Chris

TA-53/MPF-1/D111 P-25 MS H846
LANL, Los Alamos, NM  87545
505-665-9804(o) 665-4121(f) 662-0639(h)


On May 8, 2007, at 14:18:20, Aaron M. Percival wrote:

> Hello again,
>
>
>
> I?ve been playing around with the MPI support for McStas for most  
> of the day.  I believe that I have solved a few of my problems:
>
>
>
> I have defined a  ?machines? file named ?mcstas-hosts? in the perl  
> directory of the mcstas directory.  The file works as I?ve used it  
> in the test program provided with mpich.
>
>
>
> I have modified the mcstas_config.perl file as follows:
>
>
>
>     HOSTFILE => 'C:\mcstas\lib\tools\perl\mcstas-hosts',
>
>     MPIRUN => 'mpiexec.exe',
>
>     THREADS => 'no',
>
>     SSH => 'no',
>
>
>
> My problem now is that I?m not sure how to define the MPICC  
> variable, as from my understanding mpicc is unix only.  Therefore,  
> the  MPICC => 'no', is not allowing me to use MPI support.  Is  
> there a windows based compiler that I could used here, or another  
> way around this?
>
>
>
> Thanks,
>
>
>
> Aaron
>
>
>
> *************************************************
>
> Aaron M. Percival
>
> M.Sc. Candidate
>
> Dept. of Physics, Engineering Physics & Astronomy
>
> Queen's University
>
> Kingston, Ontario, Canada, K7L 3N6
>
> Office: 613-533-6000 ext. 74789
>
> Fax: 613-533-6463
>
> *************************************************
>
>
>
> _______________________________________________
> neutron-mc mailing list
> neutron-mc at risoe.dk
> http://mailman.risoe.dk/mailman/listinfo/neutron-mc






From peter.willendrup at risoe.dk  Tue May  8 22:42:37 2007
From: peter.willendrup at risoe.dk (Peter Willendrup)
Date: Tue, 08 May 2007 22:42:37 +0200 (CEST)
Subject: [neutron-mc] Addentum: MPI support
In-Reply-To: <CFC93184-13CA-4E17-934E-54BF811D5E44@lns.mit.edu>
References: <000001c791ae$08d78eb0$1a86ac10$@queensu.ca>
	<CFC93184-13CA-4E17-934E-54BF811D5E44@lns.mit.edu>
Message-ID: <Pine.LNX.4.58.0705082234470.12289@fys-lin-1.risoe.dk>

Hi guys,


I must admit that I didn't invest much time in looking at MPICH and McStas
on Win32 to date. I am very happy that you've started breaking ground for
McStas in that area Aaron.

You are probably right that mpicc is 'unix only' in the sense that it is a
shell script calling your local c compiler (in many cases gcc) with proper
lib and include flags. I will have a look around to see if I can find a
win32 equivalent or alternatively investigate what difficulties we would
have in writing up an equivalent batch.

Chris: you are absolutely right that cygwin could be an option, however
till date we have avoided it since the default install (which is what an
average Win user would select) is about 1Gb -> a lot of overhead for
something as small as McStas.

Cheers,

Peter

--
-------------------------------------
Peter Kjaer Willendrup, cand. scient

Phone: (+45) 46 77 58 62
email: peter.willendrup at risoe.dk
-------------------------------------

On Tue, 8 May 2007, Christopher Crawford wrote:

