Please can you respond to the points raised by the reviewers, and improve the paper as appropriate.

One of the reviewers remarks, "1) On page 1, lines 14 to 19: The authors still state that deriving the surface temperature of a planet "is a conceptually simple physics problem". As I said in my previous report, this is not true, and the authors do not need to say this to motivate their intercomparison project. To motivate their study the authors should rather argue that performing accurate radiative transfer calculations is the number one priority in any climate model.
"

It seems to me that you are using "conceptually simple" to mean that it is believed that the problem could theoretically be solved using known physics, given sufficient information and computational resource. You then state that the calculation is incredibly complex and that even the best estimates (that are too computationally expensive to be included in the models) presently diverge significantly from observations. I would not describe this is a situation of conceptual simplicity, so, on balance I agree with the reviewer and am hopeful that the introduction to the paper will be improved in clarity if you re-write the paragraph.

Sorry for the delay - I took a little holiday from GMD.

Testing climate model code for climates different from the present day is vital for understanding whether the code is adequate for the purpose of modelling climate change. In this case the atmospheric compositions being experimented with may be realistic for conceivable exoplanets as well as the past climate of the Earth. Despite the short expected timeframe of the experiment (maybe it may be extended..?) I hope that many modelling groups decide to contribute to this experiment! In future this work may be extended to a more comprehensive exploration of possible exoplanet atmospheres.