diff --git a/src/Action_Rmsd.cpp b/src/Action_Rmsd.cpp
index 78ddc2eafb..2341065412 100644
--- a/src/Action_Rmsd.cpp
+++ b/src/Action_Rmsd.cpp
@@ -320,6 +320,14 @@ Action::RetType Action_Rmsd::Setup(ActionSetup& setup) {
     mprintf("Warning: No atoms in mask '%s'.\n", tgtMask_.MaskString());
     return Action::SKIP;
   }
+  if ( tgtMask_.Nselected() < 3 ) {
+    mprintf("Warning: Less than 3 atoms selected for RMSD. Cannot fully"
+            "Warning:   populate the coordinate covariance matrix.\n");
+    if (debug_ == 0) {
+      mprintf("Warning: Skipping.\n");
+      return Action::SKIP;
+    }
+  }
   // Allocate space for selected atoms in the frame. This will also put the
   // correct masses in based on the mask.
   tgtFrame_.SetupFrameFromMask(tgtMask_, setup.Top().Atoms());
diff --git a/src/Cpptraj.cpp b/src/Cpptraj.cpp
index 387118094b..7798aab60a 100644
--- a/src/Cpptraj.cpp
+++ b/src/Cpptraj.cpp
@@ -17,6 +17,9 @@
 #ifdef CUDA
 # include <cuda_runtime_api.h>
 #endif
+#ifdef _OPENMP
+# include <omp.h>
+#endif
 
 /// CONSTRUCTOR - initializes all commands
 Cpptraj::Cpptraj() {
@@ -37,17 +40,20 @@ Cpptraj::~Cpptraj() { Command::Free(); }
 void Cpptraj::Usage() {
   mprinterr("\n"
             "Usage: cpptraj [-p <Top0>] [-i <Input0>] [-y <trajin>] [-x <trajout>]\n"
+            "               [-ya <args>] [-xa <args>] [<file>]\n"
             "               [-c <reference>] [-d <datain>] [-w <dataout>] [-o <output>]\n"
             "               [-h | --help] [-V | --version] [--defines] [-debug <#>]\n"
             "               [--interactive] [--log <logfile>] [-tl]\n"
             "               [-ms <mask>] [-mr <mask>] [--mask <mask>] [--resmask <mask>]\n"
-            "       cpptraj <Top> <Input>\n"
-            "\t-p <Top0>        : Load <Top0> as a topology file. May be specified more than once.\n"
-            "\t-i <Input0>      : Read input from <Input0>. May be specified more than once.\n"
-            "\t-y <trajin>      : Read from trajectory file <trajin>; same as input 'trajin <trajin>'.\n"
-            "\t-x <trajout>     : Write trajectory file <trajout>; same as input 'trajout <trajout>'.\n"
-            "\t-c <reference>   : Read <reference> as reference coordinates; same as input 'reference <reference>'.\n"
-            "\t-d <datain>      : Read data in from file <datain> ('readdata <datain>').\n"
+            "\t-p <Top0>        : * Load <Top0> as a topology file.\n"
+            "\t-i <Input0>      : * Read input from file <Input0>.\n"
+            "\t-y <trajin>      : * Read from trajectory file <trajin>; same as input 'trajin <trajin>'.\n"
+            "\t-x <trajout>     : * Write trajectory file <trajout>; same as input 'trajout <trajout>'.\n"
+            "\t-ya <args>       : * Input trajectory file arguments.\n"
+            "\t-xa <args>       : * Output trajectory file arguments.\n"
+            "\t<file>           : * A topology, input trajectory, or file containing cpptraj input.\n"
+            "\t-c <reference>   : * Read <reference> as reference coordinates; same as input 'reference <reference>'.\n"
+            "\t-d <datain>      : * Read data in from file <datain> ('readdata <datain>').\n"
             "\t-w <dataout>     : Write data from <datain> as file <dataout> ('writedata <dataout>).\n"
             "\t-o <output>      : Write CPPTRAJ STDOUT output to file <output>.\n"
             "\t-h | --help      : Print command line help and exit.\n"
@@ -56,11 +62,13 @@ void Cpptraj::Usage() {
             "\t-debug <#>       : Set global debug level to <#>; same as input 'debug <#>'.\n"
             "\t--interactive    : Force interactive mode.\n"
             "\t--log <logfile>  : Record commands to <logfile> (interactive mode only). Default is 'cpptraj.log'.\n"
-            "\t-tl              : Print length of trajectories specified with '-y' to STDOUT.\n"
+            "\t-tl              : Print length of all input trajectories specified on the command line to STDOUT.\n"
             "\t-ms <mask>       : Print selected atom numbers to STDOUT.\n"
             "\t-mr <mask>       : Print selected residue numbers to STDOUT.\n"
             "\t--mask <mask>    : Print detailed atom selection to STDOUT.\n"
-            "\t--resmask <mask> : Print detailed residue selection to STDOUT.\n\n");
+            "\t--resmask <mask> : Print detailed residue selection to STDOUT.\n"
+            "      * Denotes flag may be specified multiple times.\n"
+            "\n");
 }
 
 void Cpptraj::Intro() {
@@ -77,7 +85,10 @@ void Cpptraj::Intro() {
           "\n    ___  ___  ___  ___\n     | \\/ | \\/ | \\/ | \n    _|_/\\_|_/\\_|_/\\_|_\n\n",
           CPPTRAJ_VERSION_STRING);
 # ifdef MPI
-  mprintf("| Running on %i threads\n", Parallel::World().Size());
+  mprintf("| Running on %i processes.\n", Parallel::World().Size());
+# endif
+# ifdef _OPENMP
+  mprintf("| %i OpenMP threads available.\n", omp_get_max_threads());
 # endif
   mprintf("| Date/time: %s\n", TimeString().c_str());
   std::string available_mem = AvailableMemoryStr();
@@ -251,29 +262,93 @@ int Cpptraj::ProcessMask( Sarray const& topFiles, Sarray const& refFiles,
   return 0;
 }
 
-void Cpptraj::AddFiles(Sarray& Files, int argc, char** argv, int& idx) {
-  Files.push_back( argv[++idx] );
-  // Assume all following args without leading '-' are also files.
-  while (idx+1 != argc && argv[idx+1][0] != '-')
-    Files.push_back( argv[++idx] );
+/** Add current argument and all following arguments without leading '-' to
+  * given string array.
+  */
+void Cpptraj::AddArgs(Sarray& Args, ArgList const& cmdLineArgs, int& idx)
+{
+  Args.push_back( cmdLineArgs[++idx] );
+  while (idx+1 != cmdLineArgs.Nargs() && cmdLineArgs[idx+1][0] != '-')
+    Args.push_back( cmdLineArgs[++idx] );
+}
+
+/** \return True if argument at position matches key and is not the final argument.
+  */
+static inline bool NotFinalArg(ArgList const& cmdLineArgs, const char* key, int pos)
+{
+  return (cmdLineArgs[pos] == key && pos+1 != cmdLineArgs.Nargs());
+}
+
+/** \return 0 if unknown, 1 if topology, 2 if trajectory, 3 if input. */
+static inline int AutoDetect(std::string const& arg)
+{
+  if (File::Exists(arg)) {
+    // Check if this is a topology file.
+    ParmFile::ParmFormatType ptype = ParmFile::DetectFormat( arg );
+    if (ptype != ParmFile::UNKNOWN_PARM)
+      return 1;
+    // Check if this is a trajectory file.
+    TrajectoryFile::TrajFormatType ttype = TrajectoryFile::DetectFormat( arg );
+    if (ttype != TrajectoryFile::UNKNOWN_TRAJ)
+      return 2;
+    // Assume input file.
+    return 3;
+  }
+  return 0;
+}
+
+/** Check that number of args are the same as files. If fewer args,
+  * replicate the last arg until sizes are equal. If more args, remove the
+  * extra args. If no files, print a warning.
+  */
+void Cpptraj::ResizeArgs(Sarray const& Files, Sarray& Args, const char* type)
+{
+  if (Files.empty())
+    mprintf("Warning: No %s trajectories but %s trajectory arguments were specified.\n",
+            type, type);
+  else {
+    if (Args.size() < Files.size()) {
+      mprintf("Warning: Fewer %s trajectory arguments than %s trajectories.\n"
+              "Warning: Using final specified argument for remaining trajectories: '%s'\n",
+              type, type, Args.back().c_str());
+      Args.resize( Files.size(), Args.back() );
+    } else if (Args.size() > Files.size()) {
+      mprintf("Warning: More %s trajectory arguments specified than %s trajectories.\n",
+              type, type);
+      Args.resize( Files.size() );
+    }
+  }
+  mprintf("\tArguments for %zu %s trajectories:\n", Files.size(), type);
+  Sarray::const_iterator tf = Files.begin();
+  for (Sarray::const_iterator ta = Args.begin(); ta != Args.end(); ++ta, ++tf)
+    mprintf("\t  %s %s\n", tf->c_str(), ta->c_str());
 }
 
