Release of Fluka2006.3b

From: Alfredo Ferrari (
Date: Tue Mar 27 2007 - 22:32:47 CEST

  • Next message: "delayed (n, gamma)"

    Dear FLUKA users

    Fluka2006.3b has been released. This is not a major release, but rather
    an incremental and bug fix release. Note that it *substitutes*
    Fluka2006.3. Below please find the release notes: read carefully!

                     Best regards
                 The FLUKA development team

    P.S. This message is sent to the fluka-users and fluka-discuss lists as
          well as to some other lists internal to the FLUKA Collaboration. I
          apologize to all those who will get it twice or more

    - Release notes for Fluka2006.3b (patch version of Fluka2006.3) -

    This release contains several small fixes and a few functionality improvements
    with respect to Fluka2006.3. Most of them are likely to be irrelevant for
    the majority of the users, however it is safer and recommended to
    immediately move to Fluka2006.3b. The move should be painless since there
    is no change in the physics. Only be careful to recompile user routines
    since a couple of commons changed.

    This release is a bug fixing release for Fluka2006.3. As such it doesn't
    count as a new release, and it doesn't obsolete Fluka2005.6, for the
    purpose of the provisions of the Fluka User license. Rather, it obsoletes
    Fluka2006.3 and hence all users of Fluka2006.3 are invited to move
    to this bug fixing release from now on, particularly for publications.

    The most relevant bug fixes or functionality improvements are listed below:

    * LATTICE cards now accept a mixture of (region) names and (lattice)
       numbers in order to simplify lattice input (see )

    * The memory allocations for the (maximum) numbers of irradiation intervals
       and cooling times are now handled together, so that there is much
       more flexibility when inputting a large number of irradiation intervals
       (and relatively few cooling times) and viceversa. Small adjustments in
       the usrsuwev program have been implemented because of this

    * Creating a fluka.stop file inside the fluka_xxxx working directory
       is going to stop the current run as it did in all previous releases.
       Starting from this release, if the file is instead named rfluka.stop,
       not only the current run is stopped, but also the run sequence is
       stopped even though the total requested number of runs has not yet
       been achieved

    * When scoring some types of "pointwise" energy depositions,
       the JTRACK variable (common TRACKR) is set to a generalized particle
         JTRACK = 208 for non-transported nuclear recoils
         JTRACK = 308 for low energy neutron kerma
         JTRACK = 211 for EM particles produced below threshold
       Values 208 and 211 were already set in past versions, while the 308 flag
       is new. A new variable has been added in the TRACKR common to help identify
       these energy depositions:
       it records the ID of the particle that originated the interaction.
       Warning: this variable is normally set to 0, its value has a
       meaning only when JTRACK = 208, 211, 308.

    * A file, Version.tag, is included in the distribution in order to
       simplify (semi)automatic procedures for identifying versions and

    * The heavy ion dE/dx at low energies (below 10-30 MeV/n for projectiles
       of medium-large Z's) had a bug which slipped into the Fluka2005.6 release
       and went unnoticed till now (thanks to Ercan Pilicer for pointing it
       out), effectively disabling the effective Z algorithm.
       All users who run heavy ion beams at low energies are warned to
       moved immediately to Fluka2006.3b: results are now back to those of
       Fluka200x, x=0,1,2,3,4 and in agreement with published benchmarks

    * A couple of bugs were still lurking in the Birks law quenching
       implementation when requested through the Mgdraw routine (thanks to
       Vincenzo Patera for pointing out the problem). These bugs should have
       affected only problems with magnetic field, however users who were
       making use of this novel feature of Fluka2006.3 should better check
       if their results are still the same. All users whose runs will stop
       with the message
                 "FKBIRK, NONSENSE xxxxx CALL"
       are warmly invited to contact us through fluka-discuss since the
       messages could imply that problems are still around in the
       Mgdraw-driven quenching implementation

