arunkannawadi / kv450imsim Goto Github PK

@ifen , the source code of flensfit.c reads that it belongs to SVN revision 273. The KiDS-450 analysis was done with SVN 284. Could you double check the version installed here at Leiden?

Hi @ifen , the PSFmoments from the previous runs do not match that of the new runs, although I am using the same PSF sets. Massimo (and even Lance) had mentioned that there was a bug earlier which you had fixed. We doubt if that fix didn't make it to Leiden here. Can you point me to where the fix is, if it is indeed here?

@mviola and mine naive assumption was that the magnitudes are merely passed through lensfit for understanding the magnitude-size distribution but lensfit itself doesn't measure the magnitudes. Thus, the catmag field in the lensfit output catalogue was expected to be the same as the input magnitudes. We checked this for the observed data (THELI), and barring a few biased errors, they are fairly consistent with each other. However, on the simulations, they are drastically different. I doubt this is a bug in our pipeline though. Would you happen to know anything about this or is Lance the person to talk to? I didn't spend much time following through the C code to see what happens in there.

The implication of this is that when I compare the weighted histograms of MAG_AUTO s, they are very different compared to the unweighted histograms.