 /** Read command line args. */
 Cpptraj::Mode Cpptraj::ProcessCmdLineArgs(int argc, char** argv) {
+  // First convert argv to one continuous string.
   commandLine_.clear();
   for (int i = 1; i < argc; i++)
     commandLine_.append( " " + std::string(argv[i]) );
+  // Use ArgList to split into arguments.
+  ArgList cmdLineArgs( commandLine_ );
+  // mprintf("DEBUG: CmdLine: %s\n", cmdLineArgs.ArgLine() );
+  // Process command line flags from ArgList
   bool hasInput = false;
   bool interactive = false;
   Sarray inputFiles;
   Sarray topFiles;
   Sarray trajinFiles;
+  Sarray trajinArgs;
   Sarray trajoutFiles;
+  Sarray trajoutArgs;
   Sarray refFiles;
   Sarray dataFiles;
   std::string dataOut;
-  for (int i = 1; i < argc; i++) {
-    std::string arg(argv[i]);
+  for (int iarg = 0; iarg < cmdLineArgs.Nargs(); iarg++)
+  {
+    std::string const& arg = cmdLineArgs[iarg];
+    // ----- One-and-done flags ------------------
     if ( arg == "--help" || arg == "-h" ) {
       // --help, -help: Print usage and exit
       SetWorldSilent(true);
@@ -309,76 +384,96 @@ Cpptraj::Mode Cpptraj::ProcessCmdLineArgs(int argc, char** argv) {
       if (State_.TrajLength( topFiles[0], trajinFiles )) return ERROR;
       return QUIT;
     }
-    if ( arg == "--interactive" )
+    // ----- Single flags ------------------------
+    if ( arg == "--interactive" ) {
       interactive = true;
-    else if ( arg == "--suppress-all-output") {
+    } else if ( arg == "--suppress-all-output") {
       mprintf("Info: All further output will be suppressed.\n");
       SuppressAllOutput();
-    } else if ( arg == "-debug" && i+1 != argc) {
+    // ----- Flags that precede values -----------
+    } else if ( NotFinalArg(cmdLineArgs, "-debug", iarg) ) {
       // -debug: Set overall debug level
-      ArgList dbgarg( argv[++i] );
+      ArgList dbgarg( cmdLineArgs[++iarg] );
       State_.SetListDebug( dbgarg );
-    } else if ( arg == "--log" && i+1 != argc)
+    } else if ( NotFinalArg(cmdLineArgs, "--log", iarg) ) {
       // --log: Set up log file for interactive mode
-      logfilename_ = argv[++i];
-    else if ( arg == "-p" && i+1 != argc) {
+      logfilename_ = cmdLineArgs[++iarg];
+    } else if ( NotFinalArg(cmdLineArgs, "-p", iarg) ) {
       // -p: Topology file
-      AddFiles( topFiles, argc, argv, i );
-    } else if ( arg == "-d" && i+1 != argc) {
+      AddArgs( topFiles, cmdLineArgs, iarg );
+    } else if ( NotFinalArg(cmdLineArgs, "-d", iarg) ) {
       // -d: Read data file
-      AddFiles( dataFiles, argc, argv, i );
-    } else if ( arg == "-w" && i+1 != argc) {
+      AddArgs( dataFiles, cmdLineArgs, iarg );
+    } else if ( NotFinalArg(cmdLineArgs, "-w", iarg) ) {
       // -w: Write data file. Only one allowed. For data file conversion.
-      dataOut.assign( argv[++i] );
-    } else if ( arg == "-y" && i+1 != argc) {
+      dataOut.assign( cmdLineArgs[++iarg] );
+    } else if ( NotFinalArg(cmdLineArgs, "-y", iarg) ) {
       // -y: Trajectory file in.
-      AddFiles( trajinFiles, argc, argv, i );
-    } else if ( arg == "-x" && i+1 != argc) {
+      AddArgs( trajinFiles, cmdLineArgs, iarg );
+    } else if ( NotFinalArg(cmdLineArgs, "-ya", iarg) ) {
+      // -ya: Trajectory file in arguments.
+      AddArgs( trajinArgs, cmdLineArgs, iarg );
+    } else if ( NotFinalArg(cmdLineArgs, "-x", iarg) ) {
       // -x: Trajectory file out
-      trajoutFiles.push_back( argv[++i] );
-    } else if ( arg == "-c" && i+1 != argc) {
+      trajoutFiles.push_back( cmdLineArgs[++iarg] );
+    } else if ( NotFinalArg(cmdLineArgs, "-xa", iarg) ) {
+      // -xa: Trajectory file out arguments.
+      AddArgs( trajoutArgs, cmdLineArgs, iarg );
+    } else if ( NotFinalArg(cmdLineArgs, "-c", iarg) ) {
       // -c: Reference file
-      AddFiles( refFiles, argc, argv, i );
-    } else if (arg == "-i" && i+1 != argc) {
+      AddArgs( refFiles, cmdLineArgs, iarg );
+    } else if ( NotFinalArg(cmdLineArgs, "-i", iarg) ) {
       // -i: Input file(s)
-      AddFiles( inputFiles, argc, argv, i );
-    } else if (arg == "-o" && i+1 != argc) {
+      AddArgs( inputFiles, cmdLineArgs, iarg );
+    } else if ( NotFinalArg(cmdLineArgs, "-o", iarg) ) {
       // -o: Output file
-      FileName ofilename(argv[++i]);
+      FileName ofilename(cmdLineArgs[++iarg]);
       if (ofilename.empty()) {
         mprinterr("Error: Could not set up output file with name '%s'\n", ofilename.full());
         return ERROR;
       }
       if (OutputToFile(ofilename.full())) return ERROR;
-    } else if (arg == "-ms" && i+1 != argc) {
+    } else if ( NotFinalArg(cmdLineArgs, "-ms", iarg) ) {
       // -ms: Parse mask string, print selected atom #s
-      if (ProcessMask( topFiles, refFiles, std::string(argv[++i]), false, false )) return ERROR;
+      if (ProcessMask( topFiles, refFiles, cmdLineArgs[++iarg], false, false )) return ERROR;
       return QUIT;
-    } else if (arg == "-mr" && i+1 != argc) {
+    } else if ( NotFinalArg(cmdLineArgs, "-mr", iarg) ) {
       // -mr: Parse mask string, print selected res #s
-      if (ProcessMask( topFiles, refFiles, std::string(argv[++i]), false, true )) return ERROR;
+      if (ProcessMask( topFiles, refFiles, cmdLineArgs[++iarg], false, true )) return ERROR;
       return QUIT;
-    } else if (arg == "--mask" && i+1 != argc) {
+    } else if ( NotFinalArg(cmdLineArgs, "--mask", iarg) ) {
       // --mask: Parse mask string, print selected atom details
-      if (ProcessMask( topFiles, refFiles, std::string(argv[++i]), true, false )) return ERROR;
+      if (ProcessMask( topFiles, refFiles, cmdLineArgs[++iarg], true, false )) return ERROR;
       return QUIT;
-    } else if (arg == "--resmask" && i+1 != argc) {
+    } else if ( NotFinalArg(cmdLineArgs, "--resmask", iarg) ) {
       // --resmask: Parse mask string, print selected residue details
-      if (ProcessMask( topFiles, refFiles, std::string(argv[++i]), true, true )) return ERROR;
+      if (ProcessMask( topFiles, refFiles, cmdLineArgs[++iarg], true, true )) return ERROR;
       return QUIT;
-    } else if ( i == 1 ) {
-      // For backwards compatibility with PTRAJ; Position 1 = TOP file
-      topFiles.push_back( argv[i] );
-    } else if ( i == 2 ) {
-      // For backwards compatibility with PTRAJ; Position 2 = INPUT file
-      inputFiles.push_back( argv[i] );
     } else {
-      // Unrecognized
-      mprintf("  Unrecognized input on command line: %i: %s\n", i,argv[i]);
-      Usage();
-      return ERROR;
+      // Check if this is a file.
+#     ifdef MPI
+      int ftype;
+      if (Parallel::World().Master())
+        ftype = AutoDetect( arg );
+      Parallel::World().MasterBcast(&ftype, 1, MPI_INT);
+#     else
+      int ftype = AutoDetect( arg );
+#     endif
+      if (ftype == 1)
+        topFiles.push_back( arg );
+      else if (ftype == 2)
+        trajinFiles.push_back( arg );
+      else if (ftype == 3) {
+        if (iarg != 1) mprintf("Warning: Assuming '%s' contains cpptraj input.\n", arg.c_str());
+        inputFiles.push_back( arg );
+      } else {
+        mprinterr("Error: Unrecognized input on command line: %i: %s\n",
+                  iarg+1, cmdLineArgs[iarg].c_str());
+        Usage();
+        return ERROR;
+      }
     }
-  }
+  } // END loop over command line flags
   Cpptraj::Intro();
   // Add all data files specified on command lin.
   for (Sarray::const_iterator dataFilename = dataFiles.begin();
@@ -419,17 +514,31 @@ Cpptraj::Mode Cpptraj::ProcessCmdLineArgs(int argc, char** argv) {
                               ++refName)
     if (State_.AddReference( *refName )) return ERROR;
   // Add all input trajectories specified on command line.
-  for (Sarray::const_iterator trajinName = trajinFiles.begin();
-                              trajinName != trajinFiles.end();
-                              ++trajinName)
-    if (State_.AddInputTrajectory( *trajinName )) return ERROR;
+  // If there are fewer input trajectory arguments than input trajectories,
+  // duplicate the last input trajectory argument.
+  if (!trajinArgs.empty()) {
+    ResizeArgs( trajinFiles, trajinArgs, "input" );
+    for (unsigned int it = 0; it != trajinFiles.size(); it++)
+      if (State_.AddInputTrajectory( trajinFiles[it] + " " + trajinArgs[it] )) return ERROR;
+  } else {
+    for (Sarray::const_iterator trajinName = trajinFiles.begin();
+                                trajinName != trajinFiles.end();
+                                ++trajinName)
+      if (State_.AddInputTrajectory( *trajinName )) return ERROR;
+  }
   // Add all output trajectories specified on command line.
   if (!trajoutFiles.empty()) {
     hasInput = true; // This allows direct traj conversion with no other input
-    for (Sarray::const_iterator trajoutName = trajoutFiles.begin();
-                                trajoutName != trajoutFiles.end();
-                                ++trajoutName)
-      if (State_.AddOutputTrajectory( *trajoutName )) return ERROR;
+    if (!trajoutArgs.empty()) {
+      ResizeArgs( trajoutFiles, trajoutArgs, "output" );
+      for (unsigned int it = 0; it != trajoutFiles.size(); it++)
+        if (State_.AddOutputTrajectory( trajoutFiles[it] + " " + trajoutArgs[it] )) return ERROR;
+    } else {
+      for (Sarray::const_iterator trajoutName = trajoutFiles.begin();
+                                  trajoutName != trajoutFiles.end();
+                                  ++trajoutName)
+        if (State_.AddOutputTrajectory( *trajoutName )) return ERROR;
+    }
   }
   // Process all input files specified on command line.
   if ( !inputFiles.empty() ) {
diff --git a/src/Cpptraj.h b/src/Cpptraj.h
index 9528bc1f33..00e27f3542 100644
--- a/src/Cpptraj.h
+++ b/src/Cpptraj.h
@@ -18,8 +18,9 @@ class Cpptraj {
     static void Usage();
     static void Intro();
     static void Finalize();
+    static inline void AddArgs(Sarray&, ArgList const&, int&);
+    static inline void ResizeArgs(Sarray const&, Sarray&, const char*);
     int ProcessMask(Sarray const&, Sarray const&, std::string const&, bool,bool) const;
-    static inline void AddFiles(Sarray&, int, char**, int&);
     Mode ProcessCmdLineArgs(int, char**);
     int Interactive();
 