    * The prompt vs delayed radiation biasing selection through WHAT(4) of
       the RADDECAY card was badly broken (thanks to Stefan Roesler for pointing
       out this problem): now it is fixed and should behave as advertised
       in the manual (feedback welcome)

    Many other minor bug fixes have been implemented, but they should be
    completely transparent to end users

    A few further news of relevance for Fluka users are reported below:

    * From this release on the and mailing
       addresses no longer exist. The messages sent to those addresses were
       almost all about topics better discussed on
       Whichever problem with the web site and/or the registration and download
       procedures should be reported to the same list as well with possibly
       [SUPPORT] at the beginning of the subject line

    * A new list,, has been created. All registered
       Fluka users have been automatically subscribed to this list and new users
       will be as well. This is a low traffic, one way only list, dedicated
       to announcements (like a new release, a bug fixing respin, etc etc)
       which are deemed to be relevant for all users. We strongly invite all
       active Fluka users to not quit this list. Those no longer active, or
       anyway whoever so wishes, can unsubscribe sending a message to with "unsubscribe fluka-users" in the main body.
       As usual, all active users are strongly encouraged to subscribe to

    * A new very powerful tool for interacting with Fluka both at input
       and output stages is now available, thanks to Vasilis Vlachoudis (CERN)
       This tool, called Flair (FLuka Advanced Interface), can be downloaded

       This tool should run on whichever modern Linux distribution: users are
       strongly encouraged to test it and provide their feedback through the
       fluka-discuss list. All other tools, like FlukaGUI and TVF NMCRC, are
       obviously still available via

                 The FLUKA development team

    The release notes for Fluka2006.3 and before are reported below.
    They still apply unless explicitly superseded above. We invite the users
    to give particular attention to the PEANUT extension described below.

    - Release notes for Fluka2006.3 (release version) -

    This release is an incremental step in the FLUKA development with
    respect to Fluka2005.6. It adds a few features and there are physics
    improvements in several areas. It represents a major milestone from
    the user interface point of view with the introduction of the
    "input by names".

    The users should keep in mind that there are a significant number of new
    features in this release, and therefore some problems are expected. The
    development team has spent a significant amount of time debugging the new
    features, however we would be surprised if no issue will show up.
    Users are recommended to check the Fluka home page for possible advisories
    on patched versions.

    As it is obvious from the name, the major revision number of this release
    is 2006 and the minor revision number is 3. With this release all FLUKA
    version older or equal to 2003.1b are obsoleted and they should no longer
    be used according to the conditions spelled out in the FLUKA license.


    Among the new features of this release (with respect to Fluka2005.6):

    - New "Input by name"(fully backcompatible with the past): particle,
       materials, regions, generalized particles, binnings, and estimators can
       now be indicated in the input file through their names rather than their
       numbers. This new feature is compatible with the old input way: a mix of
       name based and numeric values can be used in the input files. The
       included example input file (example.inp) is now written "by name": the
       traditional version (exfixed.inp) and a mixed one (exmixed.inp) are also
       provided. See the manual for further details.

    - New fission model/improvements to evaporation/fragmentation
       * Actinide fission now done on first principles and no longer on
         parametrized G_fiss/G_neu
       * New fission barrier calculations following the most recent suggestions
         by Myers & Swiatecki
       * Fission level density enhancement at saddle point no longer
         excitation energy independent but now washing out with excitation
         energy coherently with the most recent studies and the recommendations
         of a IAEA working group
       * Fission product widths and asymmetric vs symmetric probabilities
         better parametrized according to the most recent data/approaches
       * New, energy dependent self-consistent, evaporation level densities
         according to the IAEA working group recommendations
       * New pairing energies consistent with the above point
       * New mass tables including calculated masses besides exp. ones till
         A=330. The use of masses calculated offline (available electronically)
         with high reliability complex models allows, a) to extend to A larger
         than those experimentally accessible, b) to minimize resorting to
         empirical mass formulae online which often generate artefacts
       * New shell corrections coherent with the new masses
       The overall result in the residual predictions in the spallation zone
       is a striking improvement for actinides (which was poor before), a nice
       improvement for non-actinides (Pb, Au etc, it was already not bad),
       and a global improvement in the mass distribution of fission fragments
       for all of them. For non fissionable light-medium mass nuclei differences
       are minor, nevertheless the new level densities appear to smooth out
       some features and in particular some excessive odd-even effect

