Teraminutes for the meeting of June 16th 2006

Wircam, Chiara
 Chiara has made a new stack with the new sky-subtracted images; photometry is much better than before
 There are still several cosmetic problems in the images which are not completely resolved
 Chiara will investigate if the noise in the final stacked image is what we expect
 Some useful comments have been exchanged with DaveThompson about the pre-processing steps
 We should look in more detail at the stacks; Henry will extract colour-colour plots from the D2 field matched with the CFHTLS to perform higher-level assessments
 The best solution is that Loic fixes his recipes with help from Dave Thompson [postscript: Dave Thompson will try to visit Loic at CFHT].
 Chiara will double check the Wircam press release before she leaves on vacation

 New megacam retina, Chiara
 Still doesn't quite work well with images with very good seeing; in this case, stars are marked as bad pixels.
 Training a retina takes around 8 days on mix1 with 60 images
 In general the training is a very time-consuming process and not sure if this is completely urgent
 It would be best to have a clear plan to explore the problem and evaluate the best parameters for Megacam.
 One way to evaluate the retinas is to use pix3 with a set of test data running QfFITs with the 2 retinas available so far

Weightwatcher, Chiara
 Weightwatcher has been modified to calculate effective area from polygon .reg region files
 .reg files can now contain overlapping polygons so weight-watcher can accept hand-generated polygons
 There is some bug related to how the header is written in weightwatcher; Emmanuel will try to look at this but it will take some time
 In any case it works if we do not apply the .reg file on MEFs
 Fixing this step could be important for the final stage in the pipeline, in particular to use the .reg files in QFITS. We will then be able to use the .reg mask file produced by Yannick for the last evaluation steps (galaxy and stellar counts, completeness limit)

Panorapix-2, Anis
 Fred was not able compile Anis' software either on the mac or on Fred's machine
 A demo of Anis' software was made on a overhead projector; zooming, contrast and gamma changing have all been implemented as programmable shaders
 A current problem is brightness and contrast cannot be altered at the same time

Machines, Fred
 All new machines have arrived; efgix, fcix2, mix8, mix9
 All Terapix+cosmix machines will be stopped on Tuesday so the new machines can be installed and the two clusters will be re-organised
 There will be a new rack installed at this time.
 The UPS should be also intalled and should run properly
 Once mix8 and mix9 installed, Fred will run the Swarp-killer on both machines to test them

Other matters
 e(b-v) values will be computed for each catalogue by Laurent using schlegel extinction values
 Discussion about which parameters / information to supplied to PIs
 Mireille asked whether Laurent ran the Annie Robin' data since she modified the grades. Need to check with Laurent [postscript: Laurent is now running Annie' data again]
 Laurent asks if scamp/swarp write in /tmp; not clear why . Fred checked and found no convincing evidence some tools write on /tmp, even for PLPLOT
 Question for Laurent: why does his programs produce tmp files which are so large?
 The 250GB disk for Mireille was ordered during the meeting [postscript: arrived on Monday 19 and installed by Fred].
 The Chiara's DELL laptop is ordered [need to check if this includes a QUERTY keyboard has been requested]
 Release catalogues W3 not supplied to CADC; Fred will supply the catalogue to CADC by Monday evening
 There seems to be some problem with the D4 at CADC; Henry will investigate

Next tera-meeting: