Difference between revisions of "Overlap"
From Gw-qcd-wiki
| Line 5: | Line 5: | ||
<math>\mathbf{D_0} = \rho(1+\gamma_5)\epsilon(H)</math> | <math>\mathbf{D_0} = \rho(1+\gamma_5)\epsilon(H)</math> | ||
| + | Where <math>\rho 4 - 1/2\kappa</math>, <math>\kappa </math> is a positive number. <math>H</math> is the hermitian Wilson operator, and <math>\epsilon(H)</math> is the matrix sign function. | ||
Given a vector <math>\vec{\eta}</math>, we want to compute the solution <math>\vec{x}</math>, which is the matrix vector multiplication | Given a vector <math>\vec{\eta}</math>, we want to compute the solution <math>\vec{x}</math>, which is the matrix vector multiplication | ||
<math>\mathbf{D_0} \vec{\eta} = \vec{x} </math> | <math>\mathbf{D_0} \vec{\eta} = \vec{x} </math> | ||
Revision as of 09:28, 31 August 2010
The intention of these notes is to outline how we construct the overlap operator in the gwu-qcd framework. The overlap operator preserves chiral symmetry on the lattice, and is the most ideal operator used to explore low pion masses. The massless overlap is defined by
\(\mathbf{D_0} = \rho(1+\gamma_5)\epsilon(H)\)
Where \(\rho 4 - 1/2\kappa\), \(\kappa \) is a positive number. \(H\) is the hermitian Wilson operator, and \(\epsilon(H)\) is the matrix sign function. Given a vector \(\vec{\eta}\), we want to compute the solution \(\vec{x}\), which is the matrix vector multiplication
\(\mathbf{D_0} \vec{\eta} = \vec{x} \)