    - An initial implementation of the BME model, available on request,
       has been performed. It covers light ion interactions up to 100 MeV/n.
       First results, when applied to positron emitter production with therapy
       beams are encouraging

    - Speed up of radioactive nuclei evolution

    - Lattices: the required transformations can now be specified associating
       each lattice with a specific roto-translation defined through
       ROT-DEFI. This is a viable alternative to a user written lattic.f when
       a limited number of transformations has to be defined. Both methods are
       and will be fully supported, in principle the user can mix and use
       predefined transformations for some lattices, and lattic.f for others.
       See the manual (LATTICE card) for details

    - An algorithm for parentheses optimization is now implemented in the
       geometry package (contributed by V.Vlachoudis, see GEOBEGIN)

    - Activity concentration 2D/3D binnings are now implemented through the
       new generalized particle types ACTIVITY (234, activity per unit volume)
       and ACTOMASS (235, activity per unit mass)

    - Residual nuclei scoring and gas production: protons are now included
       in the RESNUCLEi scoring, in order to prevent lazy users from
       obtaining nonsense results on gas production (see below)

    - Beta+/- spectra now include Coulomb and screening corrections

    - Photomuon production is now implemented limited to coherent
       (Bethe-Heitler) production for the time being (contributed by
       S.Roesler/A.Fasso`). It can be activated by the PHOTONUC card

    - Explicit primary ionization events can be requested on a material
       basis. The user must provide the number of primary ionizations
       per cm (and for some variants of the model a guess for the 1st ionization
       potential) and choose one of the four available variants. Primary
       ionization electron energies will then be stored inside common
       ALLDLT at each step in the selected materials. It can be activated
       by the IONFLUCT card. Use with care and possibly for gases only.
       The number of primary ionizations can quickly escalate, particularly
       when multiply charged ions are involved. No common saturation should
       occur since the code is piling up all the remaining primary electrons
       into the last common location if required, however CPU penalties can
       be severe if used without wisdom

    - Extension of PEANUT: last but not least, a new high energy event
       generator has been developed, based on the sophisticated nuclear
       physics of PEANUT coupled with the proved FLUKA Dual Parton Model
       description for hadron-hadron collisions and a brand new Glauber
       cascade treatment. This model will eventually substitute as default
       the old one (PEANUT is already the default below 5 GeV). It is not
       yet the default, mostly because it requires a bit more testing
       and cleaning up some FLUKA inconsistencies related to quasi-elastic
       treatment. All thin target benchmarks of the code by the development
       team are now run with the new model, the development of the old one
       being frozen. The PHYSICS cards allows to switch on the new model
       (with some caveats about the quasielastic issue) ie with:

    PHYSICS 1000. 1000. 1000. 1000. 1000. 1000. PEATHRES


    - Whenever residual nuclei (and residual dose rates) scoring is of
       importance, the heavy residue emission ("fragmentation") and the
       coalescence emission of fast complex particles should be switched on,
       through the following data cards:


       and (as a consequence of coalescence) it would be wise to link with
       rQMD-2.4 (and DPMJET) and activate ion transport and interactions. These
       suggestions are mandatory for residual nuclei benchmarking and validation.

       Those options are not on by default because the heavy evaporation carries a
       big CPU penalty which would be a waste for most problems when residuals are
       not a issue.

    - All previous versions never included scoring of A=1 residuals (supposed to
       be "elementary" hadrons and not composite nuclei) when the
       RESNUCLEi card was activated: this was clearly indicated and it was
       obvious from the output of the usrsuw and usrsuwev auxiliary programs.
       Nevertheless some users missed completely this feature, and went on
       publishing results about gas production (hydrogen production) which were
       obviously nonsensically low, without realizing they were scoring only
       H-2 and H-3 production and not H-1.