diff --git a/src/CpptrajState.cpp b/src/CpptrajState.cpp
index b760c88e22..3c16fd3333 100644
--- a/src/CpptrajState.cpp
+++ b/src/CpptrajState.cpp
@@ -515,7 +515,7 @@ int CpptrajState::RunEnsemble() {
   // In parallel only two frames needed; one for reading, one for receiving.
   FramePtrArray SortedFrames( 2 );
   FrameArray FrameEnsemble( 2 );
-  // Each thread will process one member of the ensemble, so local ensemble
+  // Each process will handle one member of the ensemble, so local ensemble
   // size is effectively 1.
   ensembleSize = 1;
 # else
@@ -533,7 +533,7 @@ int CpptrajState::RunEnsemble() {
   ActionEnsemble[0] = &actionList_;
   for (int member = 1; member < ensembleSize; member++)
     ActionEnsemble[member] = new ActionList();
-  // If we are on a single thread, give each member its own copy of the
+  // If we are on a single process, give each member its own copy of the
   // current topology address. This way if topology is modified by a member,
   // e.g. in strip or closest, subsequent members wont be trying to modify 
   // an already-modified topology.
@@ -739,19 +739,19 @@ int CpptrajState::RunEnsemble() {
 }
 #ifdef MPI
 // -----------------------------------------------------------------------------
-void CpptrajState::DivideFramesAmongThreads(int& my_start, int& my_stop, int& my_frames,
+void CpptrajState::DivideFramesAmongProcesses(int& my_start, int& my_stop, int& my_frames,
                                             int maxFrames, Parallel::Comm const& commIn) const
 {
-  my_frames = commIn.DivideAmongThreads(my_start, my_stop, maxFrames);
-  std::vector<int> frames_per_thread( commIn.Size() );
-  commIn.GatherMaster(&my_frames, 1, MPI_INT, &frames_per_thread[0]);
+  my_frames = commIn.DivideAmongProcesses(my_start, my_stop, maxFrames);
+  std::vector<int> frames_per_process( commIn.Size() );
+  commIn.GatherMaster(&my_frames, 1, MPI_INT, &frames_per_process[0]);
   // Print how many frames each rank will process.
   if (commIn.Master()) {
     mprintf("\nPARALLEL INFO:\n");
     if (Parallel::EnsembleComm().Size() > 1)
-      mprintf("  %i threads per ensemble member.\n", commIn.Size());
+      mprintf("  %i processes per ensemble member.\n", commIn.Size());
     for (int rank = 0; rank != commIn.Size(); rank++)
-      mprintf("  Thread %i will process %i frames.\n", rank, frames_per_thread[rank]);
+      mprintf("  Process %i will handle %i frames.\n", rank, frames_per_process[rank]);
   }
   commIn.Barrier();
   if (debug_ > 0) rprintf("Start %i Stop %i Frames %i\n", my_start+1, my_stop, my_frames);
@@ -773,13 +773,13 @@ int CpptrajState::PreloadCheck(int my_start, int my_frames,
     return 1;
   } else if (my_frames == n_previous_frames) {
     rprinterr("Error: Number of preload frames is same as number of processed frames.\n"
-              "Error:   Try reducing the number of threads.\n");
+              "Error:   Try reducing the number of processes.\n");
     return 1;
   } 
   if (n_previous_frames > (my_frames / 2))
     rprintf("Warning: Number of preload frames is greater than half the "
                       "number of processed frames.\n"
-            "Warning:   Try reducing the number of threads.\n");
+            "Warning:   Try reducing the number of processes.\n");
   rprintf("Warning: Preloading %i frames. These frames will NOT have Actions performed on them.\n",
           n_previous_frames);
   return 0;
@@ -805,19 +805,19 @@ int CpptrajState::RunParaEnsemble() {
   int err = NAV.AddEnsembles(trajinList_.ensemble_begin(), trajinList_.ensemble_end());
   if (Parallel::World().CheckError( err )) return 1;
 
-  // Divide frames among threads
+  // Divide frames among processes
   int my_start, my_stop, my_frames;
-  DivideFramesAmongThreads(my_start, my_stop, my_frames, NAV.IDX().MaxFrames(), TrajComm);
-  // Ensure at least 1 frame per thread, otherwise some ranks could cause hangups.
+  DivideFramesAmongProcesses(my_start, my_stop, my_frames, NAV.IDX().MaxFrames(), TrajComm);
+  // Ensure at least 1 frame per process, otherwise some ranks could cause hangups.
   if (my_frames > 0)
     err = 0;
   else {
-    rprinterr("Error: Thread is processing less than 1 frame. Try reducing # threads.\n");
+    rprinterr("Error: Process is handling less than 1 frame. Try reducing # processes.\n");
     err = 1;
   }
   if (Parallel::World().CheckError( err )) return 1;
 
-  // Allocate DataSets in the master DataSetList based on # frames to be read by this thread.
+  // Allocate DataSets in the master DataSetList based on # frames to be read by this process.
   DSL_.AllocateSets( my_frames );
   // Any DataSets added to the DataSetList during run will need to be synced.
   DSL_.SetNewSetsNeedSync( true );
@@ -917,9 +917,9 @@ int CpptrajState::RunParaEnsemble() {
   ensembleOut_.CloseEnsembleOut();
   DSL_.SetNewSetsNeedSync( false );
   sync_time_.Start();
-  // Sync Actions to master thread
+  // Sync Actions to master process
   actionList_.SyncActions();
-  // Sync data sets to master thread
+  // Sync data sets to master process
   if (DSL_.SynchronizeData( TrajComm )) return 1;
   sync_time_.Stop();
   mprintf("\nACTION OUTPUT:\n");
@@ -968,19 +968,19 @@ int CpptrajState::RunParallel() {
     if (input_traj.AddInputTraj( *traj )) { err = 1; break; }
   if (TrajComm.CheckError( err )) return 1;
 
-  // Divide frames among threads.
+  // Divide frames among processes.
   int my_start, my_stop, my_frames;
-  DivideFramesAmongThreads(my_start, my_stop, my_frames, input_traj.Size(), TrajComm);
-  // Ensure at least 1 frame per thread, otherwise some ranks could cause hangups.
+  DivideFramesAmongProcesses(my_start, my_stop, my_frames, input_traj.Size(), TrajComm);
+  // Ensure at least 1 frame per process, otherwise some ranks could cause hangups.
   if (my_frames > 0)
     err = 0;
   else {
-    rprinterr("Error: Thread is processing less than 1 frame. Try reducing # threads.\n");
+    rprinterr("Error: Process is handling less than 1 frame. Try reducing # processes.\n");
     err = 1;
   }
   if (TrajComm.CheckError( err )) return 1;
 
