How to use FGM High Resolution Software (Linux only)
| 1. | Copy to your linux box FGMdata/ and contents. I suggest that you put the files in /home/pdh/FGM. |
| 2. | Copy to your linux box bin/, bin_ds/ and contents |
| 3. | Create a folder called calibration and in this make a link called daily. This is a link to the directory where the calibration files are stored on hx. They are stored in /disk/peace1/projects/cluster/doc_server/FGM/fgm_cals_at_jsoc/. There are many ways to do this but I have created a link to this folder on my hx homespace, then linked to this link via my hx samba mount on my desktop. |
| 4. | Copy to your linux box calibration/default and contents. If there are no daily calibration files these will be used for cluster. The default FGM cal files provide a fall-back calibration set that will be picked up by the processing software for days where day -cal files are not available. Since the offsets drift with time a valid day-cal file should always be used in preference to the default cal files. |
| 5. | Copy to your linux box calibration_ds/ and contents (where the double star calibration files are stored). |
| 6. | Copy to your linux box fgm_c _v2- the idl program that calls the binary files to process cluster data. The program copies the two cds of data from hx and gunzips then untars them. The prog will process burst mode or normal mode as you say. If your period is on 2 cds (i.e. the whole day or a section over the break) two cdf files will be produced. Note that if you want a period just on cd1 (or cd2) the program will still try to Qtran files from cd1 and cd2. You will get an error but you can safely ignore this. See the pdf file in the FGM directory and: http://www.geophys.tu-bs.de/forschung/projekte/cluster/cluster2/clcoi/dp_example.html |
| 7. | Copy to your linux box fgm_ds_v2 - the idl program that calls the binary files to process doublestar data. Similar to fgm_c but the data is not on 2 cds. For TC1 you can only produce prime parameters (spin resolution), but for TC2 you can produce higher resolution. |
| 8. | Copy to your linux box QSAS_HEADERS and contents. These are used in Qtran to convert the produced ASCII files to .cdf files. |
| 9. | Copy to your linux box config_fgm.csh and config_fgm_ds.csh These paths are used by the progs in bin/ and bin_ds. You must source the correct file before you run the programs (note that these set different paths to the same name - i.e. if you source config_fgm.csh then use fgm_c you are ok. If you then try to use fgm_ds without sourcing config_fgm_ds.csh you will get errors) |
| 10. | Copy to your linux box proc_data. Where the processed data is stored. |
| 11. | Copy to your linux box String1.pro, dmy.pro and dmy.pro. These functions are useded in the progs. |
| 12. | Copy to your linux box the newest Qtran. Its in the directory there, you should not need to make/compile it, it should just run. You will need to replace your old version with this or you will not produce .cdf files. |
| 13. | If you have burst mode data and normal mode data you can combine them with Rob F's prog, fgm_join.pro. |
| 1. |
Modify the files config_fgm.csh and config_fgm_ds.csh. If you have put the files in /home/you/FGM/ you just need to change 'pdh' to your username. |
| 2. |
Modify the files fgm_c_v2.pro and fgm_ds_v2.pro. You will have to modify lines 16, 60 (again, you need to make a link to /projects/cluster/clusterops/rawdata/, for me its via my samba mount in /mnt/msslhx/rawdata), 61, 64 in fgm_c.pro, and lines 15, 57 (a link to /sddas/idfs_archive_on_hx/DOUBLESTAR/ will have to be created, for me it is via /mnt/msslhx/DOUBLESTAR/), 59, 61 in fgm_ds.pro. |
| 3. |
To get Double Star data, type: source config_fgm_ds.csh idl when this starts, type: IDL> fgm_ds_v2 and follow instructions. |
| 4. |
To get Cluster data, type: source config_fgm.csh idl when this starts, type: IDL> fgm_c_v2 and follow instructions. |
Return to PEACE CoI Resources Page
Last updated on 19-Oct-2006 by Andrew Fazakerley