Status of T0003 (Laurent)

 First question, can we change the size of the image to make the same as the T0002?
 It seems that all images generated from the wide were *actually* the correct size;
 Discovered after the meeting: Laurent's script produces RECTANGULAR images for the deep and SQUARE images for the wide
 We will restart the release with a uniform image size for all images
 Yannick has compared in detail some stacks in T0002 and T0003 and they seem to be approximately OK, although more detailed comparisons are still needed
 It will take around 1.5 months to produce all the stacks
 It seems that all the stacks will be processed on two machines, mix4 and mix6 (mix3 will not be used); may be able to speed things up by copying images between the machines
 Fred wil explore the possibilty to move some data to mix3 in order to better use mix3 after its first (only) stack be done
 In order to check colors, once Laurent will have produced the command line list and the stack list table, Mireille, henry and Yannick will do a manual selection of the stack sequence. We will then produce one Deep and one Wide in u*,g,r,i,z prior to run the whole T0003.
 Yannick remind that Mathias, Henry, Mireille and Yannick have to check the output of each stacks produced by SPICA. This is a duty that must be shared by all of us.

Preparation for SPICA post-processing (Mireille, Henry, Mathias)
 Mireille will be at CFH from 11th January and 20th February
 Henry will be in Iran from the 11th to 20th of January
 All Henry's procedures and scripts must be well documented so that processing can continue when Henry is away
 Mathias will be Paris on the 6th and 7th of December

Summary table for T0003
 Laurent will try to produce automatically the summary table for T0003
 The table will include a new column to add a link to metat-data attached to each stack. Contrary to T0001 and T0002 metadata will no longer be sent to CADC (see comment below)

Data transfer to CADC
 Apparently all the QF data (step T0002) has been transferred to CADC
 From T0002 onwards all the CFHTLS metadata will be stored at TERAPIX, and will probably be accessed through Laurent's web interface

Web-service (Jean-Christophe)
 Still some testing required; there is a bug which has to be sorted out.

Skywatcher (Jean-Christophe))

 More improvements made to the interface and the zooming facility
 RA,DEC of the mouse position now visible on the sky map
 Static linking implemented, still problems with PLPLOT

PI data (Mirelle)

 Tully data not processed completely, still very difficult to have a good astrometric solution for the most crowed fields
 Loic Albert data processed but there is a problem with the background subtration
 Seymour data processed

WIRCAM (Yannick summarizes latest news from Hawaii
 Chiara is currently processing 'science grade' wircam data
 Some debate ongoing with CFHT concerning the format the data will be shipped and deliovred to TERAPIX and/or users. Emmanuel and Chiara working on it.

Visualisation (Anis)
 Anis is investigating the use of the OpenGL to display FITS images
 No idea how much time it will take to write a test program to display an image, but it would be probably a good idea to write this kind of program. Something simple that uses CFITSIO and OpenGL to display Megacam images would be a good way to learn these tools and to practice.

Workshop in Bordeaux (Anthony)

 Mecagrid (nice); distribution of data processing across the machines
 works on hetergenous networks
 Available on licence openGL
 Also a distributed image display view 'paraview' which works on different machines
 Micromegas, mutiscale image displaying, could be useful for displaying many images at the same time. Could perhaps be useful
 Some work display on the segmentation of images.

Summary of PaRISTIC days Presentation of interesting projects for TERAPIX  In french  Hyperlinks (PowerPoint, 87.5 kb)

Defectix (Anthony)
 Chiara has provided extensive bug reports concerning Defectix which has motived Anthony to start working on the code again

Hardware (Fred)

 Fred continues to make tests on mix5 and seems have isolated the memory modules which are causing problems.
 mix7 seems to be stable; no crashes reported as yet.

comparison with SDSS (Mathias, from Besancon)

 Mathias has compared SDSS and CFHTLS photometry and computed empirically the transformation between the CFHTLS magnitude system and SDSS.
 We would like to know if these transformations agree with what you'd expect based on the SDSS and Megacam filter response curves
 Mathias will make this merged catalogue available to us probably, on the IAP ftp server.