-  // Allocate DataSets in DataSetList based on # frames read by this thread.
+  // Allocate DataSets in DataSetList based on # frames read by this process.
   DSL_.AllocateSets( my_frames );
   // Any DataSets added to the DataSetList during run will need to be synced.
   DSL_.SetNewSetsNeedSync( true );
@@ -1093,9 +1093,9 @@ int CpptrajState::RunParallel() {
   trajoutList_.CloseTrajout();
   DSL_.SetNewSetsNeedSync( false );
   sync_time_.Start();
-  // Sync Actions to master thread
+  // Sync Actions to master process
   actionList_.SyncActions();
-  // Sync data sets to master thread
+  // Sync data sets to master process
   if (DSL_.SynchronizeData( TrajComm )) return 1;
   sync_time_.Stop();
   post_time_.Start();
@@ -1120,10 +1120,10 @@ int CpptrajState::RunSingleTrajParallel() {
   // Set up single trajectory for parallel read.
   Trajin* trajin = *(trajinList_.trajin_begin());
   trajin->ParallelBeginTraj( Parallel::World() );
-  // Divide frames among threads.
+  // Divide frames among processes.
   int total_read_frames = trajin->Traj().Counter().TotalReadFrames();
   int my_start, my_stop, my_frames;
-  std::vector<int> rank_frames = DivideFramesAmongThreads(my_start, my_stop, my_frames,
+  std::vector<int> rank_frames = DivideFramesAmongProcesses(my_start, my_stop, my_frames,
                                                           total_read_frames,
                                                           Parallel::World().Size(),
                                                           Parallel::World().Rank(),
@@ -1135,7 +1135,7 @@ int CpptrajState::RunSingleTrajParallel() {
   rprintf("Start and stop adjusted for offset: %i to %i\n", traj_start, traj_stop);
   Parallel::World().Barrier();
 
-  // Allocate DataSets in DataSetList based on # frames read by this thread.
+  // Allocate DataSets in DataSetList based on # frames read by this process.
   DSL_.AllocateSets( my_frames );
 
   // ----- SETUP PHASE ---------------------------
@@ -1189,11 +1189,11 @@ int CpptrajState::RunSingleTrajParallel() {
   mprintf("TIME: Avg. throughput= %.4f frames / second.\n",
           (double)total_read_frames / frames_time.Total());
   trajoutList_.CloseTrajout();
-  // Sync data sets to master thread
+  // Sync data sets to master process
   Timer time_sync;
   time_sync.Start();
   if (DSL_.SynchronizeData( total_read_frames, rank_frames, Parallel::World() )) return 1;
-  // Sync Actions to master thread
+  // Sync Actions to master process
   actionList_.SyncActions();
   time_sync.Stop();
   time_sync.WriteTiming(1, "Data set/actions sync");
@@ -1368,7 +1368,7 @@ int CpptrajState::RunAnalyses() {
 # ifdef MPI
   // Only master performs analyses currently.
   if (Parallel::TrajComm().Size() > 1)
-    mprintf("Warning: Analysis does not currently use multiple MPI threads.\n");
+    mprintf("Warning: Analysis does not currently use multiple MPI processes.\n");
   if (Parallel::TrajComm().Master())
 # endif
     err = analysisList_.DoAnalyses();
diff --git a/src/CpptrajState.h b/src/CpptrajState.h
index a738286afe..7d274d2540 100644
--- a/src/CpptrajState.h
+++ b/src/CpptrajState.h
@@ -83,7 +83,7 @@ class CpptrajState {
     int RunNormal();
     int RunEnsemble();
 #   ifdef MPI
-    void DivideFramesAmongThreads(int&, int&, int&, int, Parallel::Comm const&) const;
+    void DivideFramesAmongProcesses(int&, int&, int&, int, Parallel::Comm const&) const;
     int PreloadCheck(int, int, int&, int&) const;
     int RunParallel();
     int RunParaEnsemble();
diff --git a/src/Exec_CrdAction.cpp b/src/Exec_CrdAction.cpp
index 442733df2c..916ee8e3f5 100644
--- a/src/Exec_CrdAction.cpp
+++ b/src/Exec_CrdAction.cpp
@@ -110,7 +110,7 @@ Exec::RetType Exec_CrdAction::Execute(CpptrajState& State, ArgList& argIn) {
   //rprintf("DEBUG: About to create new comm, ID= %i\n", ID);
   trajComm_ = Parallel::World().Split( ID );
   if (ID != MPI_UNDEFINED) {
-    mprintf("Warning: '%s' command does not yet use multiple MPI threads.\n", argIn.Command());
+    mprintf("Warning: '%s' command does not yet use multiple MPI processes.\n", argIn.Command());
     ret = ProcessArgs(State, argIn);
     if (ret != CpptrajState::OK)
       err = 1;
diff --git a/src/Exec_CrdOut.cpp b/src/Exec_CrdOut.cpp
index ad705c654c..965ec407c5 100644
--- a/src/Exec_CrdOut.cpp
+++ b/src/Exec_CrdOut.cpp
@@ -18,7 +18,7 @@ Exec::RetType Exec_CrdOut::Execute(CpptrajState& State, ArgList& argIn) {
   //rprintf("DEBUG: About to create new comm, ID= %i\n", ID);
   trajComm_ = Parallel::World().Split( ID );
   if (ID != MPI_UNDEFINED) {
-    mprintf("Warning: '%s' command does not yet use multiple MPI threads.\n", argIn.Command());
+    mprintf("Warning: '%s' command does not yet use multiple MPI processes.\n", argIn.Command());
     ret = WriteCrd(State, argIn);
     if (ret != CpptrajState::OK)
       err = 1;
diff --git a/src/Exec_RunAnalysis.cpp b/src/Exec_RunAnalysis.cpp
index dfb7f596c1..0ee40a7fc9 100644
--- a/src/Exec_RunAnalysis.cpp
+++ b/src/Exec_RunAnalysis.cpp
@@ -24,7 +24,7 @@ Exec::RetType Exec_RunAnalysis::Execute(CpptrajState& State, ArgList& argIn) {
 # ifdef MPI
   // Only master performs analyses currently.
   if (Parallel::TrajComm().Size() > 1)
-    mprintf("Warning: Analysis does not currently use multiple MPI threads.\n");
+    mprintf("Warning: Analysis does not currently use multiple MPI processes.\n");
   if (Parallel::TrajComm().Master())
 # endif
     err = DoRunAnalysis(State, argIn);
diff --git a/src/Parallel.cpp b/src/Parallel.cpp
index f1df3cce90..d5e77af9e0 100644
--- a/src/Parallel.cpp
+++ b/src/Parallel.cpp
@@ -58,10 +58,10 @@ void Parallel::dbgprintf(const char* format, ...) {
   return;
 }
 
-/** Open a file named Thread.worldrank for this thread */
+/** Open a file named Process.worldrank for this process */
 int Parallel::debug_init() {
   char outfilename[32];
-  sprintf(outfilename, "Thread.%03i", world_.Rank());
+  sprintf(outfilename, "Process.%03i", world_.Rank());
   mpidebugfile_ = fopen(outfilename, "w");
   if (mpidebugfile_ == NULL) {
     fprintf(stderr,"[%i]\tCould not open debug file:\n", world_.Rank());
@@ -74,7 +74,7 @@ int Parallel::debug_init() {
   return 0;
 }
 
-/** Close Thread.worldrank file.  */
+/** Close Process.worldrank file.  */
 int Parallel::debug_end() {
   if (mpidebugfile_ != 0)
     fclose(mpidebugfile_);
@@ -97,7 +97,7 @@ int Parallel::Init(int argc, char** argv) {
   //char processor_name[MPI_MAX_PROCESSOR_NAME];
   //int namelen;
   //MPI_Get_processor_name(processor_name, &namelen);
-  //printf("DEBUG: Thread %i of %i on %s\n", world_.Rank(), world_.Size(), processor_name);
+  //printf("DEBUG: Process %i of %i on %s\n", world_.Rank(), world_.Size(), processor_name);
   return 0;
 }
 
