Minutes for the terameeting of May 7th 2007

Present: Chiara, Mireille, Yannick, Henry, Emmanuel, Frederic, Gregory.

Emmanuel's summary document

CFHTLS T04 processing
 According to Fred, the current strategy for the transfer of weight-maps between TERAPIX and CADC (through NFS) puts a high load on clix and mix8.
 4 bands from D1-25% are done (z is missing). For the z band Fred does it by hand, for other deeps Henry has to fix the seeing rounding problem in his script. He will first redo the other 25%. In order to have a reference catalog, the extent of the output stack will not be limited to 1 degree.
 There is a memory problem to compute astrometric solutions for the Deep-85%. Mireille and Henry suggest to make a stack from existing deep and wide data around D1 and use it as an astrometric reference catalog. Emmanuel suggests to try to solve the astrometry for the deeps using INSTRINSTRU_KEY FILTER (dropping the QRUNID dependency to cut down the number of degrees of freedom), as most of the memory is actually eaten up by the normal equation matrix.
 u-band problem: according to Emmanuel it should be possible to trace the evolution of the zero-point in the scamp.xml from the W1 calibration.
 The astrometric comparison between T03 and T04 wide is made more difficult because the ldac files produced by SPICA for T03 have officially been erased. Hopefully some LDACs from Yannick and others have been recovered by Fred. Mireille will regenerate the lost T03 ldacs if necessary. Henry has a script to run SCAMP and do the comparison.

MEGACAM P.I.
 Mireille will do the quality check on the Tully data after the QFITS-in have been completed, and will do the processing (stacking image-by-image) during the week.

Software development
 Gregory has started working on the identification of large scale defects. He might receive some help from Anthony.
 According to Chiara, the impact of area computations on the processing time is less than 10% in Weightwatcher. The computed fraction of good pixels can be used for image quality assessment.

Hardware/system
 Yannick says that Fred will start cleaning up the cluster and installing new machines this week.
 Frederic says he will do the security update to MyBB this week.

Miscellaneous
 The AI/IE profile has to be written today.

WIRCAM
 Henry promises to run Herve's scripts with MAG_APER measurement today.
 Yannick will not have the time to comment on WIRCAM processing document this week, but he will do next week.