> Hi Aaron,
>    Check out 'cygwin.com'.  They provide almost full Unix
> functionality for Windows, including the standard 'gcc' compiler and
> and x-server, under the GPL.
> --Chris
>
> TA-53/MPF-1/D111 P-25 MS H846
> LANL, Los Alamos, NM  87545
> 505-665-9804(o) 665-4121(f) 662-0639(h)
>
>
> On May 8, 2007, at 14:18:20, Aaron M. Percival wrote:
>
> > Hello again,
> >
> >
> >
> > I?ve been playing around with the MPI support for McStas for most
> > of the day.  I believe that I have solved a few of my problems:
> >
> >
> >
> > I have defined a  ?machines? file named ?mcstas-hosts? in the perl
> > directory of the mcstas directory.  The file works as I?ve used it
> > in the test program provided with mpich.
> >
> >
> >
> > I have modified the mcstas_config.perl file as follows:
> >
> >
> >
> >     HOSTFILE => 'C:\mcstas\lib\tools\perl\mcstas-hosts',
> >
> >     MPIRUN => 'mpiexec.exe',
> >
> >     THREADS => 'no',
> >
> >     SSH => 'no',
> >
> >
> >
> > My problem now is that I?m not sure how to define the MPICC
> > variable, as from my understanding mpicc is unix only.  Therefore,
> > the  MPICC => 'no', is not allowing me to use MPI support.  Is
> > there a windows based compiler that I could used here, or another
> > way around this?
> >
> >
> >
> > Thanks,
> >
> >
> >
> > Aaron
> >
> >
> >
> > *************************************************
> >
> > Aaron M. Percival
> >
> > M.Sc. Candidate
> >
> > Dept. of Physics, Engineering Physics & Astronomy
> >
> > Queen's University
> >
> > Kingston, Ontario, Canada, K7L 3N6
> >
> > Office: 613-533-6000 ext. 74789
> >
> > Fax: 613-533-6463
> >
> > *************************************************
> >
> >
> >
> > _______________________________________________
> > neutron-mc mailing list
> > neutron-mc at risoe.dk
> > http://mailman.risoe.dk/mailman/listinfo/neutron-mc
>
>
>
>
> _______________________________________________
> neutron-mc mailing list
> neutron-mc at risoe.dk
> http://mailman.risoe.dk/mailman/listinfo/neutron-mc
>

Mvh/Regards

Peter

--
-------------------------------------
Peter Kjaer Willendrup, Cand. Scient

Phone: (+45) 46 77 58 62
email: peter.willendrup at risoe.dk
-------------------------------------


From peter.willendrup at risoe.dk  Thu May 10 21:27:49 2007
From: peter.willendrup at risoe.dk (Willendrup, Peter Kj=?ISO-8859-1?B?5g==?=r)
Date: Thu, 10 May 2007 21:27:49 +0200
Subject: [neutron-mc] Addentum: MPI support
In-Reply-To: <000001c791ae$08d78eb0$1a86ac10$@queensu.ca>
Message-ID: <C2693ED5.3009%PKWI@exchange.Risoe.dk>

Hello there,


Aaron Percival and I have now gotten MPICH2 to work with McStas on Windows.
(Up to now successful only on a local machine with several
processors/cores.)

I will do the needed modifications in the next stable release (mcstas-1.11)
to allow everyone to use this.


Regards,

Peter Willendrup


On 5/8/07 10:18 PM, "Aaron M. Percival" <percival at physics.queensu.ca> wrote:

> I've been playing around with the MPI support for McStas for most of the
> day.  I believe that I have solved a few of my problems:
> 
>  
> 
> I have defined a  'machines' file named 'mcstas-hosts' in the perl directory
> of the mcstas directory.  The file works as I've used it in the test program
> provided with mpich.
> 
>  
> 
> I have modified the mcstas_config.perl file as follows:
> 
>  
> 
>     HOSTFILE => 'C:\mcstas\lib\tools\perl\mcstas-hosts',
> 
>     MPIRUN => 'mpiexec.exe',
> 
>     THREADS => 'no',
> 
>     SSH => 'no',
> 
>  
> 
> My problem now is that I'm not sure how to define the MPICC variable, as
> from my understanding mpicc is unix only.  Therefore, the  MPICC => 'no', is
> not allowing me to use MPI support.  Is there a windows based compiler that
> I could used here, or another way around this?
> 
>  
> 
> Thanks,
> 
>  
> 
> Aaron
> 
>  
> 
> *************************************************
> 
> Aaron M. Percival
> 
> M.Sc. Candidate
> 
> Dept. of Physics, Engineering Physics & Astronomy
> 
> Queen's University
> 
> Kingston, Ontario, Canada, K7L 3N6
> 
> Office: 613-533-6000 ext. 74789
> 
> Fax: 613-533-6463
> 
> *************************************************
> 
>  
> 
> _______________________________________________
> neutron-mc mailing list
> neutron-mc at risoe.dk
> http://mailman.risoe.dk/mailman/listinfo/neutron-mc



