From peter.willendrup at risoe.dk Tue Jan 18 14:35:07 2005 From: peter.willendrup at risoe.dk (Peter Willendrup) Date: Tue, 18 Jan 2005 14:35:07 +0100 (CET) Subject: [neutron-mc] Introducing McKnoppix Message-ID: Hello Everyone, A very easy way to get started with McStas is now available: http://mcstas.risoe.dk/knoppix A cd image based on the Knoppix (http://www.knoppix.org) Linux distribution has been patched with the needed software to run McStas. Simply write the iso to a cd image and boot up your computer with a complete linux system with McStas installed. Ideal for students and first time users working with McStas for the first time. Enjoy, The McStas team (Note: As always, we are very open to suggestions or corrections PLEASE comment... :-) ) -- ------------------------------------- Peter Kjaer Willendrup, cand. scient Phone: (+45) 46 77 58 62 email: peter.willendrup at risoe.dk ------------------------------------- From farhi at ill.fr Thu Feb 17 11:06:04 2005 From: farhi at ill.fr (Emmanuel FARHI) Date: Thu, 17 Feb 2005 11:06:04 +0100 Subject: [neutron-mc] McStas quality insurance: documentation, bugs and tests Message-ID: <42146C8C.9090708@ill.fr> Hello Everyone, As some of you already know, a meeting on McStas (neutron ray-tracing software for instrument simulations) and code quality was held at the ILL on tuesday February 8th, 2005. The meeting raised several important issues regarding quality insurance, testing methods and concrete problems with specific McStas distributed components. The McStas team has now sketched out different solutions, strategies and policies to ensure better code quality in the future. The conclusion is more or less that a lot of problems could be avoided by improving documentation, plus hightening the level of communication among code developers and users. The software components should be qualified through test procedures, or rejected. Documentation and Test methods: 1) A reference test method / procedure with examples will be made public so that writers of a new code can test their code in a systematic way. 2) Existing components should be tested thorougly - The four McStas team members, including Klaus Lieutenant (formerly Vitess) have begun the process of testing existing components. Test results will be made part of the documentation, either as part of the installed package or as web based information. 3) New components should be tested, preferably by the authors, before inclusion into the project, a 'test' procedure should be passed. 4) Information on known bugs and limitations will be available in the documentation for each component. Communication: 1) A bugzilla system is beeing set up for reporting errors and problems with McStas parts. The idea is also to use this system for the process of contributing code, plus giving the users possibility of expressing wishes for new features of the McStas package. 2) Use neutron-mc at risoe.dk list, PLEASE! If you find problems when using the package, please report back to us. The errors and problems that may annoy you is VALUABLE information for the McStas developers and other users - reporting it to us will certainly improve the software. Register to this list at and . -- What's up Doc ? -------------------------------------------- Emmanuel FARHI, http://www.ill.fr/Computing/people/Farhi \|/ ____ \|/ CS-Group ILL4/156, Institut Laue-Langevin (ILL) Grenoble ~@-/ oO \-@~ 6 rue J. Horowitz, BP 156, 38042 Grenoble Cedex 9,France /_( \__/ )_\ Work :Tel (33/0) 4 76 20 71 35. Fax (33/0) 4 76 48 39 06 \__U_/ From peter.willendrup at risoe.dk Wed Mar 16 12:08:47 2005 From: peter.willendrup at risoe.dk (Peter Willendrup) Date: Wed, 16 Mar 2005 12:08:47 +0100 (CET) Subject: [neutron-mc] ILL-RISOE collaboration and more Message-ID: Hello everyone, To indicate that collaboration and efforts by ILL and RISOE have been increased lately, the McStas webpage has been refurbished. The webpage is now also reachable at the 'neutral' adress http://www.mcstas.org Further updating will probably take place in the near future. Other nice things to expect shortly is 1) Release of the new web frontend for McStas, the result of the work of ILL student S. Jean ILL/DS/CS (Demo install will be available at this site) 2) A new release of McStas, planned for mid -> end of may 3) Bugzilla system for error tracking and requests ... And much more... Regards on behalf of the McStas team, Peter Willendrup -- ------------------------------------- Peter Kjaer Willendrup, cand. scient Phone: (+45) 46 77 58 62 email: peter.willendrup at risoe.dk ------------------------------------- From L.C.Chapon at rl.ac.uk Tue Mar 29 13:13:34 2005 From: L.C.Chapon at rl.ac.uk (Chapon, LC (Laurent)) Date: Tue, 29 Mar 2005 12:13:34 +0100 Subject: [neutron-mc] Problems in Powder2.comp Message-ID: <57BFD82EEF0CDE41A72C4EDADF8A01B2946ABA@exchange27.fed.cclrc.ac.uk> Hi, I think there are possibly two problems in the component Powder2: Line 105 : I think it should be l=v*(dt-t0) instead of v*dt. The attenuation due to scattering was therefore much larger than expected. 