Re: [netcdfgroup] NetCDF4 for Fusion Data

• To: netcdfgroup@xxxxxxxxxxxxxxxx
• Subject: Re: [netcdfgroup] NetCDF4 for Fusion Data
• From: John Storrs <john.storrs@xxxxxxxxxxxx>
• Date: Thu, 22 Jan 2009 19:23:37 +0000

Hi again Ed

You must be getting sick of me! I'd like to make another point about fill
values. We store digitizer data as shorts, with a scale and offset,
potentially covering the full bit range from 0h to FFFFh. We need to know if
the digitizer data is stuck at the top or bottom of the range so we can
detect hardware faults. I've confirmed that we can write the fill value (eg
FFFFh) as a valid data value, and read it back. But because it's a valid
application value, we really don't want to see it as a generated fill value.
The data and nothing but the data!

A related point for John: ncdump handles unsigned short values correctly (eg
FFFFh is 65535), but the Java tool treats them as signed shorts (eg FFFFh  is
shown as -1).

Regards
John


--
John Storrs, Experiments Dept      e-mail: john.storrs@xxxxxxxxxxxx
Building D3, UKAEA Fusion                               tel: 01235 466338
Culham Science Centre                                    fax: 01235 466379
Abingdon, Oxfordshire OX14 3DB              http://www.fusion.org.uk

• References:
  • [netcdfgroup] NetCDF4 for Fusion Data
    • From: John Storrs