The paper presents a numerical study on the application of Miller cycle to a turbocharged two-stroke marine diesel engine. Different timings of exhaust valve closure were investigated, while 2-stage turbocharging and exhaust gas recirculation were also considered. 1-D and 3-D simulation models were used to evaluate NOx emissions, fuel consumption and other engine parameters. Though the subject is interesting, the manuscript has to be revised in order to be accepted.

• Introduction, line 39: NOx generation is related to maximum combustion temperature and O2 availability, but only the first aspect is mentioned.
• Introduction, line 63: what about PM penalty due to EGR application?
• Figure 1: caption is too generic. It is stated that 1-D simulation allowed to define initial conditions for 3-D calculation (line 80), but in the scheme results are also derived from 1-D model. Please better specified this point.
• Table 2 and fig.4: are data referred to full load condition? No information is given about this point.
• Line 101: “TDC is defined as 360 °CA”, but TDC shown in fig.4 is at 0 °CA.
• Line 115: two injectors are quoted, but no information is given on fuel injection system in Table 1.
• Lines 117-118: it is well known that the combustion temperature influences NOx formation. Moreover, why figure 6 is related to a short interval (from 364 °CA to 378 °CA) with a 2 °CA step, while figure 7 show results for a long range (from 370 °CA to 400 °CA) with a 10 °CA step? It is not possible to compare graphs, except for 370 °CA. In this case, there is no overlap between zones of maximum temperature and NOx levels.
• Line 144: what is CF?
• Figure 11: mean temperature curves are useless, as they don’t allow to describe the strong variations of in-cylinder temperature. Moreover, please take into account that presented values are below or very close to 1600 K, which is a common threshold to consider NOx formation.
• Figures 12/13: as for figures 6/7, there is no overlap between zones of maximum temperature and NOx levels.
• Table 3: power reduction due to the application of Miller-cycle is not considered in the discussion. Which are the reduction percentages of this parameter referring to the baseline condition? NOx variations are referred to volumetric concentrations or to specific emissions?
• Line 165: please justify the statement on 2-stage turbocharging influence on charger efficiency and fuel economy (authors calculation or references).
• Figure 15: authors should justify observed results. What about engine power?
• Line 177: please check figure 15 caption.
• Figure 16: no information is given about the selected EGR layout.
• Line 194 (caption of figure 17): what does it mean EGR rate 5 and EGR rate 10? Which is the measuring unit of EGR rate and how was EGR rate defined and computed?
• Figures 17/18: see comments on figures 6/7 and 12/13.
• Figure 19: one of the symbol is not reported in legenda (red triangle with cyan contour).
• Line 217: what is BFOC?
• What about PM Emissions?

Authors worked to revise the first version of their manuscript, but there are still several open issues. - Model validation: it is not enough to write (lines 119-120) that NOx calculated values were validated through measurements on the test bench. The comparison should be presented, while giving information on instruments and test rig. - Remark “there is no overlap between zones of maximum temperature and NOx levels”. Response: “The formation of NOx is lag compared to the variation of temperature”. No details are included in the manuscript about this point. - Fig.11: I’m still of the opinion that mean temperature curves are useless and misleading as I don’t’ agree with the authors’ response “mean temperature curves are still a reference to analyze the extent of the reduction of combustion temperature”. - The last remark on PM emissions aimed to check if PM calculation were made by the authors and to include some considerations on this aspect in the manuscript. None of these answers were given.