VNL-ATK 2014.3 更新发布

Posted · Add Comment

VNL-ATK 2014.3更新版本正式发布。此次更新修正了若干程序错误。请在更新维护期内的用户务必更新到最新版本。

主要变化

  • 自旋极化、旋轨耦合、和非共线模型的许多修正
  • BSSE和Grimme势现在可以一起使用
  • nlsave和nlread现在可以正确保存和读取MGGA-TB09的c值
  • ATKClassical、TremoloX的更新

下载地址

  • http://www.quantumwise.com/download

更新详情

Serious bugs (frequent, or affecting calculation results, or causing large disruptions in core functionality)
  • Spin-orbit, spin-polarized and non-collinear models corrected and improved
    • Updated how all 3D grid objects are handled w.r.t. spin for consistency and correctness. For instance, you no longer specify the spin parameter when you compute the grid, only when you query it (via the evaluate method).
    • Spin-orbit model for self-consistent semi-empirical Slater-Koster and Huckel corrected
    • Correctly handle the imaginary UPUP/DOWNDOWN part in sparse Fourier transforms (8975)
    • Spin-polarized electron density for the extended Huckel model corrected (10104)
    • Fixed non-collinear dense transmission coefficient (now L->R) (9318)
  • Fix for DeviceDensityOfStates with spin-orbit (9027)
  • BSSE and Grimme now work together
  • Neumann boundary conditions for FFT2D
    • Fixed for bulk configuration (9186)
    • Disable double Neumann in any direction (9231)
  • LocalStress, several bugs fixed
  • nlsave/nlread MGGA-TB09 calculations now saves and restores the “c” parameter properly (9430)
Bugs (infrequent or not affecting results, but annoying when you hit them)
  • Effective mass now requires that you specifies a direction, i.e. we no longer compute the tensor and then project, since this doesn’t work for degenerate bands
    • Also corrected the fractional coordinate in the report
  • Improved error message for phonon calculations using semi-empirical models without a pair potential (9513)
  • 1D projector constant CPU usage fixed (9226)
  • Restart polarized calculation from unpolarized initial state (9418)
  • Fixes for bands_above_fermi_level in Bandstructure/DensityOfStates (8987)
  • Make GridValues.evaluate/derivatives always return a PhysicalQuantity (10062)
  • ATKClassical/TremoloX
    • Bugfix for potentials with tags
    • Corrected a parameter in the Oligschleger_Se_1996 potential
    • Minor updates in the ATKClassical plugins
    • Fixed small error in the reference manual for the Morse potential