Dear reviewer,

thank you very much for your comments to my article. I have revised the manuscript according to your recommendations. Let me briefly reply to your main comments below.

---> Since Figs. 3, 4, and 5 take up 2.5 pages of the manuscript their discussion should be expanded.

The reason why I did not really discuss figures 3, 4 and 5 in the article was that they represent the result of a common effort of several authors - but I write this review article on my own. I thus feel that I am not in a position to deeply interpret the comparison and draw conclusions which my colleagues might disagree with.

I agree with you, however, that the figures deserve more discussion. I have thus added a paragraph describing what can be seen (that the full Monte Carlos CoREAS and ZHAireS agree well, the histogrammed models show slight deviations from the full Monte Carlos and the microscopic models predict significantly stronger pulses). I hope this is a reasonable compromise between putting the results into perspective for the common reader but not imposing my own judgement, possibly annoying my colleagues involved in the comparison.

The comment in the figure captions that the showers and time resolutions were not identical was somewhat misleading. I wanted to include this for the experts so that they do not over-interpret details on a few-percent level. But to the common reader this might indeed have sounded as if a comparison of the results was not possible at all. The differences arising from this are, however, of only minor importance (see below for more details), so I thought it best to entirely remove the comment.

I hope with these changes the manuscript is fine for publication. I also have to stress that I cannot add any more text without removing other text (the change of citation style for "these proceedings" cost a lot of space).

---> It is stated that the time resolution is different btw. models, so
     probably also the bandwidth is. Is it possible to compare the pulses for
     the same bandwidth.
  
This has only an effect on the observed "noise" in the time-pulses. The amplitudes are not significantly affected as there is only negligible power at the relevant frequencies for the figures presented here.
  
---> It is stated that the longitudinal profile is different. Does this mean,
     that also Xmax is different for different showers? How does this influence
     the pulses? Would it be more reasonable to compare iron induced showers
     (if available) instead of proton, since the spread in Xmax is expected to
     be smaller.
 
The modelers made sure that they did not simulate a strongly fluctuating shower but rather picked "typical" shower. Remaining differences in the Xmax values will lead to slightly steeper or flatter lateral distributions, but the differences should be small for the examples shown here. In the future, the comparison should be made for properly preselected showers - simulations of iron showers could indeed also help.