@@ -125,13 +125,13 @@ int Parallel::Abort(int errcode) {
 /** Trajectory and Ensemble communicators are set up orthogonal to one
   * another. In row-major notation, Trajectory communicators are set up
   * across rows, and Ensemble communicators are set up down columns. For
-  * example, if reading in 2 ensemble members with 3 threads per member
+  * example, if reading in 2 ensemble members with 3 processes per member
   * (world size 6), the layout would be:
   *   0 1 2 (member 0)
   *   3 4 5 (member 1)
-  * Threads 0 and 3 would read the first third of the trajectories, etc.
+  * Processes 0 and 3 would read the first third of the trajectories, etc.
   */
-int Parallel::SetupComms(int ngroups, bool allowFewerThreadsThanGroups) {
+int Parallel::SetupComms(int ngroups, bool allowFewerProcessesThanGroups) {
   if (ngroups < 1) {
     // If ngroups < 1 assume we want to reset comm info
     //fprintf(stdout, "DEBUG: Resetting ensemble/traj comm info.\n");
@@ -152,32 +152,32 @@ int Parallel::SetupComms(int ngroups, bool allowFewerThreadsThanGroups) {
       return 1;
     }
   } else if (world_.Size() < ngroups) {
-    // Fewer threads than groups. Make sure that # threads is a
+    // Fewer processes than groups. Make sure that # processes is a
     // multiple of ngroups. This is required for things like AllGather to work
     // properly.
-    if (!allowFewerThreadsThanGroups) {
-      fprintf(stderr,"Error: Fewer threads than groups currently not allowed.\n");
+    if (!allowFewerProcessesThanGroups) {
+      fprintf(stderr,"Error: Fewer processes than groups currently not allowed.\n");
       return 1;
     }
     trajComm_.Reset();
     ensembleComm_.Reset();
     if ( (ngroups % world_.Size()) != 0 ) {
-      fprintf(stderr,"Error: # of replicas (%i) must be a multiple of # threads (%i)\n",
+      fprintf(stderr,"Error: # of replicas (%i) must be a multiple of # processes (%i)\n",
               ngroups, world_.Size());
       return 1;
     }
     ensemble_size_ = ngroups;
-    n_ens_members_ = world_.DivideAmongThreads( ensemble_beg_, ensemble_end_, ensemble_size_ );
+    n_ens_members_ = world_.DivideAmongProcesses( ensemble_beg_, ensemble_end_, ensemble_size_ );
     int ID = world_.Rank();
     trajComm_ = world_.Split( ID );
     // NOTE: This effectively duplicates World
     ensembleComm_ = world_.Split( 0 );
     world_.Barrier();
   } else {
-    // Threads >= groups. Make sure that ngroups is a multiple of total # threads.
+    // Processes >= groups. Make sure that ngroups is a multiple of total # processes.
     if ( (world_.Size() % ngroups) != 0 ) {
       if ( world_.Master() )
-        fprintf(stderr,"Error: # of threads (%i) must be a multiple of # replicas (%i)\n",
+        fprintf(stderr,"Error: # of processes (%i) must be a multiple of # replicas (%i)\n",
                 world_.Size(), ngroups);
       return 1;
     }
@@ -233,7 +233,7 @@ int Parallel::SetupComms(int ngroups, bool allowFewerThreadsThanGroups) {
 
 /** Can be placed inside the code so debugger can be attached. */
 void Parallel::Lock() {
-  fprintf(stdout,"[%i] Thread is locked. Waiting for debugger.\n", world_.Rank());
+  fprintf(stdout,"[%i] Process is locked. Waiting for debugger.\n", world_.Rank());
   int PleaseWait = 1;
   while (PleaseWait == 1)
     PleaseWait *= 1;
@@ -300,20 +300,20 @@ void Parallel::Comm::Reset() {
 }
 
 /** Split given number of elements as evenly as possible among ranks.
-  * \return Number of elements this thread is responsible for.
+  * \return Number of elements this process is responsible for.
   */
-int Parallel::Comm::DivideAmongThreads(int& my_start, int& my_stop, int maxElts) const
+int Parallel::Comm::DivideAmongProcesses(int& my_start, int& my_stop, int maxElts) const
 {
-  int frames_per_thread = maxElts / size_;
+  int frames_per_process = maxElts / size_;
   int remainder         = maxElts % size_;
-  int my_frames         = frames_per_thread + (int)(rank_ < remainder);
-  // Figure out where this thread starts and stops
+  int my_frames         = frames_per_process + (int)(rank_ < remainder);
+  // Figure out where this process starts and stops
   my_start = 0;
   for (int rnk = 0; rnk != rank_; rnk++)
     if (rnk < remainder)
-      my_start += (frames_per_thread + 1);
+      my_start += (frames_per_process + 1);
     else
-      my_start += (frames_per_thread);
+      my_start += (frames_per_process);
   my_stop = my_start + my_frames;
   return my_frames;
 }
diff --git a/src/Parallel.h b/src/Parallel.h
index a1f02147a4..39157673af 100644
--- a/src/Parallel.h
+++ b/src/Parallel.h
@@ -51,7 +51,7 @@ class Parallel {
     static int Abort(int);
     /// Set up ensemble and trajectory communicators for given ensemble size
     static int SetupComms(int, bool);
-    /// Set up communicators - do not allow fewer threads than groups. TODO remove
+    /// Set up communicators - do not allow fewer processes than groups. TODO remove
     static int SetupComms(int n) { return SetupComms(n, false); }
     /// For DEBUG: infinite loop, gives time to attach a debugger.
     static void Lock();
@@ -61,11 +61,11 @@ class Parallel {
     static bool EnsembleIsSetup()       { return ensemble_size_ > -1; }
     /// \return total ensemble size.
     static int Ensemble_Size()          { return ensemble_size_;  }
-    /// \return First ensemble member this thread is responsible for.
+    /// \return First ensemble member this process is responsible for.
     static int Ensemble_Beg()           { return ensemble_beg_;   }
-    /// \return Last+1 ensemble member this thread is responsible for.
+    /// \return Last+1 ensemble member this process is responsible for.
     static int Ensemble_End()           { return ensemble_end_;   }
-    /// \return Total number of ensemble members this thread is responsible for.
+    /// \return Total number of ensemble members this process is responsible for.
     static int N_Ens_Members()          { return n_ens_members_;  }
     /// \return Rank in ensemble comm. for given member.
     static int MemberEnsCommRank(int i) { return memberEnsRank_[i];}
@@ -82,9 +82,9 @@ class Parallel {
     static void printMPIerr(int, const char*, int);
     static int checkMPIerr(int, const char*, int);
     static int ensemble_size_;  ///< Total number of ensemble members.
-    static int ensemble_beg_;   ///< Starting member for this ensemble thread.
-    static int ensemble_end_;   ///< Ending member for this ensemble thread.
-    static int n_ens_members_;  ///< Number of ensemble members thread is responsible for.
+    static int ensemble_beg_;   ///< Starting member for this ensemble process.
+    static int ensemble_end_;   ///< Ending member for this ensemble process.
+    static int n_ens_members_;  ///< Number of ensemble members process is responsible for.
     static int* memberEnsRank_; ///< Rank in ensemble comm for each member.
 #   ifdef PARALLEL_DEBUG_VERBOSE
     static void dbgprintf(const char*, ...);
@@ -118,7 +118,7 @@ class Parallel::Comm {
     Comm Split(int) const;
     void Reset();
     /// my_start, my_stop, maxElts
-    int DivideAmongThreads(int&, int&, int) const;
+    int DivideAmongProcesses(int&, int&, int) const;
     /// RecvBuffer, SendBuffer, Count, DataType, Op
     int ReduceMaster(void*, void*, int, MPI_Datatype, MPI_Op) const;
     /// Rank, RecvBuffer, SendBuffer, Count, DataType, Op
diff --git a/src/ParmFile.cpp b/src/ParmFile.cpp
index dc7d8ef0a0..6e182840fb 100644
--- a/src/ParmFile.cpp
+++ b/src/ParmFile.cpp
@@ -59,6 +59,14 @@ ParmIO* ParmFile::DetectFormat(FileName const& fname, ParmFormatType& ptype) {
   return 0;
 }
 
+// ParmFile::DetectFormat()
+ParmFile::ParmFormatType ParmFile::DetectFormat(FileName const& fname) {
+  ParmFormatType ptype;
+  ParmIO* pio = DetectFormat(fname, ptype);
+  delete pio;
+  return ptype;
+}
+
 // ParmFile::ReadTopology()
 int ParmFile::ReadTopology(Topology& Top, FileName const& fnameIn, 
                            ArgList const& argListIn, int debugIn) 
diff --git a/src/ParmFile.h b/src/ParmFile.h
index f54c343c88..ce2dafa717 100644
--- a/src/ParmFile.h
+++ b/src/ParmFile.h
@@ -25,8 +25,12 @@ class ParmFile {
       return WriteTopology(t, n, ArgList(), f, d);
     }
     FileName const& ParmFilename() { return parmName_; }
+    /// \return ParmFormatType of given file or UNKNOWN_PARM.
+    static ParmFormatType DetectFormat(FileName const&);
   private :
-    ParmIO* DetectFormat(FileName const&, ParmFormatType&); 
+    /// \return Allocated ParmIO if given file matches known type, 0 otherwise.
+    static ParmIO* DetectFormat(FileName const&, ParmFormatType&);
+
     FileName parmName_; ///< Topology input/output file name. 
 };
 #endif
diff --git a/src/TrajectoryFile.cpp b/src/TrajectoryFile.cpp
index 7e05fe8e43..b4c5d34a49 100644
--- a/src/TrajectoryFile.cpp
+++ b/src/TrajectoryFile.cpp
@@ -128,3 +128,11 @@ TrajectoryIO* TrajectoryFile::DetectFormat(FileName const& fname, TrajFormatType
   ttype = UNKNOWN_TRAJ;
   return 0;
 }
+
+// TrajectoryFile::DetectFormat()
+TrajectoryFile::TrajFormatType TrajectoryFile::DetectFormat(FileName const& fname) {
+  TrajFormatType ttype;
+  TrajectoryIO* tio = DetectFormat(fname, ttype);
+  delete tio;
+  return ttype;
+}
diff --git a/src/TrajectoryFile.h b/src/TrajectoryFile.h
index 132d9a2b60..6525ee79f1 100644
--- a/src/TrajectoryFile.h
+++ b/src/TrajectoryFile.h
@@ -59,5 +59,7 @@ class TrajectoryFile {
     static TrajectoryIO* AllocTrajIO(TrajFormatType t) {
       return (TrajectoryIO*)FileTypes::AllocIO(TF_AllocArray, t, true);
     }
+    /// \return TrajFormatType of given file or UNKNOWN_TRAJ
+    static TrajFormatType DetectFormat(FileName const&);
 };
 #endif
diff --git a/src/Version.h b/src/Version.h
index d950480a41..d85aaf1897 100644
--- a/src/Version.h
+++ b/src/Version.h
@@ -18,5 +18,5 @@
  * Whenever a number that precedes <revision> is incremented, all subsequent
  * numbers should be reset to 0.
  */
-#define CPPTRAJ_INTERNAL_VERSION "V4.3.11"
+#define CPPTRAJ_INTERNAL_VERSION "V4.3.12"
 #endif
diff --git a/test/Makefile b/test/Makefile
index 7b5481b4c4..cffaabb1ff 100644
--- a/test/Makefile
+++ b/test/Makefile
@@ -434,6 +434,9 @@ test.infraredspec:
 test.cphstats:
 	@-cd Test_Cphstats && ./RunTest.sh $(OPT)
 
