NAME

     mrgcube  -  merges  and  optionally  transposes  a  cube  of
     seismic data distributed over multiple files having a common
     file name stem.

SYNOPSIS

     mrgcube [ -Nfile_stem ] [ -Ootap ] [ -Mamaxmem  ]  [  -tran-
     spose ] [ -S ] [ -V ] [ -? ]

DESCRIPTION

     mrgcube  merges  N   similarly  named  data  sets  into  one
     (N<100).   All  input data sets must have the same number of
     samples and records per file.   The data cannot be piped  in
     except  for  the degenerate case of a single input file that
     fits in memory. The data can always be piped out. This  pro-
     gram    circumvents    the    need    to    run   the   flow
     'mergn;presort;sisort' to obtain output  common  offset  MBS
     gathers  from  input  common  reflection point (crp) gathers
     generated from routine sammig. The  degenerate  single  file
     capability allows for more efficient sorting of 3D poststack
     data on memory limited workstation environment than  memangs
     and on the i/o inefficient Cray environment for resorter.

     mrgcube gets both its data and its parameters  from  command
     line  arguments.  These arguments specify the inputs, output
     and verbose printout, if desired.

Command line arguments

      parameters describing explicitly defined input data

     -N file_stem
          For multiple input files, enter the file  stem  of  the
          systematically  generated input data set file names (as
          generated by routines such as sammig) immediately after
          typing  -N. Routine mrgcube will augment this file stem
          by   adding   nfiles   file   extents   of   the   form
          __1__1,__2__1,...,99_99.             Thus            if
          file_stem=/data/crp.temp, totalnodes=2 and npass=3  for
          both  nodes,  this program will look for the six files:
          /data/crp.temp__1__1,             /data/crp.temp__2__1,
          /data/crp.temp__3__1,            2/data/crp.temp__1__2,
          /data/crp.temp__2__2,   /data/crp.temp__3__2.     Total
          number  of  nodes  used  in the original computation is
          stored in line header word /fI'MxUHTm'/fR.  The  actual
          node number used to calculate a given file is stored in
          /fI'MnUHTm'/fR. The total number of passes on each node
          nfiles  is  stored  in  /fI'MxLnIn'/fR,  while the pass
          number used to calculate a given file is stored in line
          header  word  /fI'MnLnIn'/fR.  The pass number and node
          numbe must mach up with the 2 digit suffix tacked  onto
          the end of file_stem.

     For the degenerate single file case, file_stem  is the input
     file  name. Lineheader values described above do not need to
     be filled out.

     -O otap
          Enter the output data  set  name  or  file  immediately
          after typing -O.  This output file is not required when
          piping the output to another process.  The output  data
          set also requires the full path name (see above).

     -transpose
          If present, transpose the gathers after  merging  (such
          that  common  reflection  point  gathers  become common
          offset gathers, or an inline sorted  3D  cube  of  data
          becomes a crossline sorted 3D cube of data).

     -S   If present, input data consists of a single  file  with
          explicit  name given by file_stem (with no name augmen-
          tation).

     -M amaxmem
          Enter the maximum memory allowed  for  the  program  in
          Megawords. This option may be necessary when sorting to
          large common offset output data sets on the workstation
          or  Cray-XMP  environment.  The  data  may will be read
          through as many times as necessary to  fall  below  the
          memory  constraints.  (Default: amaxmem=64 Megawords on
          the Crays, amaxmem=8 Megawords on the Sparcs).

     -V   Enter the command line argument '-V' to get  additional
          printout.

     -?   Enter the command line  argument  '-?'  to  get  online
          help.   The program terminates after the help screen is
          printed.

NOTE 1:

     The non-expert user is highly encouraged to use the job sub-
     mission  xwindows  program  samx  to ensure efficient, rela-
     tively error free construction of a complete  Kirchhoff  mbs
     flow using the sammig algorithm.

NOTE 2:

     The flow  rwindow|radonf; mrgcube|rwindow  -Y  |  radonf  -Y
     will  transform  (t,x,y) data to the (tau,q,p) domain, amen-
     able for 3D post stack filtering.  The  flow   radonr  -Y  |
     rwindow -Y -R; mrgcube | radonr | rwindow -R  will transform
     the data back to (t,x,y).

BUGS:

     No bugs known at present.

SEE ALSO

     sammig, mergn, presort, sisort, memangs, resorter, samx

AUTHOR

     K. J. Marfurt (APR 1992). Originally written to allow recom-
     bination  of  multiple output data subsets running on a dis-
     tributed network. Degenerate single input file option  added
     for  efficient  transposing of 3D post stack running windows
     (1994).

COPYRIGHT

     copyright 2001, Amoco Production Company
               All Rights Reserved
          an affiliate of BP America Inc.