One-to-one comparison of the true and model-recovered age of the cold, warm, and hot orbit components for all mock data sets. Top left panel: results using model A and mock data with a MUSE-like spatial resolution from Zhu et al. (2020). Middle panels: results using model A and mock data with a CALIFA-like spatial resolution. Bottom panels: results using model B and mock data with a CALIFA-like spatial resolution. In the middle and bottom panels, the relative observational errors equal 5%, 10%, 15%, and 20% from left to right. The triangles, squares, and circles represent the mock data sets with inclination angles of 40°, 60°, and 80°. Blue, orange, and red markers denote the cold (λ_z > 0.8), warm (0.25 < λ_z < 0.8), and hot (λ_z ≤ 0.25) components. Their corresponding mean bias and statistical uncertainty are shown by the coloured texts and error bars in each panel. The dashed lines represent the equality of t_model and t_true, while the dotted lines are ±1 Gyr away from the dashed lines.