--- MITgcm/verification/exp4/README	2005/10/11 13:00:56	1.5
+++ MITgcm/verification/exp4/README	2014/03/10 01:35:16	1.8
@@ -3,24 +3,24 @@
 
 default paths are (as used in testreport):
   set build_dir = ${cwd}/build
-  set exe       = ${cwd}/build
+  set run_dir   = ${cwd}/run
   set ROOT      = ${cwd}/../../
 
 Configure and compile the code:
-  mkdir ${build_dir}
   cd ${build_dir}
-  ${ROOT}/tools/genmake2 -mods=${cwd}/code
+  ${ROOT}/tools/genmake2 -mods ../code [-of my_platform_optionFile]
   make depend
   make
+  cd ..
 
 To run:
-  mkdir ${exe_dir}
-  cd ${exe_dir}
-  cp ${cwd}/input/* .
-  ${build_dir}/mitgcmuv
+  cd ${run_dir}
+  ln -s ../input/* .
+  ${build_dir}/mitgcmuv > output.txt
+  cd ..
 
 There is comparison output in the directory:
- ..${cwd}/results
+ results      ( absolute path: ${run_dir}/../results )
 
 Comments:
 The input data is real*8.
@@ -36,5 +36,6 @@
 used for salinity and one passive tracer in the same way. Salinity is
 set to sLev at all other boundaries, while a (nearly) homogeneous
 v.Neumann condition is applied to the passive tracer (the latter is
-the default in obcs_calc.F). Open boundary parameters are set in
+the default in obcs_calc.F), with a relaxation (using pkg rbcs) in
+the Eastern part of the channel. Open boundary parameters are set in
 data.obcs.