Safe-interval path planning (SIPP) is a powerful algorithm for finding a path in the presence of dynamic obstacles. SIPP returns provably optimal solutions. However, in many practical applications of SIPP such as path planning for robots, one would like to trade-off optimality for shorter planning time. In this paper we explore different ways to build a bounded-suboptimal SIPP and discuss their pros and cons. We compare the different bounded-suboptimal versions of SIPP experimentally. While there is no universal winner, the results provide insights into when each method should be used.