    - the COMMENT card is deprecated (ordinary comments starting with "*" are
       of course supported): it is still accepted but there is no guarantee
       it works properly


    The following features are currently under active development. Some
    of them are ready but were not included for lack of time, others are ready
    and possibly already presented or going to be presented to conferences,
    but not yet tested enough for a general user version, others are in
    various stages of completeness. The Authors warn that every result obtained
    out of the production version of the code which could be heavily
    dependent on any of these features will not be representative of the actual
    performances of FLUKA and therefore shall not be published.

    List of the features under active development but not yet included in the
    production version of FLUKA:

    a) New 260 group neutron cross section library

    b) Heavy fragment emission in the preequilibrium stage

    c) Impact ionization cross sections

    d) Compton with Doppler shift

    e) Heavy ion pair production

    f) Direct resonance transport and interaction in PEANUT

    g) Updated multiple scattering model (including the so called
        polygonal approach)

    h) New hadron elastic scattering model at intermediate energies

    i) Neutral kaon regeneration (partially implemented, but still faulty
        because not yet performed at scattering amplitude level)

    j) New QMD model developed in house covering the range 50-400 MeV/n


    This version of the code should be run on the platforms for which it
    has been released, that is Linux x86 and Compaq TrueUnix. The code has
    been checked and validated for these platforms only for the time being.
    The availability of the source code shall not be exploited for tentative
    builds on other architectures or with different compilers/compiler options
    than the ones recommended by the development team. Our experience shows that
    for a code of the complexity of FLUKA the chances of hitting one or more
    compiler issues are close to unit. Therefore users shall not make use
    for every serious job, including whichever form of publication or
    presentation, of code versions built on platforms and/or with compiler
    options which have not been cleared as safe by the development team.

    The development team is actively developing and testing FLUKA also on
    other platforms which will become available in the close future.


    This is the second release of FLUKA carried out under the INFN-CERN
    Collaboration Agreement for the Maintenance and Development of the FLUKA

    The development and distribution of FLUKA is managed by two Committees,
    the former, the Fluka Coordination Committee (FCC), which is representing
    the Copyright Holders (INFN and CERN) and which is empowered for all
    major decisions, the latter, the Fluka Scientific Committee (FSC), which
    is in charge of the day-to-day development of the code and of the
    technical and scientific issues.

    The present membership of the Fluka Coordination Committee and the
    Fluka Scientific Committee are the following:


        Giuseppe Battistoni INFN/Milan (chairman)
        John Ferguson CERN/AB
        John Harvey CERN/PH
        Johannes Ranft Siegen Un.
        Paola R. Sala INFN/Milan


        Giuseppe Battistoni INFN/Milan
        Federico Carminati CERN/PH
        Alberto Fasso` SLAC
        Alfredo Ferrari CERN/AB (secretary)
        Maurizio Pelliccioni INFN/LNF
        Johannes Ranft Siegen Un.
        Stefan Roesler CERN/RP
        Paola R. Sala INFN/Milan
        Vasilis Vlachoudis CERN/AB

    The cornerstones of the INFN-CERN Agreement are the following:

    a) the code is Copyright 1989-2006 INFN and CERN, and the authors
        are Alberto Fasso`, Alfredo Ferrari, Johannes Ranft and Paola Sala:
        the Copyright and licensing conditions extend to all the work
        performed by the Authors since 1989, and therefore covers the
        vast majority of the code contained in all FLUKA releases or
        development versions since that date
    b) make the code available to the scientific community under a License
        which gives broad rights to the end user
    c) protect the code integrity and authorship inserting in the License
        proper conditions
    d) assure that only official versions of the code will be used,
        prosecuting in case, including by legal means, the use of unauthorized,
        or worse, pirated versions

    This release is available in source form for CERN staff members and INFN
    researchers: special download forms are available on the Fluka web site
    for this purpose. The source is now also available for other scientific
    Institutions, pending the technical time required to setup the download
    forms and procedures. The source release includes the Linux-x86 version
    only (which runs on x86_64 machines as well). The Fluka Coordination
    Committee, the Fluka Scientific Committee and the Authors kindly invite
    all users to refer to any of them for whichever question or doubt about
    the source release and its condition of use.

