Right ventricular longitudinal strain (RVLS) by 2D speckle-tracking echocardiography (2D-STE) is a useful parameter for assessing systolic function. However, the exact method to perform it is not well defined as some authors evaluate only free wall (FW) segments while others include all six RV segments. To compare the assessment of RVLS at rest and during exercise by these two approaches. Echocardiography was performed on 80 healthy subjects at rest and during exercise. The analysis consisted of standard and 2D-STE assessment of RV global and segmental strain tracing only RVFW and also tracing all six RV segments. At rest, RVLS could be assessed in 78 (feasibility 97.5%) subjects by both methods. However, during exercise, RVLS by RVFW method was feasible in 67 (83.8%) as compared to 74 (92.5%) by RV6S approach. Both at rest and during exercise, RVLS values by the two methods showed excellent correlation (r =  > 0.90). However, RVLS values assessed by RV6S were lower (absolute values) than those by RVFW approach (RV6S vs. RVFW; rest: − 27.0 ± 3.9 vs. − 9.5 ± 3.9, p < 0.001 and exercise: − 30.7 ± 5.2 vs. − 33.3 ± 5.1, p < 0.001). Furthermore, basal strain was higher and apical strain lower (absolute values) by RV6S approach. At rest, reproducibility for RVLS was excellent and similar for the two methods. However, during exercise, reproducibility for RVFW method was poorer, especially at the apex. The two currently described methods for RVLS assessment by 2D-STE demonstrated excellent agreement. However, the RV6S approach seemed to be more feasible and reproducible, particularly during exercise. Moreover, global and segmental strain values are different with both methods and should not be interchanged.