work, work, work
Right, have continued and completed interpolation of the Grodent atmosphere. Now need to wrap a program round it, but first my mind, such as it is, turns towards the windows-born mess of a program that is MICdev - the display system for the spectrograph, created by a postdoc and improved in various ways since. I have now been taught in the dark arts of setting it up and off and my job is to get it working on a unix system.
Initial attempts yesterday were not promising. Asking it to compile is the easiest way to hear a computer laugh (after asking the latex template of the thesis to compile...). I was going to grab the apl laptop, but my one single bit of idl knowledge (that it uses unix and so hate uppercase letters) was brought into play first. I renamed every single file and now it works. To an extent. There are still problems.
Windows retain is not employed. Most annoying.
The output is logged to the terminal, but needs to be stuck into a file for easy reading.
Image output is to a strange file format. IanF assumes this means no images can be output, I will investigate the widget to see whether or not this is so. If it is, my own widget will replace the current one. If not, then I will tweak the output format into something a little more useful.
Flatfielding/dark current subtraction creates some interesting stuff to happen (due to over corrections with the flats). 'Correcting' a five second exposure with four hour long exposures certainly provides a flat field, but not quite the one expected...
Lets see if I can tackle this one in the future, but now I've got it working (and have tested with some data) I can at least use it as a display mechanism, to an extent.
ok, sorted out the last of the compilation errors...
retaiun=2 now stuck in 'spectro.pro', where previously the decompose=0 version was used. Now works fine, all images are retained. The program is usable! Usable enough to find the problem with darks and flats - just need to reautoscale the thing.
Right, now that works and has been tested enough (for now, twitching to do more, but have other jobs) back to Grodentish atmosphere.
Here we go again. Makefile:
randgen - no changes
atom - change to five species
cross - change to five species
electron - as above
strike - ditto
proton - ditto
proton1 - ditto
hit - ditto
back - ditto
selectron - ditto
recoil - ditto
So, ones that are ready:
randgen - done!
atom - done!
cross - recombination sorted, dissociation sorted
Still to do:
cross - scattering, ionisation and excitation related stuff
electron
strike
proton
proton1
hit
back
selectron
recoil
Initial attempts yesterday were not promising. Asking it to compile is the easiest way to hear a computer laugh (after asking the latex template of the thesis to compile...). I was going to grab the apl laptop, but my one single bit of idl knowledge (that it uses unix and so hate uppercase letters) was brought into play first. I renamed every single file and now it works. To an extent. There are still problems.
Windows retain is not employed. Most annoying.
The output is logged to the terminal, but needs to be stuck into a file for easy reading.
Image output is to a strange file format. IanF assumes this means no images can be output, I will investigate the widget to see whether or not this is so. If it is, my own widget will replace the current one. If not, then I will tweak the output format into something a little more useful.
Flatfielding/dark current subtraction creates some interesting stuff to happen (due to over corrections with the flats). 'Correcting' a five second exposure with four hour long exposures certainly provides a flat field, but not quite the one expected...
Lets see if I can tackle this one in the future, but now I've got it working (and have tested with some data) I can at least use it as a display mechanism, to an extent.
ok, sorted out the last of the compilation errors...
retaiun=2 now stuck in 'spectro.pro', where previously the decompose=0 version was used. Now works fine, all images are retained. The program is usable! Usable enough to find the problem with darks and flats - just need to reautoscale the thing.
Right, now that works and has been tested enough (for now, twitching to do more, but have other jobs) back to Grodentish atmosphere.
Here we go again. Makefile:
randgen - no changes
atom - change to five species
cross - change to five species
electron - as above
strike - ditto
proton - ditto
proton1 - ditto
hit - ditto
back - ditto
selectron - ditto
recoil - ditto
So, ones that are ready:
randgen - done!
atom - done!
cross - recombination sorted, dissociation sorted
Still to do:
cross - scattering, ionisation and excitation related stuff
electron
strike
proton
proton1
hit
back
selectron
recoil
0 Comments:
Post a Comment
<< Home