In the ATLAS experiment at the Large Hadron Collider (LHC) at CERN, direct searches for the elusive Higgs boson will be conducted, as well as for physics beyond the Standard Model. The charged Higgs boson (H±) is interesting both as a part of the Higgs sector, and as a clear sign of new physics. This thesis focuses on H± searches, with H± production in top-antitop pair events, and in particular the  bW± bH±, H±→τhadν, W±→qq channel. Its potential was investigated  as  part of a larger study of the expected performance of the entire ATLAS experiment. Full simulation of the ATLAS detector and trigger was used, and all dominant systematics considered. It was shown to be the most promising H± discovery channel for mH±<mt. As hadronic τ decays are important for H± searches, their correct identification is critical. Possibilities of improving tau-jet identification in  pile-up and top-antitop pair events were investigated. Redundant or even performance-reducing variables in the default likelihood identification were identified, as were new variables showing discriminatory power. This allows for increased rejection of QCD jets in these environments, and higher robustness of the method. Before any physics studies, a commissioned and well-understood detector is required. The Lorentz angle of the ATLAS Semi-Conductor Tracker (SCT) barrel was measured using 2008 cosmic-ray data. It is an important observable for the performance of several detector aspects. Potential sources of systematics were investigated and evaluated. The Lorentz angle in the SCT barrel was measured as θL = 3.93 ± 0.03(stat) ± 0.10(syst) degrees, agreeing with the model prediction. The Compact Linear Collider (CLIC) is a proposed successor to the LHC. The potential for charged and heavy neutral Higgs bosons at CLIC was investigated, in terms of both discovery and precision measurement of parameters like tanβ or the Higgs masses, up to and beyond 1 TeV, which would be challenging at the LHC