Claus Vohwinkel

The mass of the T2+ glueball in SU(2) Yang-Mills theory in intermediate volumes from an analytic calculation

Abstract Koller and van Baal carried out an intermediate volume calculation for pure SU(2) gauge theory. I re-analyze the boundary conditions they imposed on wave functions, and obtain a mass for the T2+ glueball which is in agreement with recent Monte Carlo results.

Tunneling and deconfinement

Abstract For non-abelian lattice gauge theory in a finite box one finds, as the box size increases, a rapid increase of the string tension due to tunneling between different toron sectors. The tunneling transition is a signal for the deconfinement phase transition. For various lattice shapes, we compare numerical results for this signal with those obtained from the popular analysis of Polyakovloop expectation values.

A fast cyber 205-ETA10 program for SU(3) lattice gauge theory☆

Abstract We describe a Monte Carlo program that simulates SU(3) lattice gauge theoryon CYBER 205 and ETA 10 computers. The program uses the Cabibbo Marinari algorithm for the updating and a novel way of gathering neighbors. It achieves a 17.1 μs updating time per link on a 2-pipe CYBER 205 using 32-bit arithmetic. The updating time per link on the ETA 10 at Florida State University is 8.8 μs.

A fast method to gather neighbors in vectorized Monte Carlo simulations

Abstract An algorithm is presented, which gathers neighbors for n -dimensional lattices with periodic boundary conditions. The method used does not need the usual GATHER instructions. With the exception of a few bit-vectors no additional storage is needed. The algorithm can be easily extended to more than nearest neighbor interactions. In a simulation of SU(3) lattice gauge theory on a 2-pipe CDC CYBER 205, using 32-bit arithmetic, the time spent on gathering neighboring links is reduced from 5.7 μs to 1.3 μs per link per update, and an update time of 17 μs per link is obtained.

Monte Carlo calculation for systems consisting of several coordinate patches

Abstract I investigate the time step dependence of Monte Carlo simulations for coordinate spaces consisting of several patches. It is shown that a naive kinetic term does not necessarily converge to the same spectrum as a Hamiltonian calculation. Then an improved kinetic term is presented which allows one to connect the Monte Carlo and Rayleigh-Ritz results of intermediate volume SU(2) gauge theory.

Non-Abelian gauge theory : mass spectrum and deconfining temperature. I

Abstract Using the small volume expansion the mass spectrum and deconfining temperature of 4 d non-Abelian gauge theory are evaluated. Including non-perturbative features by restoring symmetries which were broken in perturbation theory we obtain results which are valid up to intermediate volumes. The mass spectrum obtained is in good agreement with Luschers small volume limit and with Monte Carlo Data for medium sized volumes. Asymmetric volumes enable us to calculate the deconfining temperature and we find reasonable agreement with Monte Carlo calculations. Paper I outlines the general concepts and presents results for the SU (2) gauge group. Paper II shall generalize the approach to SU (3).

The mass of the T2+ glueball in SU(2) Yang-Mills theory

Abstract Koller and van Baal carried out an intermediate volume calculation for pure SU(2) gauge theory. I re-analyze the boundary conditions they imposed on the wave functions, and obtain a mass for the T2+ glueball which is in agreement with recent Monte Carlo results.