...
Unless converting the tools to use absolute times internally is easier than I think it is, I'd still propose that we have the tools and simulator check the value of MJDREF and issue a warning if it is not 51910.
| Panel |
|---|
|
Our tools have or will have multimission capabilities, and other FTOOLS will be able to process our
data files. Our tools and data files live in an environment larger than the SAE, and should follow
the standards of that environment. Therefore we should NEVER hardwire our MET standard into our tools,
nor should any of our files EVER assume our MET standard (i.e., MDJREF = 51910 should be in all our
files). That being said, we should ALWAYS use our MET standard in all of our data files, even if it is not
real data. This is the purpose of a standard. The result is that times will always be enormous,
but this is the reason that times are double precision. There are ways of dealing with smaller numbers inside of a code and yet writing data files that
follow the correct standard. For example, one can use TZEROn for a given FITS column, and the
FITSIO package that reads the column will automatically add the TZEROn value to all values in that
column. Thus a simulation package could use time relative to the beginning of the simulation, with
TZEROn equal to the beginning of the simulation (relative to MJDREF). Note that any code reading
the FITS file with a FITSIO package will see a large time value relative to MJDREF. David Band, 6 Oct 2005 |
2. Specification of a simulation starting time
...