CCL: New release of the ORCA program

 Sent to CCL by: "Frank  Neese" [neese::thch.uni-bonn.de]
 Dear CCLers and ORCA community,
 We are happy to announce a new release of the general purpose quantum
 chemistry program ORCA (version 2.7.0) as a beta release. As indicated
 by the beta status we do not exlude the possibility that minor
 adjustments have to be made in the coming weeks. Please let us know if
 you run into bugs and inconsistencies! As always, we appreciate positive
 as well as negative feedback.
 This release presents numerous improvements as well as new features that
 should make the code much more attractive for large-scale computational
 chemistry applications. With this release we believe that we have
 reached state-of-the-art performance (or better) in all major code
 branches (HF, DFT, MP2, CCSD(T); within the still persistent limitation
 to C1 symmetry). Thus, the emphasis has shifted somewhat away from
 specialized spectroscopy applications to general large-scale
 computational chemistry.  We have, of course, also fixed the bugs that
 have come to our attention.
 The number of ORCA users continues to grow very fast and the list has
 now reached more than 2500 registered individuals, research groups and
 computer centers worldwide. We encourage users who have tried ORCA
 before and were not fully convinced to test it again.
 The program can be downloaded free of charge (for academic users) at:
 http://www.thch.uni-bonn.de/tc/orca/
 There are serial and parallel version of Linux and Mac OS X available as
 well as a serial version for Windows. Parallelization is based either on
 MPICH or (on request of several users) also for OpenMPI.
 In addition to an extensive manual (>400 pages), we now provide a
 "Jump-start" guide to the use of ORCA together with the executables of
 the program.
 RI-DFT:
 ------
 * The numerical integration has been completely rewritten and is up to
 30% faster than before.
 Hybrid-DFT, double hybrid DFT,  Hartree-Fock
 -------------------------------------------------------------
 * The entire ORCA program now works up to L=8 in the orbital and fitting
 bases.
 * We are very happy to have entered a collaboration with Prof. Ed.
 Valeev who has generously contributed his LIBINT integrals to ORCA. For
 TZV(2df,2pd) (or cc-pVTZ) the performance gain is roughly a factor of 2
 and for QZV(3d2f1g,3p2d1f) (or cc-pVQZ) it is  factor of four or more.
 * The RIJCOSX approximation greatly speeds up HF and hybrid DFT
 calculations (up to a factor of fifty for large basis sets and
 molecules) while maintaining chemical accuracy in the total energies. It
 is available for energies, gradients and response calculations including
 TD-DFT.
 * The RI-JK approximation proposed by Weigend et al. is available for
 RHF and UHF calculations and also significantly speeds up Hartree-Fock
 and hybrid DFT calculations.
 * Specialized code for generally contracted basis sets has been developped.
 * The parallelization has been improved.
 MP2 and RI-MP2
 -------------
 * Efficient (and parallel) frozen core gradients
 * Improved parallelization
 Coupled-Pair and Coupled Cluster
 ---------------------------
 * RHF and UHF based CCSD(T), QCISD(T), CEPA and CPF methods are now
 fully optimized and show state-of-the-art performance
 * AO direct mode, AO conventional mode, full integral transformation, RI
 approximation for balancing memory, CPU and disk requirements.
 * Parallelization has been done, but data is replicated on all nodes
 * MP3 and SCS-MP3 are now available
 CASSCF
 ------
 * Much improved RI transformation for efficient calculations on large
 molecules
 * RIJCOSX and RI-JK approximations for efficient Fock operator
 approximations
 * Improved parallelization
 * Slightly improved convergence behavior (but still room for improvement)
 MR-CI and MR-MP2
 ---------------
 * Specialized MR-MP2 code. Efficient for not too large reference spaces.
 * Magnetic fields, Spin-orbit coupling, electron-electron spin-spin
 coupling.
 * Picture change effects in MRCI-DKH2 spin-orbit calculations
 * Manual excitation energy input for quasi-degenerate perturbation theory
 * Individual natural orbitals for each state
 Optimization
 -----------
 * Improved coordinate setup and detection of linear angle problems
 * Improved constrained optimization and relaxed surface scan calculations
 * Hybrid Hessian for large transition state optimizations
 * Fragment based optimization
 * Minimum energy crossing point (MECP) location between two potential
 energy surfaces with SCF and CASSCF methods
 General features
 -------------
 * More rigid convergence thresholds and improved SCF Convergence
 * More extensive and consistent basis set library
 * Scalar relativistic basis sets for third row transition metals and
 lanthanides
 * Accurate atomic natural orbital basis sets for H-Zn.
 * Automatic extrapolation of the SCF and single reference correlation
 energies to the basis set limit.
 * Resonance Raman intensities, excitation profiles, absorption
 bandshapes, fluorescence bandshapes
 * X-ray absorption spectra with TD-DFT
 * Nuclear resonance vibrational spectra
 * Normal mode trajectory calculations
 * Generally improved parallelization in most parts of the program.
 * Convenient QM/MM interface to Gromacs
 * Semiempirical QM/MM calculations
 * CHELPG charges
 * Multiple temperatures in thermochemical calculations
 * Multiple XYZ file scan feature (e.g. series of single point
 calculations from relaxed surface scans)
 * Several new double hybrid functionals
 * More complete support for .47 files in NBO analysis
 * Finite temperature electron smearing for difficult SCF cases
 * Numerical gradient for high level correlation methods
 Coming later this year: (things that are currently being tested)
 ------------------
 * Effective core potentials
 * A variety of improved perturbation theoretical methods
 * Local correlation methods
 * Analytic TD-DFT gradients
 * .... surprise   :-)
 We hope that with this release we have further increased the
 attractiveness of ORCA for the computational chemistry community and
 will now intensify our efforts to produce a high-performance,
 user-friendly and versatile quantum chemistry program.
 With our best regards,
 The ORCA development team:
 Frank Neese
 Frank Wennmohs
 Ute Becker
 Dmitry Ganyushin
 Andreas Hansen
 Simone Kossmann
 Dimitrios G. Liakos
 Dimitrios Pantazis
 Taras Petrenko
 Christoph Reimann
 Christoph Riplinger
 Kanthen Sivalingam