From percival at physics.queensu.ca  Tue May 22 18:11:35 2007
From: percival at physics.queensu.ca (Aaron M. Percival)
Date: Tue, 22 May 2007 12:11:35 -0400
Subject: [neutron-mc] Virtual Output file
Message-ID: <000001c79c8b$e1a78d70$a4f6a850$@queensu.ca>

Hello,

 

I was wondering if it were possible to merge files created by the virtual
output instrument?  Is there a correct way to do this, is it even possible
at all?  Can I simply append the neutron data from one file to another?

 

Thanks,

 

Aaron

 

*************************************************

Aaron M. Percival

M.Sc. Candidate

Dept. of Physics, Engineering Physics & Astronomy

Queen's University

Kingston, Ontario, Canada, K7L 3N6

Office: 613-533-6000 ext. 74789

Fax: 613-533-6463

*************************************************

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman2.mcstas.org/pipermail/mcstas-users/attachments/20070522/52256e9f/attachment.html>

From chris2 at lns.mit.edu  Tue May 22 18:59:10 2007
From: chris2 at lns.mit.edu (Christopher Crawford)
Date: Tue, 22 May 2007 10:59:10 -0600
Subject: [neutron-mc] Virtual Output file
In-Reply-To: <000001c79c8b$e1a78d70$a4f6a850$@queensu.ca>
References: <000001c79c8b$e1a78d70$a4f6a850$@queensu.ca>
Message-ID: <291562E4-20CE-4712-9C25-AA225A8B157A@lns.mit.edu>

Hi Aaron,
   If you are using binary files, it is a simple 'cat a b > out'.  I  
had no luck creating my own 'text' files externally.
--Chris

TA-53/MPF-1/D111 P-25 MS H846
LANL, Los Alamos, NM  87545
505-665-9804(o) 665-4121(f) 662-0639(h)


On May 22, 2007, at 10:11:35, Aaron M. Percival wrote:

> Hello,
>
>
>
> I was wondering if it were possible to merge files created by the  
> virtual output instrument?  Is there a correct way to do this, is  
> it even possible at all?  Can I simply append the neutron data from  
> one file to another?
>
>
>
> Thanks,
>
>
>
> Aaron
>
>
>
> *************************************************
>
> Aaron M. Percival
>
> M.Sc. Candidate
>
> Dept. of Physics, Engineering Physics & Astronomy
>
> Queen's University
>
> Kingston, Ontario, Canada, K7L 3N6
>
> Office: 613-533-6000 ext. 74789
>
> Fax: 613-533-6463
>
> *************************************************
>
>
>
> _______________________________________________
> neutron-mc mailing list
> neutron-mc at risoe.dk
> http://mailman.risoe.dk/mailman/listinfo/neutron-mc





From peter.willendrup at risoe.dk  Tue May 22 19:38:34 2007
From: peter.willendrup at risoe.dk (Willendrup, Peter Kj=?ISO-8859-1?B?5g==?=r)
Date: Tue, 22 May 2007 19:38:34 +0200
Subject: [neutron-mc] Virtual Output file
In-Reply-To: <291562E4-20CE-4712-9C25-AA225A8B157A@lns.mit.edu>
Message-ID: <C278F73A.31AD%PKWI@exchange.Risoe.dk>

Hi guys,


Alternatively, the mcformat tool should actually be able to handle merging
of ascii event files. It was introduced in 1.10 but has definately evolved
since then.

If you want to handle the ascii stuff, simply get the dev version from
http://www.mcstas.org/cvs or wait just a little longer for 1.11...


