The determination of design wind pressures for component and cladding loads is a crucial aspect of building design. In the case of high-rise buildings, these pressures are often assessed through wind tunnel tests. While several methods exist to estimate peak pressures, the Gumbel distribution, and the generalized least square method, also known as the BLUE method, are commonly employed for estimating peak pressure data. The wind tunnel test duration and the number of peaks considered in the estimation are key parameters in this process. While guidelines for low-rise building peak pressure estimation are commonly followed in the case of high-rise buildings, it is essential to acknowledge the unique challenges posed by the latter. High-rise buildings typically involve smaller model scales compared to low-rise buildings, necessitating special considerations in peak pressure investigation. For reliable statistical analysis, recent studies have proposed the use of extended duration to estimate one-hour statistics more accurately for reliable peak pressure coefficients. The objective of this study is to investigate how varying the number of peaks utilized in peak estimation impacts the estimated peak pressures for data with both highly and mildly non-Gaussian characteristics, specifically focusing on one-hour length datasets. Additionally, the study aims to explore the non-Gaussian features of high-rise buildings across different building heights.