Implement the following suggestions

[dmitry-kabanov]

The following suggestions from an anonymous reviewer must be implemented.

Evaluation of the ASOFI3D codes

The authors provided necessary components for using the codes: The installation guide, the test example, the user manual. They also implement experiments for checking the accuracy and efficiency of the modeling. Following is my personal comments and suggestions.

The test examples they provided cost a lot time. Although the installation is quite

easy by following the guidelines. It seems they do not well design or set parameters for the test examples, it takes a long time to finish in my workstation.

The manual is not updated regarding to the codes fully. Some of the folders such

as “genmod”, “overnightbuilt”, “scripts” described in the manual do not exist in the codes they provided. The authors need keep the manual and codes the same version.

The code only provided with standalone installation and lacks API interfaces.

However, we want to use it as a library. In this aspect, the codes are more or like a software with source code rather than an API which we can use in our codes.

A user manual with general introduction is not enough. They provided a user

manual which is good however is not enough. In order to use it conveniently, we need a manual in detail which gives description of each function used in the codes and their input and output should also be clearly specified. Some automatic documentation tool can achieve this, e.g., “Doxygen”.

The code is not well organized. In order to facilitate others, it is good to organize the source code and header files based on their functions. In the code they provided, it seems they just open a new header and source file as they like. No good structure.

The code is not clean and there are some comments left in the codes. This makes

people suspect the accuracy of the codes.

In general, the authors do make improvement to the original ASOFI3D codes. However, the code may be more useful for producing seismic records while for using in the inversion, specially using it as library and integrated with the FWI codes. It seems not user-friendly and inconvenient.

[vkazei]

1. FIx CI tests
2. “genmod”, “overnightbuilt”, “scripts” described in the manual do not exist in the codes they provided. The authors need keep the manual and codes the same version.
3. The code is not well organized. In order to facilitate others, it is good to organize the source code and header files based on their functions. In the code they provided, it seems they just open a new header and source file as they like. No good structure.
4. The code is not clean and there are some comments left in the codes.

[vkazei]

2. Fixed