$Id: Readme.txt 1813 2014-01-20 14:09:14Z lendl $

This directory contains various miniSEED examples useful for (unit-) testing
the two "IEEE floating point" codecs.


 Note: All ASCII dumps are in the same format as the GIPPtools 'edl2ascii'
       utility will produce.


Test 001 -- miniSEED file created by 'libmseed'
-----------------------------------------------

  A series of 6000 samples in IEEE single and IEEE double floating point 
  number encoding. The miniSEED file was created with the 'msrepack' tool 
  by Chad Trabant (IRIS DMC) from a Steim-1 encoded input file (see test 
  case "002-qlib2-repacked.mseed" in the Steim-1 test case subdirectory).

  Since input for the 'msrepack' tool were (Steim-1 encoded) integer values, 
  non of the samples in the miniSEED files given below have any fractional 
  part! This information might aid debugging...
  
  Possible pitfall:
  
  - Both floating point codecs, unlike the Steim family of codecs, do not 
    depend on aligning data to 64 byte long "frames". A consequence is, that
    the "begin-of-data" is already at offset 56 bytes and not at the usual
    (for Steim encoding) 64 byte offset. This is possible as there is only a 
    single blockette #1000 in the record.
     
  001-libmseed-single.mseed    miniSEED file (IEEE single precision)
  001-libmseed-single.header   ASCII dump of the header
  001-libmseed-single.data.gz  ASCII dump of the sample values (gzipped)
  
  001-libmseed-double.mseed    miniSEED file (IEEE double precision)
  001-libmseed-double.header   ASCII dump of the header 
  001-libmseed-double.data.gz  ASCII dump of the sample values (gzipped)

  Note:  To create the two miniSEED file the following commands were used:

          msrepack -v -t  -E 4  -o 001-libmseed-single.mseed  input.mseed
          msrepack -v -t  -E 5  -o 001-libmseed-double.mseed  input.mseed

  Note:  An attempt was made to also encode the same time series using Doug 
         Neuhauser's "QLib2 Library", however this failed! After a few hundred 
         samples of correctly converted values the resulting miniSEED file
         contains suddenly only garbage. The command used during the failed 
         attempt was:
    
           qmerge -T -v -O IEEE_FP_SP -r -o 001-qlib2-single.mseed  input.mseed


Test 002 -- Double precision floating point encoded miniSEED
------------------------------------------------------------

  A series of 5 miniSEED records (2525 samples) encoded in IEEE double 
  precision floating point format. There are no pitfalls, no problematic
  values and no strings attached.
  
  The five miniSEED records are from a real world data file and unlike the 
  test case #001 above, contain values that have fractional parts. Sampling 
  rate is 1 Hz.
  
  002-plain-double.mseed      miniSEED file (IEEE double precision)
  002-plain-double.header     ASCII dump of the header 
  002-plain-double.data.gz    ASCII dump of the sample values (gzipped)