+test.cmdline:
+	@-cd Test_CmdLine && ./RunTest.sh $(OPT)
+
 # Every test target should go here.
 COMPLETETESTS=test.general \
               test.strip \
@@ -573,7 +576,8 @@ COMPLETETESTS=test.general \
               test.phipsi \
               test.control \
               test.infraredspec \
-              test.cphstats
+              test.cphstats \
+              test.cmdline
 
 test.all:
 	$(MAKE) test.complete summary
diff --git a/test/Test_2DRMS/RunTest.sh b/test/Test_2DRMS/RunTest.sh
index 9c4ee0b8be..4011f1f4f2 100755
--- a/test/Test_2DRMS/RunTest.sh
+++ b/test/Test_2DRMS/RunTest.sh
@@ -6,7 +6,7 @@
 CleanFiles rms.in rmsd1.dat rmsd2.dat ref.nc rmsd.mass.dat dme.dat trp.dat nofit.dat
 
 TESTNAME='2D RMSD tests'
-Requires netcdf
+Requires netcdf maxthreads 10
 TOP="../tz2.parm7"
 CRD="../tz2.nc"
 INPUT="rms.in"
diff --git a/test/Test_Align/RunTest.sh b/test/Test_Align/RunTest.sh
index 8b1942a894..68ce1636b4 100755
--- a/test/Test_Align/RunTest.sh
+++ b/test/Test_Align/RunTest.sh
@@ -6,7 +6,7 @@
 CleanFiles rms.in rmsd.dat rmsd.mass.dat rmsd.reftraj.dat
 
 TESTNAME='Align tests'
-Requires netcdf
+Requires netcdf maxthreads 10
 TOP="../tz2.truncoct.parm7"
 INPUT="rms.in"
 
diff --git a/test/Test_AnalysisRunAvg/RunTest.sh b/test/Test_AnalysisRunAvg/RunTest.sh
index b9607b6037..3452d8dac4 100755
--- a/test/Test_AnalysisRunAvg/RunTest.sh
+++ b/test/Test_AnalysisRunAvg/RunTest.sh
@@ -4,7 +4,7 @@
 
 CleanFiles runavg.in running_avg.dat cumulative_avg.dat distances.dat
 TESTNAME='Analysis Running Average'
-Requires netcdf
+Requires netcdf maxthreads 14
 INPUT="runavg.in"
 TOP="../tz2.parm7"
 cat > $INPUT <<EOF
diff --git a/test/Test_Closest/RunTest.sh b/test/Test_Closest/RunTest.sh
index 90e0cece5d..4281d6d4a9 100755
--- a/test/Test_Closest/RunTest.sh
+++ b/test/Test_Closest/RunTest.sh
@@ -9,7 +9,7 @@ CleanFiles closest.in first.Closest.pdb.1 closestmols.dat \
 
 INPUT="-i closest.in"
 TESTNAME='Closest tests'
-Requires netcdf
+Requires netcdf maxthreads 10
 
 UNITNAME='Closest command tests using first/all solvent atoms'
 CheckFor maxthreads 1
diff --git a/test/Test_CmdLine/RunTest.sh b/test/Test_CmdLine/RunTest.sh
new file mode 100755
index 0000000000..84ae786588
--- /dev/null
+++ b/test/Test_CmdLine/RunTest.sh
@@ -0,0 +1,47 @@
+#!/bin/bash
+
+. ../MasterTest.sh
+
+# Clean
+CleanFiles cmd.in test1.crd.save test1.crd res res.save
+
+TESTNAME='Command line tests'
+# Required environment 
+Requires maxthreads 8
+
+
+# First test without command line
+cat > cmd.in <<EOF
+parm ../tz2.parm7
+trajin ../tz2.crd 2 13 3
+trajin ../tz2.crd 98 last
+trajout test1.crd.save title MyTitle
+EOF
+INPUT='-i cmd.in'
+RunCpptraj "Command line trajectory args test, part 1"
+
+# Now test using the command line. Hijack the INPUT variable.
+INPUT="-p ../tz2.parm7 -y ../tz2.crd -ya \"2 13 3\" -y ../tz2.crd -ya \"98 last\" -x test1.crd -xa \"title MyTitle\""
+RunCpptraj "Command line test trajectory args test, part 2"
+DoTest test1.crd.save test1.crd
+
+# Test loading multiple topology files
+cat > cmd.in <<EOF
+parm ../tz2.parm7
+parm ../tz2.pdb
+resinfo out res.save parmindex 0
+resinfo out res.save parmindex 1
+EOF
+INPUT='-i cmd.in'
+RunCpptraj "Command line wildcard test, part 1"
+cat > cmd.in <<EOF
+resinfo out res parmindex 0
+resinfo out res parmindex 1
+EOF
+INPUT='../tz2.p* cmd.in'
+RunCpptraj "Command line wildcard test, part 2"
+DoTest res.save res
+
+EndTest
+
+exit 0
diff --git a/test/Test_Datafile/RunTest.sh b/test/Test_Datafile/RunTest.sh
index 4fbcd8758f..cd36bb51fa 100755
--- a/test/Test_Datafile/RunTest.sh
+++ b/test/Test_Datafile/RunTest.sh
@@ -5,12 +5,14 @@
 # Clean
 CleanFiles prec.in prec.dat a1.dat a1.agr xprec.dat
 
+TESTNAME='Data file tests'
+
 TOP="../tz2.truncoct.parm7"
 INPUT="prec.in"
 
 # Test 1
 UNITNAME='Data file output precision test'
-CheckFor netcdf
+CheckFor netcdf maxthreads 10
 if [ $? -eq 0 ] ; then
   cat > prec.in <<EOF
 noprogress
diff --git a/test/Test_Diffusion/RunTest.sh b/test/Test_Diffusion/RunTest.sh
index 5ad3b6d309..461c245507 100755
--- a/test/Test_Diffusion/RunTest.sh
+++ b/test/Test_Diffusion/RunTest.sh
@@ -7,7 +7,7 @@ INPUT="diffusion.in"
 CleanFiles $INPUT diff_?.xmgr diff.dat diff.1.dat diff.2.dat diff.3.dat nw.dat \
            WAT_O.agr DC.dat Nonortho.agr Nonortho.dat noimage.agr noimage.dat
 TESTNAME='Diffusion tests'
-Requires netcdf
+Requires netcdf maxthreads 10
 
 # Basic ptraj diffusion test
 # creates <prefix>_X.xmgr, X = {a,r,x,y,z}
diff --git a/test/Test_DihCovar/RunTest.sh b/test/Test_DihCovar/RunTest.sh
index c2fe123f38..d4d2d9264e 100755
--- a/test/Test_DihCovar/RunTest.sh
+++ b/test/Test_DihCovar/RunTest.sh
@@ -6,7 +6,7 @@ CleanFiles matrix.in mtest.dat.save mtest.*.dat dihcovar.dat modes.dihcovar.dat
            dih.project.dat dih.project.agr dihedrals.dat
 
 TESTNAME='Dihedral Covariance Matrix Test'
-Requires mathlib
+Requires mathlib maxthreads 10
 
 INPUT="-i matrix.in"
 if [ -z "$DO_PARALLEL" ] ; then
diff --git a/test/Test_Ensemble_MREMD/RunTest.sh b/test/Test_Ensemble_MREMD/RunTest.sh
index e5813f9c8a..7359858965 100755
--- a/test/Test_Ensemble_MREMD/RunTest.sh
+++ b/test/Test_Ensemble_MREMD/RunTest.sh
@@ -81,18 +81,22 @@ EOF
 
