komatits
commented
6 years ago Dear Lei,
Thank you very much! Very useful. I will make the changes in the official version of the code tomorrow.
Best regards, Dimitri.
On 05/09/2018 09:51 PM, Lei Zhang wrote:
To whom it may concern， This is Lei Zhang, a user of SPECFEM3D from Department of Hydraulic Engineering, Tsinghua University. Thank you for having been developing this package. I found two tiny bugs in /src/shared/read_parame ter_file.F90. They occur only when USE_EXTERNAL_SOURCE_FILE is set to TRUE. (1) LINE 1156: we should skip 10 lines when USE_EXTERNAL_SOURCE_FILE is .true. because there is an extra line for source time function file name. (2) LINE 1137: "NSOURCES = icounter / NLINES_PER_CMTSOLUTION_SOURCE" should be put into the if-then-else-endif between LINE 1129 and LINE
- Besides, LINE 1134: (NLINES_PER_FORCESOLUTION_SOURCE+1) should be (NLINES_PER_CMTSOLUTION_SOURCE+1). (3) BTW, at page 40 of the pdf manual, the sentence “the source time function file must contain 2 columns with the current time and the amplitude of the soure function” might be something inexact. Because there should be only ONE column containing the amplitude of the source at that time step, and on its first line it must contain the time step used. I think the comment at the Sources section in Par_file is better. Attached is a fixed version of (1) and (2). I am a freshman in SPECFEM3D, so please correct me if I am wrong. Regards, Lei read_parameter_file.F90.zip https://github.com/geodynamics/specfem3d/files/1989261/read_parameter_file.F90.zip
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/geodynamics/specfem3d/issues/1235, or mute the thread https://github.com/notifications/unsubscribe-auth/AFjDKXIrzXGfWd36ZR0bijXqb3VYfSFTks5tw0iwgaJpZM4T45d9.
-- Dimitri Komatitsch, CNRS Research Director (DR CNRS) Laboratory of Mechanics and Acoustics, Marseille, France http://komatitsch.free.fr
To whom it may concern， This is Lei Zhang, a user of SPECFEM3D from Department of Hydraulic Engineering, Tsinghua University. Thank you for having been developing this package. I found two tiny bugs in /src/shared/read_parame ter_file.F90. They occur only when USE_EXTERNAL_SOURCE_FILE is set to TRUE. (1) LINE 1156: we should skip 10 lines when USE_EXTERNAL_SOURCE_FILE is .true. because there is an extra line for source time function file name. (2) LINE 1137: "NSOURCES = icounter / NLINES_PER_CMTSOLUTION_SOURCE" should be put into the if-then-else-endif between LINE 1129 and LINE 1135. Besides, LINE 1134: (NLINES_PER_FORCESOLUTION_SOURCE+1) should be (NLINES_PER_CMTSOLUTION_SOURCE+1). (3) BTW, at page 40 of the pdf manual, the sentence “the source time function file must contain 2 columns with the current time and the amplitude of the soure function” might be something inexact. Because there should be only ONE column containing the amplitude of the source at that time step, and on its first line it must contain the time step used. I think the comment at the Sources section in Par_file is better. Attached is a fixed version of (1) and (2). I am a freshman in SPECFEM3D, so please correct me if I am wrong. Regards, Lei read_parameter_file.F90.zip