    Code snippets setting an expiration day for this release
    version are present in a few routines: obviously the availability
    of the source code allows to change it, however users should be aware
    that under the licensing conditions this is not permitted. The code
    expiration date (around january 2008 for this release) is there as a
    reminder to use up-to-date versions. In no way it is intended as a
    robust protection, the code distribution is done as always on a mutual
    trust basis.

    The "FLUKA User Routines" mentioned at point 3) in the FLUKA User License
    are obviously those (and only those) contained in the directory usermvax,
    both in the source and binary versions of the code.

    Copyright statements referring to one of more of the Authors (A.Fasso`,
    A.Ferrari, J.Ranft, P.R. Sala) contained in individual routines, must
    always be interpreted as:

    Copyright INFN and CERN, Authors: ...

    since the Authors have transferred their rights to INFN and CERN at the
    time of (and subject to) the enactement of the INFN-CERN agreement of

    A proper re-elaboration of all those Copyright statements is going on.


    The source code for DPMJET and for the version of rQMD-2.4 used together
    with FLUKA is not yet available for this release. The development
    team is finalizing the proper distribution conditions for these codes,
    which will be possibly included in the next release. The binary
    libraries are anyway available as usual, hence we do not expect any serious
    inconvenience for users who need heavy ion capabilities

    There are several routines in this release which are apparently
    useless and not required for a successful link of the code. Most of them
    are new developments which either are activated in the development
    version only for the time being, or are under test in isolation with
    suitable drivers which are not included in the released version.
    Please ignore them.


    All FLUKA versions older than Fluka2005.6 (Fluka2003.1b included), and
    starting since 1989, are declared obsolete and will no longer be
    Therefore they shall no longer be used for any publication according to
    the FLUKA User License and associated Requests of the Authors.


    The use of the FLUKA code must be acknowledged explicitly by quoting
    at least the following set of references

         - A. Fasso`, A. Ferrari, J. Ranft, and P.R. Sala,
          "FLUKA: a multi-particle transport code",
           CERN 2005-10 (2005), INFN/TC_05/11, SLAC-R-773

         - A. Fasso`, A. Ferrari, S. Roesler, P.R. Sala,
           G. Battistoni, F. Cerutti, E. Gadioli, M.V. Garzelli,
           F. Ballarini, A. Ottolenghi, A. Empl and J. Ranft,
           "The physics models of FLUKA: status and recent developments",
           Computing in High Energy and Nuclear Physics 2003 Conference
           (CHEP2003), La Jolla, CA, USA, March 24-28, 2003, (paper MOMT005)
           eConf C0303241 (2003), arXiv:hep-ph/0306267

    Additional FLUKA references can be added, provided they are relevant for
    this FLUKA version.

    If FLUKA is used together with rQMD-2.4, DPMJET-2.53, or DPMJET-3 the
    following references should be quoted:


         - H. Sorge, H. Stoecker, and W. Greiner, Annals of Physics 192, 266 (1989)


         - J. Ranft. Physical Review D51, 64 (1995)


         - S.Roesler, R.Engel, J.Ranft: "The Monte Carlo Event Generator DPMJET-III"
           in Proceedings of the Monte Carlo 2000 Conference, Lisbon, October 23-26
           2000, A. Kling, F. Barao, M. Nakagawa, L. Tavora, P. Vaz eds.,
           Springer-Verlag Berlin, 1033-1038 (2001).

                     The FLUKA development team

      === Part of the Release Notes for Fluka2005.6 are reported here ===
      === for convenience ===
      === All provisions not explicitly overriden in the above text ===
      === are still valid ===

    This release is a quantum leap forward in the FLUKA development with
    respect to the last release, Fluka2003.1b, both from the technical and
    scientific point of view, and for the interaction with the users.