 # Test M-REMD process, no sort, running average (tests preload in parallel)
 RunAvgTest() {
-  cat > mremd.in <<EOF
+  UNITNAME='M-REMD no sort, running average test'
+  CheckFor maxthreads 24
+  if [ $? -eq 0 ] ; then
+    cat > mremd.in <<EOF
 parm rGACC.nowat.parm7
 ensemblesize 8
 ensemble rGACC.nowat.001 nosort
 runavg window 3
 rms RA first :1-4&!@H= out RA.dat
 EOF
-  RunCpptraj "M-REMD no sort, running average test"
-  if [ -z "$DO_PARALLEL" ] ; then
-    DoTest RA.dat.save RA.dat
-  else
-    cat RA.dat.? > all.RA.dat && DoTest all.RA.dat.save all.RA.dat
+    RunCpptraj "$UNITNAME"
+    if [ -z "$DO_PARALLEL" ] ; then
+      DoTest RA.dat.save RA.dat
+    else
+      cat RA.dat.? > all.RA.dat && DoTest all.RA.dat.save all.RA.dat
+    fi
   fi
 }
 
diff --git a/test/Test_Esander/RunTest.sh b/test/Test_Esander/RunTest.sh
index 4c577effa7..18efb1232c 100755
--- a/test/Test_Esander/RunTest.sh
+++ b/test/Test_Esander/RunTest.sh
@@ -10,7 +10,7 @@ INPUT="-i ene.in"
 
 TestPME() {
   UNITNAME='SANDER energy test, PME'
-  CheckFor pnetcdf
+  CheckFor pnetcdf maxthreads 10
   if [ $? -eq 0 ] ; then
     cat > ene.in <<EOF
 parm ../tz2.truncoct.parm7
@@ -35,14 +35,18 @@ EOF
 }
 
 TestStrip() {
-  cat > ene.in <<EOF
+  UNITNAME="SANDER energy test after 'strip', PME."
+  CheckFor maxthreads 10
+  if [ $? -eq 0 ] ; then
+    cat > ene.in <<EOF
 parm ../tz2.truncoct.parm7
 trajin ../tz2.truncoct.nc
 strip :WAT
 esander NoWat out NoWat.dat
 EOF
-  RunCpptraj "SANDER energy test after 'strip', PME."
-  DoTest NoWat.dat.save NoWat.dat
+    RunCpptraj "$UNITNAME"
+    DoTest NoWat.dat.save NoWat.dat
+  fi
 }
 
 TestGB
diff --git a/test/Test_FixAtomOrder/RunTest.sh b/test/Test_FixAtomOrder/RunTest.sh
index b6a0052026..7de75f97a5 100755
--- a/test/Test_FixAtomOrder/RunTest.sh
+++ b/test/Test_FixAtomOrder/RunTest.sh
@@ -4,6 +4,8 @@
 
 CleanFiles fix.in order.in reorder.outoforder.parm7 reorder.mdcrd
 
+TESTNAME='Out of order molecules test'
+Requires maxthreads 10
 INPUT='-i fix.in'
 
 cat > fix.in <<EOF
@@ -16,7 +18,7 @@ trajin min1.crd 1 10
 fixatomorder outprefix reorder
 trajout reorder.mdcrd
 EOF
-RunCpptraj "Out of order molecules test (will print error messages)"
+RunCpptraj "$TESTNAME (will print error messages)"
 DoTest reorder.outoforder.parm7.save reorder.outoforder.parm7 -I %VERSION
 DoTest reorder.mdcrd.save reorder.mdcrd
 
diff --git a/test/Test_GromacsTrr/RunTest.sh b/test/Test_GromacsTrr/RunTest.sh
index 2d6701b2b0..beef3f8af1 100755
--- a/test/Test_GromacsTrr/RunTest.sh
+++ b/test/Test_GromacsTrr/RunTest.sh
@@ -4,6 +4,9 @@
 
 CleanFiles ptraj.in cpptraj.nc mop.trr temp.crd total?.out
 
+TESTNAME='GROMACS TRR tests'
+Requires maxthreads 10
+
 INPUT="-i ptraj.in"
 
 GmxTrrRead() {
diff --git a/test/Test_Hbond/RunTest.sh b/test/Test_Hbond/RunTest.sh
index 4b4419b174..528162042c 100755
--- a/test/Test_Hbond/RunTest.sh
+++ b/test/Test_Hbond/RunTest.sh
@@ -42,7 +42,7 @@ EOF
 # Solute-Solvent test
 TestUV() {
   UNITNAME='Solute-solvent hbond test'
-  CheckFor netcdf
+  CheckFor netcdf maxthreads 10
   if [ $? -eq 0 ] ; then
     cat > hbond.in <<EOF
 parm ../tz2.ortho.parm7
@@ -76,7 +76,7 @@ EOF
 # Nointramol test
 TestNointramol() {
   UNITNAME='Hbond, no intramolecular hydrogen bonds test'
-  CheckFor netcdf
+  CheckFor netcdf maxthreads 10
   if [ $? -eq 0 ] ; then
     cat > hbond.in <<EOF
 parm ../FtuFabI.NAD.TCL.parm7
diff --git a/test/Test_LIE/RunTest.sh b/test/Test_LIE/RunTest.sh
index 3f37ed11f4..64e9808ad7 100755
--- a/test/Test_LIE/RunTest.sh
+++ b/test/Test_LIE/RunTest.sh
@@ -6,7 +6,7 @@
 CleanFiles lie.in TCL.out
 
 TESTNAME='LIE test'
-Requires netcdf
+Requires netcdf maxthreads 10
 
 INPUT="lie.in"
 TOP=../FtuFabI.NAD.TCL.parm7
diff --git a/test/Test_Mask/RunTest.sh b/test/Test_Mask/RunTest.sh
index 94c4e7c8c8..941d96f9db 100755
--- a/test/Test_Mask/RunTest.sh
+++ b/test/Test_Mask/RunTest.sh
@@ -6,7 +6,7 @@
 CleanFiles mask.in mask.out mask.pdb.1 mask.mol2.1 M.dat
 
 TESTNAME='Mask command tests'
-Requires netcdf
+Requires netcdf maxthreads 10
 
 INPUT="-i mask.in"
 # Test 1
diff --git a/test/Test_Matrix/RunTest.sh b/test/Test_Matrix/RunTest.sh
index ab80a71c18..352e980a8e 100755
--- a/test/Test_Matrix/RunTest.sh
+++ b/test/Test_Matrix/RunTest.sh
@@ -4,6 +4,9 @@
 
 CleanFiles matrix.in mtest.dat.save mtest.*.dat evecs.10.dat
 
+TESTNAME='Matrix Tests'
+Requires maxthreads 10
+
 INPUT="-i matrix.in"
 cat > matrix.in <<EOF
 parm 1rrb_vac.prmtop
@@ -23,7 +26,7 @@ matrix mwcovar @CA out mtest.11.dat
 matrix dist @N @C out mtest.12.dat 
 matrix distcovar :1-4@CA out mtest.13.dat
 EOF
-RunCpptraj "Matrix Tests."
+RunCpptraj "$TESTNAME"
 DoTest mtest.dat.0.save mtest.0.dat
 DoTest mtest.dat.1.save mtest.1.dat
 DoTest mtest.dat.2.save mtest.2.dat
diff --git a/test/Test_Molsurf/RunTest.sh b/test/Test_Molsurf/RunTest.sh
index 304b8764a3..2576450a5d 100755
--- a/test/Test_Molsurf/RunTest.sh
+++ b/test/Test_Molsurf/RunTest.sh
@@ -4,7 +4,7 @@
 
 CleanFiles molsurf.in msurf.dat
 TESTNAME='Molsurf test'
-Requires netcdf
+Requires netcdf maxthreads 10
 # Molsurf test
 cat > molsurf.in <<EOF
 parm ../tz2.parm7
diff --git a/test/Test_Multidihedral/RunTest.sh b/test/Test_Multidihedral/RunTest.sh
index f1b9760c06..b13e52d189 100755
--- a/test/Test_Multidihedral/RunTest.sh
+++ b/test/Test_Multidihedral/RunTest.sh
@@ -4,6 +4,9 @@
 
 CleanFiles dih.in multidih.dat dihedral.dat custom.dat dihedral2.dat all.dat chin.dat
 
+TESTNAME='Multidihedral tests'
+Requires maxthreads 10
+
 INPUT="-i dih.in"
 UNITNAME='Multidihedral test'
 CheckFor netcdf
diff --git a/test/Test_NetcdfTraj/RunTest.sh b/test/Test_NetcdfTraj/RunTest.sh
index 739698ca45..3b5e93d328 100755
--- a/test/Test_NetcdfTraj/RunTest.sh
+++ b/test/Test_NetcdfTraj/RunTest.sh
@@ -6,7 +6,7 @@
 CleanFiles ptraj_netcdf.in trajectory.netcdf trajectory_test.mdcrd
 
 TESTNAME='NetCDF tests'
-Requires netcdf pnetcdf
+Requires netcdf pnetcdf maxthreads 10
 
