Real-time PCR assays using 3'-minor groove binder (MGB) or Taqman probes are widely used for clinical virological testing and mutation/polymorphism detection. We compared a 3'-MGB probe to a conventional Taqman probe for linearity, sensitivity, specificity and dynamic range. The performance of the two assays was compared using plasmids containing different mismatches or using human genomic DNA as a template. Comparable linearity and sensitivity were observed for the MGB and the Taqman probe assays. Using standard conditions, none of the assays were sequence-specific. Up to five mismatches generated a detectable signal in the Taqman probe assay. The performance of the Taqman as well as the MGB probe assay was influenced by the complexity of the template, the latter, however, to a lesser degree. Overall, these results highlight the advantages of the MGB probe over the Taqman probe regarding mismatch discrimination, but suggest that optimization of reaction conditions and verification of the specificity are necessary also for MGB probes.