    A good fraction of the new features of FLUKA as presented in recent
    conferences are in, some of them are still out just because they are still
    felt as experimental.

    The most relevant new features are the following:

    a) Radioactive products online evolution and associated remnant doses

    b) PEANUT extension to pbar/nbar and, for p,n, pi's, pbar/nbar
        elimination of Nucriv. Among the many consequences of this development,
        the threshold for nbar transport and interactions can now be set as low
        as the user wishes (!!! Please do no longer use 50 MeV, and please,
        note that the threshold for Kshort/long transport can be as low as
        the user wish since 2000 !!!)

    c) ElectroMagnetic dissociation of heavy ions

    d) Elimination of PEMF

    e) New photon cross sections obtained out of EPDL97 (LLNL)

    f) New photon coherent scattering model

    g) Updated photon photoelectric effect model

    h) Updated photon pair production model

    i) Elimination of the OUTLEVEL and ACCURACY input options

    l) Obsoleting the EXTRAWEI option (see USERWEIG)

    m) Parentheses in geometry (eventually!)

    n) Introduction of (simple) preprocessor directives in the input stream

    o) New random number generator

    p) New routines for mathematical special functions (adapted from SLATEC)

    q) Interface with DPMJET-3 (the one with DPMJET-2.53 is still



    A major cleanup of FLUKA commons and routines has been performed, trying to
    make the code structure more rational and symmetric (among the various
    parts) in its implementation. A significant amount of obsolete code has been

    Problems with variable or commons which no longer exists or have
    changed their names could occur within preexisting user routines.

    In particular:

    a) as a consequence of the elimination of PEMF, all material properties
        which were given in the PEMF input file are now taken from the FLUKA
        input file. The user MUST provide only the production thresholds
        for photons and electrons/positrons (the AE and AP parameters of
        the PEMF input file) using the EMFCUT card with sdum = PROD-CUT.
        Forgetting to introduce this card into your old inputs could
        cause wrong results. The code will indeed select defaults for
        the production cuts which could be easily far different from the
        ones previously used in the corresponding PEMF file.
        Please look at the manual for details about how to input the
        EM production threshold with EMFCUT. The "-p" option in the rfluka
        script has been kept for future extensions, where an automatically
        generated data file will be produced in order to speedup the
        initialization of further runs with the same input. Please don't
        use it.

    b) the common BEAM has been renamed BEAMCM and the following variable
        name changes occurred:


    c) the Fluka and Emf stacks have been completely reshuffled. For most
        user routines, it should be enough to include (FLKSTK) where
        (STACK) was included and perform the following modifications:

        XA -> XFLK TX -> TXFLK TKE -> TKEFLK
        YA -> YFLK TY -> TYFLK ILO -> ILOFLK
        ZA -> ZFLK TZ -> TZFLK LO -> LOFLK

        Please be careful that there are new variables which are critical
        and which need to be set for the proper operation of the code, ie
        when a user source routine is used

    d) the lattic user routine, which is needed for the exploitation of the
        repetition capabilities of the FLUKA geometry, has now one further
        argument in the calling sequence. User should adapt their lattic
        versions, looking at the example supplied inside usermvax

    e) for those entitled to download the source version: the source
        distribution contains a Makefile which allows building the code
        on Linux-x86 (x86_64, with the -m32 option as required by the binary
        files). In order to run the code successfully one has to download
        the binary version as well (which contains the data files, the
        auxiliary tools etc, which are not duplicated in the source tar file


    |  Alfredo Ferrari                ||  Tel.: +41.22.767.6119                  |
    |  CERN-AB                        ||  Fax.: +41.22.767.7555                  |
    |  1211 Geneva 23                 ||  e-mail:        |
    |  Switzerland                    ||     |

  • Next message: "delayed (n, gamma)"

    This archive was generated by hypermail 2.1.6 : Wed Mar 28 2007 - 09:01:17 CEST