 # Convert MDCRD to NETCDF and back again.
 INPUT="ptraj_netcdf.in"
diff --git a/test/Test_Outtraj/RunTest.sh b/test/Test_Outtraj/RunTest.sh
index d49eff6792..313ee0d771 100755
--- a/test/Test_Outtraj/RunTest.sh
+++ b/test/Test_Outtraj/RunTest.sh
@@ -6,7 +6,7 @@
 CleanFiles rms.in rmsd.dat test.crd T1.crd T2.crd maxmin.in maxmin.crd
 
 TESTNAME='Outtraj tests'
-Requires netcdf
+Requires netcdf maxthreads 10
 TOP="../tz2.truncoct.parm7"
 
 # Test 1
diff --git a/test/Test_Parallel/RunTest.sh b/test/Test_Parallel/RunTest.sh
index 1f65617a3d..37840875b0 100755
--- a/test/Test_Parallel/RunTest.sh
+++ b/test/Test_Parallel/RunTest.sh
@@ -30,7 +30,7 @@ EOF
 
 Test2() {
   UNITNAME='Parallel test with trajectory offset'
-  CheckFor pnetcdf
+  CheckFor pnetcdf maxthreads 22
   if [ $? -eq 0 ] ; then
     cat > para.in <<EOF
 noprogress
diff --git a/test/Test_Principal/RunTest.sh b/test/Test_Principal/RunTest.sh
index 56b768e46a..ed6ed7addd 100755
--- a/test/Test_Principal/RunTest.sh
+++ b/test/Test_Principal/RunTest.sh
@@ -5,6 +5,9 @@
 
 CleanFiles principal.in Ctest.pdb principal.dat Ctest.crd eigen.dat
 
+TESTNAME='Principal axes tests'
+Requires maxthreads 10
+
 TOP="../Test_IRED/1IEE_A_prot.prmtop"
 INPUT="principal.in"
 
diff --git a/test/Test_RMSD/RunTest.sh b/test/Test_RMSD/RunTest.sh
index b9c0747f84..18cc641399 100755
--- a/test/Test_RMSD/RunTest.sh
+++ b/test/Test_RMSD/RunTest.sh
@@ -11,13 +11,12 @@ CleanFiles rms.in rmsd.dat rms.mass.in rmsd.mass.dat rms.reftraj.in \
 TESTNAME='RMSD tests'
 Requires netcdf
 
-INPUT="rms.in"
-
 # Test rmsd, mass-weighted rmsd, rmsd to reference traj.
 UNITNAME='Basic RMSD tests'
 CheckFor maxthreads 10
 if [ $? -eq 0 ] ; then
-  TOP="../tz2.truncoct.parm7"
+  TOP='../tz2.truncoct.parm7'
+  INPUT='rms.in'
   cat > rms.in <<EOF
 noprogress
 trajin ../tz2.truncoct.nc
@@ -40,9 +39,12 @@ EOF
 fi
 
 # Test RMS rotate/norotate, generation of rotation matrices
-TOP=""
-INPUT="-i rms.in"
-cat > rms.in <<EOF
+UNITNAME='RMS coordinate rotation/rotation matrices test'
+CheckFor maxthreads 10
+if [ $? -eq 0 ] ; then
+  TOP=''
+  INPUT='-i rms.in'
+  cat > rms.in <<EOF
 parm ../tz2.parm7 [NOWAT] 
 reference ../tz2.nc parm [NOWAT] 1 [first] 
 parm ../tz2.truncoct.parm7 [WAT]
@@ -53,12 +55,14 @@ outtraj tz2.norotate.crd parm [WAT]
 rms ROT ref [first] out rms.dat @CA savematrices matricesout rmatrices.dat
 outtraj tz2.rotate.crd parm [WAT]
 EOF
-RunCpptraj "RMS coordinate rotation/rotation matrices test."
-DoTest tz2.norotate.crd.save tz2.norotate.crd
-DoTest tz2.rotate.crd.save tz2.rotate.crd
-DoTest rmatrices.dat.save rmatrices.dat
+  RunCpptraj "$UNITNAME"
+  DoTest tz2.norotate.crd.save tz2.norotate.crd
+  DoTest tz2.rotate.crd.save tz2.rotate.crd
+  DoTest rmatrices.dat.save rmatrices.dat
+fi
 
 # Test RMS nomod
+TOP=''
 INPUT="-i rms.in"
 cat > rms.in <<EOF
 parm ../tz2.parm7
diff --git a/test/Test_RemdTraj/RunTest.sh b/test/Test_RemdTraj/RunTest.sh
index dac1c79d3d..4234e107ae 100755
--- a/test/Test_RemdTraj/RunTest.sh
+++ b/test/Test_RemdTraj/RunTest.sh
@@ -6,6 +6,9 @@
 CleanFiles remd.in d1.offset.dat d1.crd.dat d1.nc.dat temp.crd.* \
            temp0.crd.* d1.ensemble.dat d1.ensemble.dat.? all.dat 
 
+TESTNAME='Replica exchange trajectory tests'
+Requires maxthreads 10
+
 INPUT="-i remd.in"
 
 # Test 0
diff --git a/test/Test_Rotate/RunTest.sh b/test/Test_Rotate/RunTest.sh
index 444858d483..57a807f477 100755
--- a/test/Test_Rotate/RunTest.sh
+++ b/test/Test_Rotate/RunTest.sh
@@ -4,6 +4,9 @@
 
 CleanFiles rotate.in fromMatrices.crd TCS.rotated.mol2 inverse.crd
 
+TESTNAME='Coordinates rotation tests'
+Requires maxthreads 10
+
 INPUT="-i rotate.in"
 cat > rotate.in <<EOF
 parm ../tz2.truncoct.parm7
diff --git a/test/Test_Runavg/RunTest.sh b/test/Test_Runavg/RunTest.sh
index f803c0e02e..725b551967 100755
--- a/test/Test_Runavg/RunTest.sh
+++ b/test/Test_Runavg/RunTest.sh
@@ -4,7 +4,7 @@
 
 CleanFiles runavg.in runavg.crd R1.dat
 TESTNAME='Coordinate running average tests'
-Requires netcdf
+Requires netcdf maxthreads 20
 TOP="../tz2.parm7"
 INPUT="runavg.in"
 
diff --git a/test/Test_SPAM/RunTest.sh b/test/Test_SPAM/RunTest.sh
index b35fc6eedd..b1bd1dc2ca 100755
--- a/test/Test_SPAM/RunTest.sh
+++ b/test/Test_SPAM/RunTest.sh
@@ -7,7 +7,7 @@ CleanFiles spam.in spam.dat spam.info spampure.dat test.mdcrd summary.dat
 
 # Check libraries
 TESTNAME='SPAM tests'
-Requires netcdf
+Requires netcdf maxthreads 10
 
 INPUT="spam.in"
 # SPAM test
diff --git a/test/Test_Temperature/RunTest.sh b/test/Test_Temperature/RunTest.sh
index f0f39cd6b0..ab5021454a 100755
--- a/test/Test_Temperature/RunTest.sh
+++ b/test/Test_Temperature/RunTest.sh
@@ -5,7 +5,7 @@
 CleanFiles temp.in T.dat T2.dat T3.dat
 
 TESTNAME='Temperature tests'
-Requires netcdf
+Requires netcdf maxthreads 10
 
 INPUT="-i temp.in"
 cat > temp.in <<EOF
diff --git a/test/Test_TrajinOffset/RunTest.sh b/test/Test_TrajinOffset/RunTest.sh
index 6c0be43182..c32225d10b 100755
--- a/test/Test_TrajinOffset/RunTest.sh
+++ b/test/Test_TrajinOffset/RunTest.sh
@@ -5,6 +5,8 @@
 # Clean
 CleanFiles cpptraj.offset.in rem.crd.combined gzip.rem.crd.combined bzip2.rem.crd.combined
 
+TESTNAME='Trajectory read with offset tests'
+Requires maxthreads 13
 # Test 1
 cat > cpptraj.offset.in <<EOF
 noprogress
diff --git a/test/Test_Watershell/RunTest.sh b/test/Test_Watershell/RunTest.sh
index 8c112677fa..a2d19ba398 100755
--- a/test/Test_Watershell/RunTest.sh
+++ b/test/Test_Watershell/RunTest.sh
@@ -5,7 +5,7 @@
 CleanFiles ws.in ws.agr ws.ortho.agr ws.noimage.agr
 
 TESTNAME='Watershell tests'
-Requires netcdf
+Requires netcdf maxthreads 10
 INPUT="ws.in"
 
 # Non-orthorhombic imaging
diff --git a/test/Test_systemVF/RunTest.sh b/test/Test_systemVF/RunTest.sh
index 21c0f6dced..343b9985ce 100755
--- a/test/Test_systemVF/RunTest.sh
+++ b/test/Test_systemVF/RunTest.sh
@@ -4,7 +4,7 @@
 
 CleanFiles systemVF.in test.systemVF.nc test.systemVf.ncdump
 TESTNAME='Amber NetCDF trajectory with Velocity/Force test'
-Requires netcdf pnetcdf
+Requires netcdf pnetcdf maxthreads 10
 
 cat > systemVF.in <<EOF
 parm systemVF.parm7