This paper investigates the sensitivity of a simulated tropical precipitation climatology focusing on the intraseasonal oscillation (ISO) to four convective parameterization schemes: simplified Arakawa-Schubert (SAS), relaxation Arakawa-Schubert (RAS), new Kain-Fritsch (KF2), National Center for Atmospheric Research (NCAR) Climate Model version 3 (CCM). An 8-year boreal summer climatology from 1997 to 2004 is constructed using an oceanatmosphere coupled global climate model (GCM). The simulated tropical precipitation climatology shows that all four experiments capture the observed climatology fairly well, with the pattern correlation coefficients greater than 0.8. The ensemble mean of results from the four experiments does not reveal a benefit in reproducing the observed precipitation climatology or the ISO signals. Although the KF2 scheme has been most widely tested and updated in mesoscale modeling communities, its capability in tropical climate simulation is shown to be relatively good in terms of sea surface temperature (SST) and precipitation. Results from the SAS and KF2 schemes show similar patterns in terms of climatology and ISO signals, with a greater precipitation variance than that from other experiments. The ISO signals from the RAS run show relatively realistic ISO signals, but with too strong intensity. Our study implies that the appropriate partitioning of deep convection due to cumulus parameterization scheme and stratiform precipitation due to microphysics scheme should be taken into account when developing or revising physics algorithms in coupled GCMs.