During the development of the European Rail Traffic Management System (ERTMS), several changes and improvements have been made to the driver–machine interface (DMI). The intention is to refine the train driver support by adding, removing, or clarifying signal information. In ERTMS Baseline 3, major DMI changes have been made between onboard versions 3.4 and 3.6. This includes clarification about speed targets and a simplified color strategy for target speed monitoring. This paper presents the effects of the DMI version shift on running time, driver target braking toward a European Train Control System target, and driver workload. In an electrical multiple unit train driver simulator, 39 student train drivers tested two different DMI versions on a 16 km railway line. In addition, the drivers rated the driver task workload using the NASA Task Load Index. From the results of this study, it can be concluded that drivers, despite a longer braking phase, go faster with version 3.6 than with version 3.4. The running time difference was about 1%. Although the driving task was short, with a low workload demand, the train driver workload was clearly affected by the onboard version, such that with the later version (3.6), the workload was statistically significantly lower.