Cheers,

Peter


On 5/22/07 6:59 PM, "Christopher Crawford" <chris2 at lns.mit.edu> wrote:

> Hi Aaron,
>    If you are using binary files, it is a simple 'cat a b > out'.  I
> had no luck creating my own 'text' files externally.
> --Chris
> 
> TA-53/MPF-1/D111 P-25 MS H846
> LANL, Los Alamos, NM  87545
> 505-665-9804(o) 665-4121(f) 662-0639(h)
> 
> 
> On May 22, 2007, at 10:11:35, Aaron M. Percival wrote:
> 
>> Hello,
>> 
>> 
>> 
>> I was wondering if it were possible to merge files created by the
>> virtual output instrument?  Is there a correct way to do this, is
>> it even possible at all?  Can I simply append the neutron data from
>> one file to another?
>> 
>> 
>> 
>> Thanks,
>> 
>> 
>> 
>> Aaron
>> 
>> 
>> 
>> *************************************************
>> 
>> Aaron M. Percival
>> 
>> M.Sc. Candidate
>> 
>> Dept. of Physics, Engineering Physics & Astronomy
>> 
>> Queen's University
>> 
>> Kingston, Ontario, Canada, K7L 3N6
>> 
>> Office: 613-533-6000 ext. 74789
>> 
>> Fax: 613-533-6463
>> 
>> *************************************************
>> 
>> 
>> 
>> _______________________________________________
>> neutron-mc mailing list
>> neutron-mc at risoe.dk
>> http://mailman.risoe.dk/mailman/listinfo/neutron-mc
> 
> 
> 
> _______________________________________________
> neutron-mc mailing list
> neutron-mc at risoe.dk
> http://mailman.risoe.dk/mailman/listinfo/neutron-mc



From kim.lefmann at risoe.dk  Tue May 22 21:13:04 2007
From: kim.lefmann at risoe.dk (Kim Lefmann)
Date: Tue, 22 May 2007 21:13:04 +0200 (CEST)
Subject: [neutron-mc] Virtual Output file
In-Reply-To: <C278F73A.31AD%PKWI@exchange.Risoe.dk>
References: <C278F73A.31AD%PKWI@exchange.Risoe.dk>
Message-ID: <Pine.LNX.4.53.0705222112180.3472@linux-01.risoe.dk>

Hi guys,

I have a feeling that the absolute intensities may be jeopardized by this 
method. You can try this out very easily (I do not have the time, sorry)

best, kim
-- 
Kim Lefmann
Professor, Senior Scientist
Materials Research Department
Building 227
Risoe National Laboratory
Technical University of Denmark

Phone: +45 46 77 47 26
Fax:   +45 46 77 57 58


From farhi at ill.fr  Tue May 22 22:43:22 2007
From: farhi at ill.fr (Emmanuel FARHI)
Date: Tue, 22 May 2007 22:43:22 +0200 (CEST)
Subject: [neutron-mc] Virtual Output file
In-Reply-To: <Pine.LNX.4.53.0705222112180.3472@linux-01.risoe.dk>
References: <C278F73A.31AD%PKWI@exchange.Risoe.dk>
	<Pine.LNX.4.53.0705222112180.3472@linux-01.risoe.dk>
Message-ID: <49531.193.49.43.124.1179866602.squirrel@195.83.126.3>

hi,

yes intensities should be checked. And the mcformat is an easy way to
merge ascii event files. Other way is to remove the 'header' (# lines) and
then do a 'cat'.

binary may use 'cat' indeed.

E.

> Hi guys,
>
> I have a feeling that the absolute intensities may be jeopardized by this
> method. You can try this out very easily (I do not have the time, sorry)
>
> best, kim
> --
> Kim Lefmann
> Professor, Senior Scientist
> Materials Research Department
> Building 227
> Risoe National Laboratory
> Technical University of Denmark
>
> Phone: +45 46 77 47 26
> Fax:   +45 46 77 57 58
> _______________________________________________
> neutron-mc mailing list
> neutron-mc at risoe.dk
> http://mailman.risoe.dk/mailman/listinfo/neutron-mc
>


-- 
FARHI Emmanuel <farhi at ill.fr>
Groupe DS/CS, ILL4/156, Tel 04 76 20 71 35
ILL, Grenoble