103 dt = rand01()*(t1 - t0) + t0; /* Time of scattering */ 104 PROP_DT(dt); /* Point of scattering */ 105 l = v*dt; /* Penetration in sample */ The other problem is on line 170 : When using focus_aw and focus_ah. The angles should be provided in radians and not degrees as suggested in the component. Otherwise, DEG2RAD should be use to convert the focus_aw and focus_ah parameters when calling the routine randvec_target_rect_angular. 170 randvec_target_rect_angular(&vx, &vy, &vz, &solid_angle, 171 aim_x, aim_y, aim_z, focus_aw, focus_ah, ROT_A_CURRENT_COMP); ______________________________________ Dr Laurent Chapon Instrument Scientist ISIS department Bldg R3 Rutherford Appleton Laboratory Chilton, Didcot OX11 0QX UK +44(0)1235-446090 -------------- next part -------------- An HTML attachment was scrubbed... URL: From peter.willendrup at risoe.dk Wed Mar 30 09:48:06 2005 From: peter.willendrup at risoe.dk (Peter Kjaer Willendrup) Date: Wed, 30 Mar 2005 09:48:06 +0200 (CEST) Subject: [neutron-mc] Problems in Powder2.comp In-Reply-To: <57BFD82EEF0CDE41A72C4EDADF8A01B2946ABA@exchange27.fed.cclrc.ac.uk> References: <57BFD82EEF0CDE41A72C4EDADF8A01B2946ABA@exchange27.fed.cclrc.ac.uk> Message-ID: Hello Laurent, Thanks for reporting. I have only had time to look at the problems briefly, but you are probably right in your claims... A new release is in the works, due aproximately may 15th. We will fix the problems in that release. Regards, Peter On Tue, 29 Mar 2005, Chapon, LC (Laurent) wrote: > Hi, > > > > I think there are possibly two problems in the component Powder2: > > > > Line 105 : I think it should be l=v*(dt-t0) instead of v*dt. The > attenuation due to scattering was therefore much larger than expected. > > > > 103 dt = rand01()*(t1 - t0) + t0; /* Time of scattering */ > > 104 PROP_DT(dt); /* Point of scattering */ > > 105 l = v*dt; /* Penetration in sample */ > > > > The other problem is on line 170 : When using focus_aw and focus_ah. The > angles should be provided in radians and not degrees as suggested in the > component. Otherwise, DEG2RAD should be use to convert the focus_aw and > focus_ah parameters when calling the routine > randvec_target_rect_angular. > > > > 170 randvec_target_rect_angular(&vx, &vy, &vz, &solid_angle, > > 171 aim_x, aim_y, aim_z, > focus_aw, focus_ah, ROT_A_CURRENT_COMP); > > > > > > ______________________________________ > > Dr Laurent Chapon > > Instrument Scientist > > ISIS department > > Bldg R3 > > Rutherford Appleton Laboratory > > Chilton, Didcot > > OX11 0QX > > UK > > +44(0)1235-446090 > > > > -- Mvh/Regards Peter -- ------------------------------------- Peter Kjaer Willendrup, Cand. Scient Phone: (+45) 46 77 58 62 email: peter.willendrup at risoe.dk ------------------------------------- From L.C.Chapon at rl.ac.uk Thu Mar 31 19:24:28 2005 From: L.C.Chapon at rl.ac.uk (Chapon, LC (Laurent)) Date: Thu, 31 Mar 2005 18:24:28 +0100 Subject: [neutron-mc] Other problems with Powder modules Message-ID: <57BFD82EEF0CDE41A72C4EDADF8A01B2946BED@exchange27.fed.cclrc.ac.uk> Dear all, following last email, I think there are possibly two more problems in the Powder modules in McStas : First in the definition of the coherent scattering cross section, the expression: my_s_v2 = PI*PI*PI*pack*DW/(Vc*Vc*V2K*V2K)*(j_1*F2_1/q_1) is not quite right. I think a factor 4 is missing, it should be 4PI^3. In the definition of the absorbing and incoherent, my_inc and my_a_v, the multiplication by the packing fraction is missing. 80 my_a_v = sigma_a/Vc*2200; /* Is not yet divided by v */ 81 my_inc = sigma_inc/Vc; Best, ______________________________________ Dr Laurent Chapon Instrument Scientist ISIS department Bldg R3 Rutherford Appleton Laboratory Chilton, Didcot OX11 0QX UK +44(0)1235-446090 -------------- next part -------------- An HTML attachment was scrubbed... URL: