Special NAUG - 2007-Jan-03
Progress/Status EVLA TST
- General
- calibrator models - high frequency images are available:
Calibrator |
Q |
K |
U |
C |
X |
L |
3C48 0137+331 (J2000) 0134+329 (B1950) |
X |
X |
X |
|
|
|
3C138 0521+166 (J2000) 0518+165 (B1950) |
X |
X |
X |
|
|
|
3C147 0542+498 (J2000) 0538+498 (B1950) |
X |
X |
X |
|
|
|
3C286 1331+305 (J2000) 1328+307 (B1950) |
X |
X |
X |
|
|
|
- data selection - Will keep the index specifiers (fieldid, etc) for now.
fieldid:---> field: takes a single string (spaces separate the names of the fields)
it does a minimum match on each string so:
field='133'
field='1331+305'
field='1331+30500002_0' (Field Name reported in listobs)
are all equivalent
Multiple fields are specified as:
field='133 144'
will get you fields: 1331+30500002_0, 1445_09900002_0
Implemented in: clean, contsub, exportuvfits, flagdata, flagxy, ft, invert,
mosaic, plotxy, setjy, split
-
-
- added field,spwid (to eventually replace msselect) in:
- bandpass, blcal, correct, fringecal, gaincal, pointcal, uvmodelfit
- Refant now refers to antenna name; Note: still needs calplot to use antenna names not indices
- bandpass, blcal, fringecal, gaincal, pointcal
-
error trapping within tasks 20070105
- makemask task - support for clean boxes.Also next week
- added cleanbox parameter to clean, mosaic; compliant with clean boxes in AIPS BOXFILE parameter - Note: can not currently handle the different 'cards' (C,F,W, etc).
fieldname blc-x blc-y trc-x trc-y #as in AIPS BOXFILE
0 60 60 75 75 # or
0 18:30:24.0 -05.15.00 18:30:26.0 -05:15:15
Note: earlier list of Alpha issues was well aligned with other test comments so the test reinforced/resolved some needed areas. We'll update that list in addition to the test spread sheet
Question: When is 64-bit computing coming? March 2007, maybe earlier for RH.
CASA 2007 Calendar
Lots of commitments for only the third day of the year. Michael has sent a list of EVLA related targets that need to be integrated into this. Danger zone is the simulator. Must coordinate with Crystal/CV for the pre-test. Early February looks okay (not before the 3rd).
Testing Proposal
- How to keep the momentum - User Support Fridays
- NAUG members will test/CASA supports
Seems reasonable. Not exclusive to Friday but guaranteed for Friday. It is also a good day for doing verifications for resolved issues and for focused testing in some areas (e.g., pointing calibration).
- Question: Does this cover your flagging interactions; if not, please state it.
- Question: What algorithms should be ported from AIPS first?
- Interface
- data selection by:
- antennanames (antennaid), spwid, fieldnames (fieldid), uvdist, time, correlation, (nchan, start, width)
- any other parameters needed?
- manual flagging:
- antenna, autocorrelation, baseline, fieldnames, spwid, channels, freqs, timerange, timeslot, clip, quack, unflag
- other parameters?
- algorithms
- timemedian, freqmedian, spectralrejection, uvbin
- what others first?
- summary/statistics
- get percentage of data flagged vs antenna, baseline, spwid, channel, correlations, time
- other statistics?
- queries
- return list of antennas, channels, baselines, times with more or less than a specified percentage of flagged data.
- extend flags
- use queries to extend flags (e.g, flag all channels that have more than 70% of the data flagged)
- extend flag scope
- extend by time, channel, baseline, polarization, spwid, spectral/time averages
- other functions
- flag history - append to history file
- flag reset - reset flags
Lots of expected feedback here.
--
JosephMcMullin - 03 Jan 2007